• [00:01:25] * Olipro_ (n=Olipro@unaffiliated/olipro) has joined #beagle
  • [00:07:00] * Olipro (n=Olipro@unaffiliated/olipro) Quit (Nick collision from services.)
  • [00:07:02] * Olipro_ is now known as Olipro
  • [00:07:47] * bobbarker (n=bob@adsl-69-109-116-44.dsl.pltn13.pacbell.net) Quit ()
  • [00:36:29] * asfidididisfs (n=jborn@adsl-75-2-247-191.dsl.chcgil.sbcglobal.net) has joined #beagle
  • [00:37:01] * asfidididisfs is now known as JoeBorn
  • [01:07:45] * cian (n=cianh@cian.ws) Quit ()
  • [01:23:30] * dcordes (n=dcordes@unaffiliated/dcordes) Quit (Remote closed the connection)
  • [01:27:26] * cian (n=cianh@cian.ws) has joined #beagle
  • [02:00:29] <vlad_> man, more failure
  • [02:00:55] * _Ade_ (n=Ade@78-86-195-216.zone2.bethere.co.uk) Quit ()
  • [02:01:06] <vlad_> handhelds mojo guys are compiling new eabi debian packages but using 'arm' as the arch, and not 'armel'
  • [02:06:27] * cian (n=cianh@cian.ws) Quit ()
  • [02:19:17] * JuanG (n=Juan@nat/ti/x-0514848c98ba4e72) has joined #beagle
  • [02:19:21] * JuanG (n=Juan@nat/ti/x-0514848c98ba4e72) has left #beagle
  • [02:33:04] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [02:37:30] * JoeBorn (n=jborn@adsl-75-2-247-191.dsl.chcgil.sbcglobal.net) Quit (Read error: 104 (Connection reset by peer))
  • [02:58:39] * prpplague (n=dave123_@ppp-70-249-146-43.dsl.rcsntx.swbell.net) Quit ()
  • [03:11:21] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [03:12:13] * JoeBorn__ (n=jborn@75.2.240.113) has joined #beagle
  • [03:16:07] * bobbarker (n=bob@adsl-69-109-116-44.dsl.pltn13.pacbell.net) has joined #beagle
  • [03:17:07] <bobbarker> Woot! I got my beagleboard's serial communications working!
  • [03:18:01] <bobbarker> Special thanks to mru and vlad_
  • [03:18:07] <davr> what turned out to be the problem?
  • [03:18:41] <bobbarker> The "null modem cable" I bought from Radio Shack wasn't a null modem.
  • [03:18:51] <bobbarker> Swapped the 2 and 3 pin and it works!
  • [03:20:04] <bobbarker> Now to decide on the first sample to try.
  • [03:26:36] * gmaxwell (n=NT4TN@wikimedia/KatWalsh/x-0001) has joined #beagle
  • [03:27:17] <gmaxwell> Anyone have the DSP doing anything yet?
  • [03:29:08] <bobbarker> What does DSP stand for?
  • [03:30:22] <davr> digital signal processor. it's a unit inside the OMAP
  • [03:30:33] <davr> aren't they using it for video decoding?
  • [03:31:11] <gmaxwell> Last I checked there was no kernel support for it. :)
  • [03:31:39] <bobbarker> Thanks, couldn't find it.
  • [03:31:54] <gmaxwell> I'd like to get an audio codec (celt-codec.org) I've been working on running on c64x and I thought beagleboard would be a good platform to do that on once the DSP is usable. :)
  • [03:32:41] <jkridner|work> DSP Link has been released.
  • [03:32:56] <jkridner|work> and so has DSP Bridge.
  • [03:33:30] <jkridner|work> There have been demonstrations with video codecs running on the DSP integrated with Angstrom (built with OE flow).
  • [03:33:54] <jkridner|work> compiler is out there too.
  • [03:34:23] <jkridner|work> Ogg Theora project has started on OSD2, but should move over to Beagle Board as well once done.
  • [03:34:58] <gmaxwell> jkridner, Compiler? The no-cost-for-non-commercial-use TI one or some port of GCC?
  • [03:36:21] <jkridner|work> still talking about the free, non-commercial-use TI compiler.
  • [03:36:37] <jkridner|work> GCC port was started at a university, but I don't think it is useful yet.
  • [03:37:29] <jkridner|work> I'd be happy to give some support if someone picks it up.
  • [03:37:48] <gmaxwell> jkridner, The instruction set looks rather complex. I've never worked with 64x before....
  • [03:38:10] <jkridner|work> VLIW is tough, but TI C compiler and intrinsics work pretty well.
  • [03:38:39] <jkridner|work> ignoring the instruction scheduling, the instruction set isn't all that complex (from my point of view).
  • [03:38:54] <jkridner|work> scheduling is tough.
  • [03:38:56] <gmaxwell> Well needing to handle scheduling is what I was referring to.. :)
  • [03:39:48] <jkridner|work> personally, I'd just use C and write the inner loops with intrinsics, then adjust based on what the compiler outputs.
  • [03:40:00] <ds2> Hmmm is the 64x in the OMAP3 close enough to a standalone 64x dsp to use it as a dev board for the 64x stuff?
  • [03:40:23] <jkridner|work> I did a bit of hand scheduling on the C6201--fun, but time consuming.
  • [03:40:49] <jkridner|work> ds2: yes, if what you are looking to learn is the ISA.
  • [03:41:01] <jkridner|work> ISA is the same as standalone C64x devices.
  • [03:41:20] <jkridner|work> most recently announced C64x devices add floating point.
  • [03:41:46] <ds2> I see...
  • [03:41:46] <jkridner|work> C67x devices are floating point as well, but didn't have some of the other optimizations in C64x.
  • [03:42:13] <ds2> I wonder if octave can generate pseudo code to feed to the c compiler and use the the omap3 as a signal processing "bread board"
  • [03:42:16] <jkridner|work> I think the L2 architecture might be a bit different...
  • [03:42:30] <jkridner|work> but, up to that point, they are the same as the stand-alone C64x devices.
  • [03:42:45] <ds2> for someone just learning DSPs, those details don't matter
  • [03:45:06] <gmaxwell> I've never used a DSP on a system like the omaps running Linux. How does scheduling code on the DSP work?
  • [03:45:22] <gmaxwell> Can multiple apps use it, etc?
  • [03:45:46] <jkridner|work> depends on what management software they use with it.
  • [03:46:01] <jkridner|work> Link and Bridge provide both communications and resource management capabilities.
  • [03:46:15] <jkridner|work> TI provides an RTOS for free use on the C64x. (DSP/BIOS)
  • [03:46:15] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [03:46:51] <ds2> hmmm, stupid question, but that does that mean one does not need to write cold start code for the DSP?
  • [03:47:41] <jkridner|work> Accelerated DirectFB was written for DM6446 (same DSP, different ARM sub-system) without using Link or any "standard" resource sharing, so it would be difficult to run other things on the DSP at the same time.
  • [03:48:03] <jkridner|work> ds2: you can leverage existing cold start code for the DSP.
  • [03:49:53] <ds2> jkridner: trying to figure out how easy it is to run a basic algorithym on the DSP... say just plain convolution of a finite set of data with a fixed filter and return that back
  • [03:50:10] <gmaxwell> jkridner|work, I suspected that the video stuff might be written to monopolize the DSP .. fortunately most audio codecs will easily run on the ARM... but I thought it might be useful to get audio onto the DSP to leave the arm free for crypto or UI.
  • [03:50:18] <ds2> if there is no need to figure out how to start it like in a standalone DSP, that makes it much easier
  • [03:59:34] <vlad_> I s'pose I should actually get a cross-compilation env set up instead of building xorg on the device
  • [04:00:04] <vlad_> though it's probably a tossup as to which is faster :p
  • [04:00:26] <gmaxwell> oh is the default setup for beagle board self hosting?
  • [04:00:36] <vlad_> there is no default setup
  • [04:02:01] <gmaxwell> No default OE build scripts or the like?
  • [04:09:14] <bobbarker> So does the board come with an OS other than the boot loader?
  • [04:10:21] <jkridner|work> gmaxwell: xDM-compliant codecs are designed to share the DSP.
  • [04:10:49] <jkridner|work> those codecs allow for doing both audio and video or multi-stream video on the DSP.
  • [04:12:04] <bobbarker> Or should I start with an image on an SD?
  • [04:12:16] <jkridner|work> vlad_: Xorg is already building reasonably well with OE. It would be interesting to see some better combinations of cross/native compilation that would coordinate when one was better than the other.
  • [04:12:35] <jkridner|work> bobbarker: the board comes with only a boot-loader.
  • [04:13:28] <jkridner|work> the preferred method for me is to create 2 partitions on an SD card with MLO/u-boot.bin/uImage on the first (FAT/FAT32) partition and a root file system on the second (ext3) partition.
  • [04:13:49] <jkridner|work> that makes it relatively easy to switch kernels and file systems.
  • [04:14:10] <jkridner|work> you could put the kernel into an ext2 file system, but I don't trust that yet.
  • [04:14:32] <jkridner|work> you could also just put it in the on-board flash (and the rootfs too), but I'm switching between too often.
  • [04:15:04] <bobbarker> K, why 2 partitions, and what's the problem with switching?
  • [04:15:46] <jkridner|work> if you were going to boot off of SD, it would need to be from a FAT formatted partition.
  • [04:16:00] <jkridner|work> but, FAT isn't a good format for a Linux root file system.
  • [04:16:21] <bobbarker> Oh, ok.
  • [04:16:26] <jkridner|work> using on-board flash is difficult to switch because you must overwrite it to switch.
  • [04:16:39] <vlad_> jkridner|work: yeah, I'm just not familiar with OE at the moment.. also would like to end up with something that I can get symbol info with using oprofile
  • [04:16:41] <jkridner|work> I have a small stack of SD cards.
  • [04:16:51] <vlad_> without going through gymnastics
  • [04:17:12] <vlad_> SD cards are cheap enough anyway :)
  • [04:17:20] <bobbarker> That sounds dangerous. I'll buy an SD card tomorrow. Thanks, good night all.
  • [04:17:39] <jkridner|work> good night bobbarker.
  • [04:17:43] <bobbarker> ;)
  • [04:17:45] * bobbarker (n=bob@adsl-69-109-116-44.dsl.pltn13.pacbell.net) has left #beagle
  • [04:22:19] <vlad_> jkridner: do you know if Xomap runs at on beagle? (Or, really, if it even offers any advantages over Xfbdev?)
  • [04:23:36] <vlad_> I'm hoping to set up an environment at least reasonably close to that on n810, but one that's easier to work with (e.g. has keyboard and display :)
  • [04:36:31] <jkridner|work> vlad_: sorry, I don't know what the features are of Xomap. I know that it is possible to run much of the N810 image on Beagle. The few proprietary/multimedia portions don't work.
  • [04:36:44] <jkridner|work> that is, they don't work as-is with an N810 image.
  • [04:37:05] <jkridner|work> but, Xomap runs as part of Maemo.
  • [04:37:16] <vlad_> ok, great
  • [04:37:17] <jkridner|work> Do the Mamona guys use Xomap as well?
  • [04:37:39] <vlad_> that'll help a decent bit.. if this works out I can see us (mozilla) getting a bunch of beagles for test boards
  • [04:39:05] * Guest62082 (n=root@rrcs-208-125-50-2.nys.biz.rr.com) Quit (Read error: 110 (Connection timed out))
  • [04:39:11] <jkridner|work> if you are looking to setup a farm of boards, I've found the Linksys USB200M can be powered from Beagle directly without much headache.
  • [04:39:59] <jkridner|work> with the stuff I've been doing this weekend, I don't have a monitor/keyboard/mouse connected, so I didn't even connect a USB hub.
  • [04:40:28] <vlad_> nice; i've been using a .. uh, "USB 2.0 Ethernet Adapter" probably isn't useful, where's the box for this thing
  • [04:40:32] <vlad_> "AirLink 101"
  • [04:40:33] * jkridner|work is anxious to play with boards built with a bit more RAM.
  • [04:40:35] <vlad_> seems to work great
  • [04:41:06] <jkridner|work> isn't an "AirLink 101" wireless?
  • [04:41:23] <vlad_> nope
  • [04:41:33] <jkridner|work> k. I have one that is. :)
  • [04:41:41] <vlad_> of course, I had to touch it, and now it just went down
  • [04:41:45] <jkridner|work> so, very generic. :)
  • [04:41:48] <vlad_> so maybe I take back the 'seems to work great' :)
  • [04:42:05] <jkridner|work> hmmm. connector issue?
  • [04:44:00] <vlad_> not sure; that was odd, everything seemed to be good
  • [04:44:05] <vlad_> I just reseated everything and it seems to be back
  • [04:53:50] <rsalveti> jkridner|work, vlad_: yep, at mamona we run xomap
  • [04:54:05] <rsalveti> but the same recipe at OE
  • [04:54:42] <jkridner|work> rsalveti: how/why would I switch from building Angstrom with OE to building Mamona?
  • [04:55:29] <rsalveti> jkridner|work: that's what I'm doing now, mamona branch doesn't have all patches and stuff to let it working fine on beagle
  • [04:56:00] <rsalveti> I create another branch to get the latest oe branch and integrate it with mamona's, this will be the branch we'll be using after 0.2 release
  • [04:56:45] <rsalveti> jkridner|work: if you get it today, you're going to use mostly what we got from omap3430sdp, but no armv7 support
  • [04:56:49] <rsalveti> it's a quite old branch
  • [04:57:07] <jkridner|work> btw, I really love to see S-Video going with a browser. Very important for those who cannot afford DVI-D monitors.
  • [05:00:07] <rsalveti> guess that until next weekend we'll have mamona running with beagle with all support we need (mostly what we already have at angstrom today)
  • [05:00:15] * Olipro_ (n=Olipro@unaffiliated/olipro) has joined #beagle
  • [05:03:19] <rsalveti> jkridner|work: do you know the state of s-video support with the latest omap kernel?
  • [05:03:46] <jkridner|work> it is not supported in the linux-omap kernel.
  • [05:03:51] <rsalveti> we got it working but only with an old linux-omap tarball from ti website
  • [05:04:54] <jkridner|work> yeah... code is too ugly today.
  • [05:04:58] <rsalveti> yep, at the time I was playing with it the driver was not integrated with linux-omap branch yet
  • [05:05:05] <rsalveti> hehe, true
  • [05:12:50] <vlad_> good thing I can leave this overnight, because it's taking just -pixman- like 5-10 min to build
  • [05:13:09] <vlad_> crap, I don't have the right armv7
  • [05:13:18] <vlad_> cflags
  • [05:16:33] <ds2> you can get S-video working with the top of the tree l-o but you will have it apply a patch
  • [05:17:26] * Olipro (n=Olipro@unaffiliated/olipro) Quit (Read error: 110 (Connection timed out))
  • [05:19:20] <rsalveti> ds2: do you have the patch or the address where I can get it?
  • [05:24:10] <jkridner|work> see the mailing list.
  • [05:25:46] <rsalveti> jkridner|work: true, just got it here, thanks
  • [05:29:50] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [05:32:19] <jkridner|work> good morning khasim
  • [05:42:35] <khasim> hi jkridner
  • [05:45:04] * jeh is now known as chakie_work
  • [05:52:50] * ap02 (n=abhijit@c-24-6-175-156.hsd1.ca.comcast.net) has joined #beagle
  • [06:03:47] * diego__ (n=diego@host-84-222-19-249.cust-adsl.tiscali.it) has joined #beagle
  • [06:14:30] <vlad_> rsalveti: do you have a canonical source location for Xomap?
  • [06:14:41] <vlad_> I am failing to find a repo, patch, or anything
  • [06:16:54] <rsalveti> vlad_: http://dev.openbossa.org/mamona/gitweb?p=mamona.git;a=tree;f=packages/xorg-xserver;h=d10e49bfa5f1aa7b75d83132e9b73b0670b723c6;hb=HEAD
  • [06:17:03] <vlad_> yeah, just got there
  • [06:17:07] <rsalveti> xserver-kdrive-xomap_1.3.99.0.bb
  • [06:17:23] <vlad_> so the canonical source is .. http://repository.maemo.org/pool/maemo4.0/free/x/xorg-server/xorg-server_1.3.99.0~git20070321-0osso20074202.tar.gz ?
  • [06:17:31] <vlad_> fun
  • [06:17:53] * diego_ (n=diego@host-84-222-18-165.cust-adsl.tiscali.it) Quit (Read error: 101 (Network is unreachable))
  • [06:18:19] <rsalveti> vlad_: yep
  • [06:18:27] <vlad_> wonder why it hasn't been merged upstream yet
  • [06:20:14] <rsalveti> vlad_: at oe?
  • [06:20:53] <vlad_> no, into xorg proper
  • [06:22:15] <rsalveti> vlad_: hm, don't know exactly, need to see the code
  • [06:22:36] <rsalveti> could be just because it's easier to maintain another package than pushing it upstream
  • [06:23:10] <rsalveti> ugly hacks, don't know, already saw a lot of ugly code at maemo :-)
  • [06:23:17] <vlad_> yeah
  • [06:23:33] <vlad_> that's what I'm afraid of :)
  • [06:23:39] <rsalveti> :-)
  • [06:24:32] <vlad_> huh, X build finished.. tha wasn't so bad
  • [06:25:02] <vlad_> and starts just fine
  • [06:25:13] <rsalveti> cool
  • [06:25:22] <vlad_> Xfbdev that is, haven't done Xomap yet
  • [06:41:44] * trickie (n=trickie@basesoft.xs4all.nl) has joined #beagle
  • [06:42:10] <vlad_> my usb hub is going to cause me to have seizures
  • [06:42:41] <vlad_> http://flickr.com/photos/vladphone/2769749382/ is what I get when I buy the cheapest hub I see (the light in the middle? it blinks and fades in and out)
  • [06:48:17] <rsalveti> haha :-)
  • [06:52:21] * jkridner|work (n=a0321898@nat/ti/x-9ce1603e543e615d) Quit (Remote closed the connection)
  • [06:58:17] <chakie_work> vlad_: wtf is that? a christmas tree decoration?
  • [07:15:07] <ds2> I have a hub that does odd things but that's cuz the hub doesn't meet spec and is back driving
  • [07:23:20] * sakoman_ (n=sakoman@static-74-41-60-154.dsl1.pco.ca.frontiernet.net) Quit (Read error: 110 (Connection timed out))
  • [07:24:16] <vlad_> (I had no idea the hub was going to do that until I plugged it in -- at worst I was expecting 7 bright blue LEDs)
  • [07:24:50] * sakoman (n=sakoman@static-74-41-60-154.dsl1.pco.ca.frontiernet.net) Quit (Read error: 110 (Connection timed out))
  • [07:32:27] <chakie_work> duct tape it
  • [08:00:50] <mru> morning
  • [08:06:02] * methril|laptop (n=methril@229.72.220.87.dynamic.jazztel.es) has joined #beagle
  • [08:12:22] * ap02 (n=abhijit@c-24-6-175-156.hsd1.ca.comcast.net) Quit ("Leaving")
  • [08:24:16] * methril|laptop (n=methril@229.72.220.87.dynamic.jazztel.es) Quit ("Ex-Chat")
  • [08:25:21] * methril|laptop (n=methril@229.72.220.87.dynamic.jazztel.es) has joined #beagle
  • [08:27:04] * methril|laptop (n=methril@229.72.220.87.dynamic.jazztel.es) Quit (Client Quit)
  • [08:27:12] * methril|laptop (n=methril@229.72.220.87.dynamic.jazztel.es) has joined #beagle
  • [08:27:17] * RogerMonk (n=a0740758@nat/ti/x-74a35f80fbcc2983) has joined #beagle
  • [09:02:14] <koen> good morning all
  • [09:03:53] * guillaum1 (n=gl@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  • [09:30:42] * RogerMonk (n=a0740758@nat/ti/x-74a35f80fbcc2983) Quit (Remote closed the connection)
  • [09:41:44] <ahu> koen: I'm not good with soldering iron, do you know a place in .nl where I can get the beagleboard serial cable?
  • [09:48:42] <kulve> I think that's quite standard connector that most PC motherboards still have?
  • [09:49:17] <kulve> i.e. header->rs232
  • [09:49:32] <chakie_work> i got one as part of some random newish motherboard
  • [09:49:33] * Beagle3 (n=Beagle3@195.7.58.230) has joined #beagle
  • [09:49:59] * Beagle3 (n=Beagle3@195.7.58.230) Quit (Client Quit)
  • [09:50:09] <ahu> hmmm ok - I think the mother boards I have don't have a ribbon cable to their rs232 connector
  • [09:50:16] <ahu> but I might be able to dig one up
  • [10:00:30] * Olipro__ (n=Olipro@unaffiliated/olipro) has joined #beagle
  • [10:07:52] * Fragguel (n=rafa@185.114.218.87.dynamic.jazztel.es) has joined #beagle
  • [10:09:24] <Fragguel> Hi, i've just received my beagleboard. I've connected by s-video to TV and when power on i get only some color bars in the TV. Why?
  • [10:10:04] <Fragguel> Is there any way to switch output to s-video?
  • [10:11:19] <ahu> Fragguel: same here
  • [10:11:22] <ahu> real pretty bars though
  • [10:13:32] <Fragguel> what have you done with it?
  • [10:13:47] <ahu> I need to get a serial connector going
  • [10:14:45] <Fragguel> it's possible redirect output with rs232?
  • [10:15:06] <ahu> it is more the other way aroun
  • [10:15:06] <ahu> d
  • [10:15:17] <ahu> everything goes out the rs232 unless something else is in place
  • [10:16:00] <Fragguel> but how do you get the output?
  • [10:17:15] * Olipro_ (n=Olipro@unaffiliated/olipro) Quit (Read error: 110 (Connection timed out))
  • [10:19:24] <ahu> Fragguel: right now, I don't
  • [10:19:54] <Fragguel> ok, but it's possible to get the output with rs232, how?
  • [10:20:10] <Fragguel> to my computer? which software is needed?
  • [10:20:41] <ahu> hyperterm, minicom, ser2net etc
  • [10:20:51] <Fragguel> thank you!
  • [10:48:02] <b00berfry> teraterm pro
  • [10:48:05] <b00berfry> = best
  • [10:48:07] * b00berfry is now known as uberfry
  • [10:48:47] * Fragguel_ (n=rafa@10.114.218.87.dynamic.jazztel.es) has joined #beagle
  • [10:50:53] * Fragguel (n=rafa@185.114.218.87.dynamic.jazztel.es) Quit (Read error: 110 (Connection timed out))
  • [10:51:44] * sweetlilm (n=sweet@196.37.229.15) has joined #beagle
  • [10:55:14] * Guest62082 (n=root@rrcs-208-125-50-2.nys.biz.rr.com) has joined #beagle
  • [11:00:09] * JuanG (n=Juan@nat/ti/x-0991d1dee539800a) has joined #beagle
  • [11:00:14] * JuanG (n=Juan@nat/ti/x-0991d1dee539800a) has left #beagle
  • [11:26:54] * esden`away is now known as esden
  • [11:49:56] * cian (n=cian@cian.ws) has joined #beagle
  • [11:50:32] <cian> is there any way for me to force the OTG port into host mode without shorting pins 4/5 while booting?
  • [12:06:55] * cian (n=cian@cian.ws) Quit (Remote closed the connection)
  • [12:11:22] * Fragguel_ (n=rafa@10.114.218.87.dynamic.jazztel.es) Quit (Remote closed the connection)
  • [12:23:08] * BThompson (n=BThompso@nat/ti/x-3e680c4477a965bb) has joined #beagle
  • [12:28:59] * gmaxwell (n=NT4TN@wikimedia/KatWalsh/x-0001) Quit (Remote closed the connection)
  • [12:44:34] * methril|laptop (n=methril@229.72.220.87.dynamic.jazztel.es) Quit (Read error: 104 (Connection reset by peer))
  • [12:45:32] * DDevine (n=devine@123-243-187-85.static.tpgi.com.au) has joined #beagle
  • [12:48:04] <DDevine> Hey, I just want to make sure that when I connect a screen into S-Video will that work during boot?
  • [12:49:41] * methril|laptop (n=methril@229.72.220.87.dynamic.jazztel.es) has joined #beagle
  • [12:50:58] * keesj (n=keesj@ip49-193-210-87.adsl2.static.versatel.nl) Quit (Read error: 113 (No route to host))
  • [12:53:11] <DDevine> oh "you should see coloured stripes"
  • [12:54:17] <jkridner> I have an OE mystery. http://tinderbox.openembedded.net/packages/102059/ failed due to soup-date.h being renamed soup-date.`
  • [12:55:02] <jkridner> DDevine: correct.
  • [12:55:10] <jkridner> good morning all
  • [12:56:00] <DDevine> do I need a null modem? or can i just interface through USB?
  • [12:58:34] * jkridner|work (n=a0321898@nat/ti/x-45f179a99756dafe) has joined #beagle
  • [12:59:37] <jkridner> using existing software, you'll need a null-modem.
  • [13:01:29] <DDevine> what do you mean existing software?
  • [13:01:50] <DDevine> like "what is installed in Windows"?
  • [13:02:16] <jkridner|work> no, as in Angstrom, or the diagnostic kernel or other existing beagle board distros/projects.
  • [13:02:41] <DDevine> ah
  • [13:02:43] <jkridner|work> no one has done a good enough USB bootloader to avoid needing a null-modem.
  • [13:03:03] <DDevine> i though you could make a TCP/IP connection though the USB.
  • [13:03:40] * BotOnBeagle (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
  • [13:03:45] <jkridner|work> you can.
  • [13:03:59] <jkridner|work> but, you need to boot a kernel first.
  • [13:04:20] <DDevine> ok.
  • [13:06:30] * jkridner|work finds joy in having his Beagle running a copy of the web server and connecting to the IRC channel to collect logs.
  • [13:06:56] <DDevine> I will definitely ask what stuff I need to buy when It comes time to order my beagle.
  • [13:18:50] * sakoman (n=sakoman@static-74-41-60-154.dsl1.pco.ca.frontiernet.net) has joined #beagle
  • [13:21:42] * cbrake_away is now known as cbrake
  • [13:25:06] * Fragguel (n=rafa@84.114.218.87.dynamic.jazztel.es) has joined #beagle
  • [13:27:33] * [-ip-] (n=ts@mnhm-590e08cb.pool.einsundeins.de) has joined #beagle
  • [13:29:37] * BotOnBeagle (n=PircBot@c-76-31-18-64.hsd1.tx.comcast.net) Quit (Remote closed the connection)
  • [13:33:19] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
  • [13:37:05] * NishanthMenon (n=gnat@nat/ti/x-a7cb5db4763afe61) has joined #beagle
  • [13:40:36] * rsalveti (n=salveti@189.70.119.44) Quit (Remote closed the connection)
  • [13:44:37] * Olipro__ is now known as Olipro
  • [13:45:10] * suihkulokki (i=rvoipio@nat/movial/x-0e0cfe2c2570d3f3) Quit (Read error: 60 (Operation timed out))
  • [13:47:02] <Crofton|work> koen, back from the party?
  • [14:06:10] * rsalveti (n=salveti@200.184.118.132) has joined #beagle
  • [14:27:07] * sakoman_ (n=sakoman@static-74-41-60-154.dsl1.pco.ca.frontiernet.net) has joined #beagle
  • [14:38:21] * Fragguel (n=rafa@84.114.218.87.dynamic.jazztel.es) Quit ("Ex-Chat")
  • [14:46:35] * JoeBorn__ is now known as JoeBorn
  • [14:48:36] * dirk2 (n=dirk@F33af.f.strato-dslnet.de) has joined #beagle
  • [14:54:55] * jkridner|work just found https://www-a.ti.com/downloads/sds_support/targetcontent/psp/omap35x/index.html, such that you don't have to have a registered board serial number.
  • [14:55:20] * suihkulokki (i=rvoipio@nat/movial/x-b448bceb91309c1a) has joined #beagle
  • [14:55:51] <dcramer> anyone have instructions on how to get ethernet gadget running on linux
  • [14:56:01] <dcramer> specifically the bridging so the device can get out ?
  • [14:56:13] * sakoman_ (n=sakoman@static-74-41-60-154.dsl1.pco.ca.frontiernet.net) Quit (Remote closed the connection)
  • [14:57:19] <dirk2> While sharing new links, I have http://www.eetimes.eu/showArticle.jhtml?articleID=210100653&cid=NL_eeteu
  • [14:57:40] * johnx (n=john@p2172-ipbf2302hodogaya.kanagawa.ocn.ne.jp) has joined #beagle
  • [15:03:53] <dirk2> jkridner|work: ping
  • [15:04:32] <jkridner> pong
  • [15:05:23] <jkridner> dirk2: ?
  • [15:06:12] <sweetlilm> just got my board, and SD card formatted correctly etc. I don't have a serial port connect, is this required to boot the board off SD (seems so?)?
  • [15:06:26] <dcramer> yes
  • [15:06:30] <sweetlilm> bugger
  • [15:06:35] <jkridner> dcramer: http://groups.google.com/group/beagleboard/browse_thread/thread/aaf290080d1f0c24
  • [15:06:38] <dcramer> well, you don't need one
  • [15:06:40] <sweetlilm> I think I have a cable at home somewhere
  • [15:06:50] <dcramer> well, the cable is not common
  • [15:06:55] <dcramer> make sure you look carefully
  • [15:07:04] <sakoman> dirk2: any comments on Mani's approach to consolidate common omap3 u-boot files. Do you think it will be acceptable upstream?
  • [15:07:09] <sweetlilm> not standard IDC->DB9 ?
  • [15:07:21] <dcramer> sweetlilm: yes, not standard
  • [15:07:27] <dirk2> jkridner: What do you think about moving Beagle FAQ to wiki, e.g. to http://elinux.org/BeagleBoardFAQ ? It would be easier to edit and contribute for people.
  • [15:07:27] <jkridner> not all IDC->DB9 are the same.
  • [15:07:34] <jkridner> it *is* standard: AT/Everex.
  • [15:07:45] <jkridner> just may not be the same as what you already have.
  • [15:07:46] <dcramer> jkridner: well, I said not common
  • [15:07:46] <sweetlilm> I was going to use one off an old PC
  • [15:07:53] <jkridner> it is what is used for PC risers.
  • [15:07:58] <dcramer> sweetlilm: tis about 5 min soldering work
  • [15:08:01] <sweetlilm> aha, then I should be good
  • [15:08:44] <sweetlilm> actually... I have and 10 pin IDC connector with a ribbon on it... and some DB9 shells... :)
  • [15:08:54] <dcramer> sweetlilm: there ya go
  • [15:09:03] <jkridner> how about I put a link to the eLinux FAQ on http://beagleboard.org/support/faq?
  • [15:09:14] <sweetlilm> people look at me oddly when I fire up the iron at work :)
  • [15:09:15] <dirk2> sakoman: Had a quick look to his first, NAND, patch. Looks to me that this is the one we already applied. Have to do a detailed diff to check if there any detailed differences. And if there are no differences, I like to ask Mani if he tested this patch at runtime
  • [15:09:24] <jkridner> make sure you think about the serial cable as well. null modem was used.
  • [15:09:58] <sweetlilm> i was going to plug it into a USB->serial converter
  • [15:10:49] <dirk2> jkridner: We don't have any FAQ at eLinux yet (you can link to ;) ). I first want your opinion about this before I create something like http://elinux.org/BeagleBoardFAQ
  • [15:11:21] <jkridner> I agree that people have an easier time editing the elinux.org faq.
  • [15:11:34] <jkridner> I like that beagleboard.org is editable, but that people have to think about it.
  • [15:12:15] <dirk2> jkridner: So you like the idea to create eLinux FAQ and linking to it from beagleboard.org ?
  • [15:12:21] <jkridner> yes.
  • [15:13:06] <dirk2> :) Let's try it then. Have to think about some first entries, though. Don't want to steal your content ;)
  • [15:14:51] <dirk2> jkridner: And an other question: Can you share what you have in mind for http://beagleboard.org/project/documentation/ ?
  • [15:16:25] <dirk2> sakoman: Will have a look to Mani's second, the make common stuff common, patch, too
  • [15:17:29] <jkridner> I plan to start with the full source for the hardware reference manual...
  • [15:17:44] <jkridner> then, I plan to work with people building software tutorials.
  • [15:18:14] <jkridner> given the lift we've gotten out of Angstrom, I think we'd give step-by-step and background on using OE on Beagle.
  • [15:18:50] * trickie (n=trickie@basesoft.xs4all.nl) Quit ("Leaving")
  • [15:18:51] <jkridner> I don't have a time-line, but when I start to engage with more universities, I'll be looking for some champions for various portions of that text.
  • [15:19:29] <dirk2> Maybe I like to add some text, too.
  • [15:19:34] <jkridner> please feel free to steal any of my content if it is useful in any way. :)
  • [15:19:43] <jkridner> great!
  • [15:20:00] <jkridner> We could also use text in German.
  • [15:20:22] <jkridner> In fact, I'd suggest you source in whatever language you feel comfortabe.
  • [15:20:25] <jkridner> comfortable.
  • [15:20:35] <jkridner> I believe cnx.org understands how to deal with translations.
  • [15:22:23] <dirk2> The software tutorials sounds great. But I wonder if we have to 'copy' the whole hardware TRM with all its details. Yes, we need a hardware section. But do we need all the datails, or will it be sufficent to link to existing documentation? E.g. you will link to OMAP3 TRM, too, and not copy it.
  • [15:24:08] <jkridner> I wouldn't bring in the TRM.
  • [15:24:35] <jkridner> you can choose what "chapters" to pull into the manual.
  • [15:24:56] <jkridner> I think there is good information in the board hardware reference that also needs to be extended over time.
  • [15:25:46] <jkridner> I think it would be a good starting point for a text, but I do want to work towards reformatting into material most suitable for a textbook, perhaps with a somewhat large reference section.
  • [15:26:01] <dirk2> okay. Sounds we need a "starting" chapter we can use as example and then extend it.
  • [15:26:20] <jkridner> I could see bringing in some portions of the TRM, but focused on software developers.
  • [15:26:41] <jkridner> good point.
  • [15:28:08] * feig (n=ejf3@215.sub-75-222-127.myvzw.com) has joined #beagle
  • [15:29:04] <sakoman> dirk2: OK, thanks. I'll await your further comments
  • [15:42:50] <dirk2> jkridner: Hmm, for http://beagleboard.org/faq I only get the login page (even if I'm already logged in). I think yesterday I got the FAQ content (?). My fault or anything wrong here?
  • [15:42:52] * Beagle3 (n=Beagle3@63.65.238.105) has joined #beagle
  • [15:43:13] <jkridner> guess I should put a forward link. FAQ is at http://beagleboard.org/support/faq
  • [15:43:28] <jkridner> I'll create a http://beagleboard.org/faq.
  • [15:44:14] <dirk2> http://beagleboard.org/support links to http://beagleboard.org/faq
  • [15:44:50] <Crofton|work> Here's an avi for beagle testing ....
  • [15:44:53] <Crofton|work> http://www.balister.org/~balister/shoulder-xray-dye/ax-ta.avi
  • [15:48:29] <jkridner> dirk2: it does? Can you check again? do you need to shift-RELOAD?
  • [15:48:42] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [15:49:39] <kulve> it does
  • [15:50:34] <dirk2> jkridner: Works now :)
  • [15:52:33] <dcramer> btw, the instructions for briding eth0 and usb on linux didn't work for me
  • [15:53:08] <jkridner> :( can you reply to the e-mail and describe what you did and what the behavior was?
  • [15:53:16] <dirk2> dcramer: Link?
  • [15:53:34] <dcramer> jkridner: I did
  • [15:53:35] <dcramer> http://groups.google.com/group/beagleboard/browse_thread/thread/aaf290080d1f0c24
  • [15:53:41] <jkridner> thanks.
  • [15:53:58] <dcramer> jkridner: I think the host has to run a dhcp server for that to work
  • [15:54:01] <dcramer> not sure though
  • [15:54:08] <dcramer> packets were getting to the host, but not any further
  • [16:01:00] <feig> Hey, I'm new here and have a *hopefully* quick question, I've been looking to pick up a board, but digi-key is showing 0 available, i read on the google-groups page that the qty is not likely accurate, but i also read that after a month delay, there will be more manufactured
  • [16:01:17] <feig> will the next manufactured batch be the rev c boards?
  • [16:04:36] <dirk2> feig: I don't think so. Most probably rev C boards will be there earliest at november. And I don't think that next boards will have a month delay, but jkridner will have better information.
  • [16:04:49] * cian (n=cian@cian.ws) has joined #beagle
  • [16:04:51] <feig> well, that post was Jul 30
  • [16:05:08] <feig> so, i should jump now? or are they actually sold out?
  • [16:06:00] <cian> is g_ether compiled into the angstrom provided kernel? I can't seem to find it in /lib/modules...
  • [16:07:16] <dirk2> feig: Wait some minutes, maybe jkridner can give you more up to date information
  • [16:07:24] <feig> sure
  • [16:10:55] * quickdev (n=quickdev@p5B053902.dip.t-dialin.net) has joined #beagle
  • [16:12:18] <dirk2> sakoman: Mani's NAND patch is 100% what we have already git, except the Signed-off-by . wonder ...
  • [16:12:21] * keesj (n=keesj@ip49-193-210-87.adsl2.static.versatel.nl) has joined #beagle
  • [16:18:04] * robclark (n=robclark@nat/ti/x-5be14ec66906c341) has joined #beagle
  • [16:20:23] <sakoman> dirk2: OK, won't bother with that one :-)
  • [16:20:32] <sakoman> Wonder if it works :-)
  • [16:21:36] <quickdev> are there some interesting use cases of the beagle board? :)
  • [16:23:42] <dirk2> sakoman: If it works: Me too ;) So I asked him.
  • [16:25:01] <Crofton|work> quickdev, all depends on what you consider interesting
  • [16:25:27] <dirk2> sakoman: Regarding the make common stuff common: Looking only at the DIFF STAT results, hoping that only code move and no functional changes are in this patch
  • [16:26:10] <sakoman> dirk2: is the structure something that would be accepted upstream?
  • [16:26:12] <dirk2> First, yes, looks okay to me. At least to separate common stuff.
  • [16:26:42] <jkridner> feig, dirk2: what was the question?
  • [16:26:52] <jkridner> oh...
  • [16:26:54] <jkridner> Rev C.
  • [16:27:09] <jkridner> the next 1600 boards built will all be Rev B5.
  • [16:27:24] * asterick (i=lemoncar@svn.rkiq.net) has joined #beagle
  • [16:27:44] * quickdev (n=quickdev@p5B053902.dip.t-dialin.net) Quit ("Verlassend")
  • [16:27:47] <jkridner> those will likely sell out pretty quickly.
  • [16:27:58] <asterick> what are we talking about?
  • [16:27:59] <jkridner> boards sold in November will likely be Rev C.
  • [16:28:04] <dirk2> jkridner: Any eta for the next boards available at digi key?
  • [16:28:18] <jkridner> I've heard this week.
  • [16:28:38] <dirk2> sakoman: (a) have to look at sys_info.c if it can be moved to common
  • [16:29:33] <dirk2> sakoman: (b) Then we can check if overo.c, omap3530beagle.c and omap3evm.c have some additional common stuff.
  • [16:30:19] <dirk2> sakoman: Maybe we just can rename then to omap3/beagle/beagle.c omap3/overo/overo.c omap3/evm/evm.c
  • [16:31:44] <dirk2> sakoman: (c) Then we should check if board/omap3/common is okay. Davinci uses cpu/arm926ejs/davinci
  • [16:32:13] <sakoman> dirk2: do you know who to check with?
  • [16:32:52] <dirk2> sakoman: There is already cpu/omap3 Maybe we should just move stuff in board/omap3/common to cpu/omap3
  • [16:33:06] <sakoman> dirk2: regarding the rename -- shouldn't the names match the official linux machine names?
  • [16:34:28] <sakoman> dirk2: I actually like the idea of moving the common stuff to cpu/omap3 No sense creating additional directory complexity
  • [16:34:32] <koen> Crofton|work: yeah, I got back this morning straight after my shift :)
  • [16:34:43] <dirk2> sakoman: to check with and names: I think, while U-Boot list is quite strict with patch rules (40k limit, git-send-mail etc.), they are not so strict with patch content, e.g file names and directory layout
  • [16:34:51] <Crofton|work> fun
  • [16:34:51] <Crofton|work> good time?
  • [16:34:54] <ds2> b5? what's the diff between b4 and b5?
  • [16:34:55] <sakoman> dirk2: which do you think would be viewed more favorably by the gods upstream?
  • [16:35:18] <koen> Crofton|work: we had time to do like http://www.youtube.com/watch?v=V1RJxtCq_x0
  • [16:35:22] <feig> thanks for the answer
  • [16:35:43] <jkridner> ds2: removal of C70 to fix 32kHz clock (avoids GPT1 issue).
  • [16:35:50] <dirk2> feig: Sounds like time to order :)
  • [16:35:56] <feig> sounds like it =)
  • [16:36:07] <asterick> did they fix the C70 issue in the current \ next rev?
  • [16:36:11] <feig> it's going to be the brains of a robot my friend and i are working on
  • [16:36:13] <feig> i'm excited
  • [16:36:27] <asterick> Nice. :D
  • [16:36:36] <asterick> Do you have enough left over I/O for a robot?
  • [16:36:42] <asterick> (I assume you're muxing it with an FPGA or something)
  • [16:37:03] <feig> microcontrollers for the lower level stuff
  • [16:37:13] <feig> beagleboard is to be used for higher level control
  • [16:37:31] <feig> my friend is doing all the low level stuff, he wants to clean out his parts cabinets
  • [16:37:32] <ds2> jkridner: so is that the finally "blessed" fix for the GPT1 issue on rev Bn boards?
  • [16:38:36] <jkridner> it has been confirmed and it is known to help.
  • [16:38:37] <dirk2> sakoman: Let us try with common stuff in cpu/omap3 If anybody complains we can say "we did the same like DaVinci" ;)
  • [16:39:22] <koen> any big beaglenews the past few days?
  • [16:39:40] <jkridner> I'm not 100% comfortable that we shouldn't be doing something extra with the GPT1 to prevent serial hang on ill-behaved clocks.
  • [16:40:36] <sakoman> dirk2: should we ask Mani to re-craft a patch, or should I just start moving common files to cpu/omap one at a time?
  • [16:40:38] <jkridner> asterick: I/O resources depend on bandwidth required for control as well. You can always throw on I2C-based GPIO expanders, etc.
  • [16:41:15] <asterick> jkridner: indeed.
  • [16:41:22] <dirk2> sakoman: I propose to apply Mani's patch and then move files. Is it easy in git? Then we don't need an additional patch for the move.
  • [16:41:33] <jkridner> koen: big news may be that the boards are selling rather quickly and we keep running out.
  • [16:41:59] * paul_pwsan (n=paul@utopia.booyaka.com) has joined #beagle
  • [16:42:00] <jkridner> or, it could be getting help with silly Angstrom issues. :)
  • [16:42:36] <sakoman> dirk2: it is extremely easy to move files in git
  • [16:42:37] <ds2> what was the original goal of C70? a simple 1st order LPF to keep down harmonics?
  • [16:43:00] <dirk2> sakoman: Then let's do it without additional patch
  • [16:43:06] <jkridner> given all the folks that are here, is it fair to say that ASoC is the next Beagle kernel priority, or are people looking ahead at the power management stuff?
  • [16:43:22] <Crofton> ASoC ....
  • [16:43:43] <asterick> ALSA SoC layer
  • [16:43:48] <dirk2> ASoC, Display configuration & timing
  • [16:43:56] * koen wants cpufreq working
  • [16:43:57] <sakoman> dirk2: I agree it will be easier to understand what is going on if I move common files one at a time using the git commnads
  • [16:44:07] <jkridner> I'm on "vacation" this week, so I might get to play. :)
  • [16:44:18] * dcordes (n=dcordes@unaffiliated/dcordes) has joined #beagle
  • [16:44:21] <sakoman> So I will start with the really obvious "safe" files that can be shared
  • [16:44:43] * jkridner says this despite having already been on 3 hours of conference calls today.
  • [16:44:47] <koen> sakoman: the install issues could be that one of yor machines doesn't have install-native (or another package that provides install.sh) in staging
  • [16:44:53] <sakoman> jkridner: I agree with dirk2's list
  • [16:45:30] <koen> sakoman: I used to have the same problem and opted tp patch -m0644 to -m 0644 instead of finding the root cause
  • [16:45:51] <sakoman> koen: I resolved the issue that way :-)
  • [16:45:53] <dirk2> sakoman: Then, in board/omap3 we can remove the omapXXXX from the file/directory names
  • [16:45:54] <jkridner> I think I'm going to have to break down and start drawing a diagram of the clock trees to help understand the display timings.
  • [16:46:19] <jkridner> let me know if anyone has started such a thing.
  • [16:46:51] <sakoman> koen: you will likely run into the same issue when you upgrade your kernel recipe to linux-omap 2.6.27-rc3
  • [16:47:06] <dirk2> sakoman: Result: omap3/overo/overo.c omap3/beagle/beagle.c omap3/evm/evm.c
  • [16:47:16] <sakoman> You will also see the omapfbplay will break because of the header file moves
  • [16:48:05] <koen> sakoman: poke mru about that :)
  • [16:48:10] <dcordes> koen: do you have a fbset mode file for the demo image? I can't figure how to set the resolution
  • [16:48:22] <koen> dcordes: it's a KConfig option
  • [16:48:30] <dirk2> sakoman: Then I will have a look to if there is more common stuff in sys_info.c and (overo.c evm.c beagle.c) that can be moved to cpu/omap3
  • [16:48:45] <dcordes> koen: so with the demo image, only xga is possible?
  • [16:48:52] <dcordes> s/image/kernel/
  • [16:49:10] <koen> yes
  • [16:49:32] <dirk2> sakoman: And then ..... the first attempt to send a patch (which one? evm? beagle? overo?) to U-Boot list ?
  • [16:50:33] <dirk2> sakoman: And then, depending on the comments from U-Boot list, we re-arrange everything again ;))
  • [16:50:35] <dcordes> another thing: how does the c70 problem show? I seem to have a solid system with the demo image on beagle.
  • [16:50:47] <sakoman> dirk2: your choice :-) I would vote Overo, but this is #beagle so it should probably be beagle ;-)
  • [16:52:23] <dirk2> sakoman: btw: We will get cpu/omap3/nand.c (the "common" nand). But where is OneNAND located in U-Boot?
  • [16:56:50] <sakoman> dirk2: we are able to use the standard onenand stuff in drivers/mtd/onenand
  • [16:58:47] <sakoman> dirk2: I probably won't get to the changes till this evening
  • [16:59:13] <sakoman> I'll do them in a separate branch for your review and then we can merge them if you find them acceptable
  • [16:59:35] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [17:00:01] <dirk2> sakoman: Okay, give me some time, too ;)
  • [17:00:48] <dirk2> sakoman: Just had a look to http://www.sakoman.net/cgi-bin/gitweb.cgi?p=u-boot-omap3.git;a=commit;h=cfa460adfdefcc30d104e1a9ee44994ee349bb7b
  • [17:01:09] <dirk2> sakoman: There is one board/*/common
  • [17:01:25] <dirk2> sakoman: and two nand drivers in cpu *
  • [17:01:35] <sakoman> Crofton: FWIW, the asoc driver seems to work a bit better with the latest kernel. On Overo it does not crash using either mplayer -ao alsa or mplayer -ao oss
  • [17:01:51] <dirk2> sakoman: So, most probably, U-Boot upstream will accept both.
  • [17:02:22] <dirk2> sakoman: Let us do it as discussed above: Move common stuff to cpu/omap3, then we are in sync with DaVinci
  • [17:03:33] <sakoman> dirk2: OK, works for me
  • [17:03:50] <sakoman> we have a 50% chance of getting it right ;-)
  • [17:04:01] * valhalla (n=valhalla@81-174-21-16.dynamic.ngi.it) has joined #beagle
  • [17:04:36] <sakoman> Crofton: but sadly dialtone still crashes. IIRC it crashes on TI's alsa driver too, doesn't it?
  • [17:05:20] <dcordes> is it possible to play sounds with oss only in the kernel?
  • [17:05:37] <paul_pwsan> jkridner, jkridner|work: i believe mru has a patch for some of the DSS clock timings
  • [17:05:47] <cian> maybe you already know this, but older u-boot versions cause kernel panics when using xmms
  • [17:06:00] <cian> latest from git works fine
  • [17:06:50] <jkridner> paul_pwsan: k. I believe those patches did not get to the 74.25MHz pixel clock required for some 720p HDTVs.
  • [17:07:15] <dcordes> cian: you mean latest kernel source fixes alsa?
  • [17:07:42] <cian> no, with the same kernel version I see crashes when using the stock u-boot that shipped on my nand
  • [17:07:58] <cian> but no crashes using a u-boot i built from git
  • [17:08:31] <dcordes> why would the bootloader affect a problem with alsa?
  • [17:08:34] <jkridner> any of the OE guys seen problems finding boost (http://tinderbox.openembedded.net/public/logs/775614.txt) ?
  • [17:08:38] <cian> i have no idea
  • [17:08:47] <cian> but i noticed this by chance today
  • [17:09:02] <dcordes> that's very strange I think
  • [17:09:12] <cian> strange but reproducable
  • [17:09:33] <cian> using the U-Boot from June that ships on the Rev B4 boards
  • [17:09:44] <cian> and angstrom from http://www.angstrom-distribution.org/demo/beagleboard/
  • [17:10:08] <cian> cat /dev/dsp will cause a kernel panic
  • [17:10:15] <dcordes> yes
  • [17:10:51] <cian> if you upgrade u-boot.bin, and keep the same kernel, no panic
  • [17:11:01] <sakoman> cian: what is the kernel panic?
  • [17:11:03] * grummel (n=christia@cl-13.dus-01.de.sixxs.net) Quit (brown.freenode.net irc.freenode.net)
  • [17:12:15] <cian> sakoman: http://pastebin.ca/1177720
  • [17:12:26] <paul_pwsan> jkridner: very possible. commits from mru's git btw are 2b7b958dc79e51127d7a4ecf88ce12dbc6c31426, e2de5e5578fbaa9b4b75074796da0608fc93e6ae, 01ee28c50701caa94739e764c3dae9298edd8216
  • [17:12:40] <paul_pwsan> i have not tested them personally
  • [17:13:23] <paul_pwsan> i suppose that using the DSI PLL to generate the exact frequency would be the next step
  • [17:13:35] <sakoman> cian: thanks
  • [17:14:07] <cian> as you can see that crash was caused by running xmms
  • [17:14:18] <cian> but cat /dev/dsp does the same thing
  • [17:14:27] <dcordes> sakoman: I get the same with the shipped uboot+demo image.
  • [17:15:07] <sakoman> what kernel/rootfs are you using?
  • [17:15:17] * grummel (n=christia@cl-13.dus-01.de.sixxs.net) has joined #beagle
  • [17:15:19] <sakoman> is it the OE demo image?
  • [17:15:58] <dcordes> sakoman: http://pastebin.ca/1177723 OE demo image/kernel and shipped uboot here.
  • [17:16:06] <cian> I'm using the rootfs from http://www.angstrom-distribution.org/demo/beagleboard/
  • [17:16:23] <cian> and the Kernel is Linux version 2.6.26-omap1 (koen@bitbake) (gcc version 4.2.1) #1 Tue Aug 5 20:47:04 CEST 2008
  • [17:16:32] <dcordes> koen: is there a way to make the windowmanager faster? it seems very wonky
  • [17:17:03] <koen> dcordes: dunno, you'd have to ask raster
  • [17:17:32] <dcordes> koen: raster?
  • [17:19:15] <sakoman> dcordes, cian: thanks. more data is good to have!
  • [17:19:42] <cian> sakoman: out of interest, what is the best source of new kernels be
  • [17:19:43] <dcordes> sakoman: did you see cian sayin latest uboot git code fixes the problem?
  • [17:22:10] <cian> sakoman, dcordes: to clarify, U-Boot 1.3.3 (Jul 10 2008 - 16:33:09) doesn't work, U-Boot 1.3.4-rc2 (Aug 8 2008 - 08:25:54) (from gstrom-distribution.org/demo/beagleboard) and git from a week ago both do
  • [17:22:20] <charliex> heyo, someone asked me to ask again today, digikey are out of stock on the board, they said 7-9 weeks lead time, does anyone know if thats correct?
  • [17:22:50] <asterick> from what I've heard, it's a pretty safe estimate.
  • [17:22:58] <asterick> low turnover, high demand.
  • [17:23:45] <charliex> ok thanks, can't blame me for being hopeful ;)
  • [17:23:51] <asterick> hehe.
  • [17:24:12] <asterick> I'm kinda glad they are impossible to get right now. Keeps me from impulse buying one with money I don't have. xD
  • [17:25:25] <charliex> hah, yeah i know what you mean, i'm switching from a PC104 ARM 200Mhz SBC that i had to guy buying extra bits for it, and eventually come to realise its just not upto what i want, so i asked for rec's and someone pointed me to the beagle, it looks v interesting though, but i know i'll restarting a lot of the, need this, need that etc
  • [17:25:33] * khasi1 (n=a0393720@192.163.20.231) has joined #beagle
  • [17:26:16] <asterick> hehe
  • [17:26:31] <asterick> At least it's expandable. could be a LOT worse than the BB.
  • [17:26:40] <asterick> I have so many dev boards I don't use.
  • [17:26:48] <asterick> Just auctioned off like 2 of my old FPGA boards on ebay. xD
  • [17:27:29] <charliex> me too, comfile, display 3000, ts arm, lots of little AVR boards, a couple of FPGA's etc.
  • [17:27:32] <dirk2> charliex, asterick: Eta of new boards at digi key: http://www.beagleboard.org/irclogs/index.php?date=2008-08-18#T16:26:42
  • [17:27:46] * asterick is scared to look.
  • [17:27:58] <asterick> I'm tempted to try to just get one of the first released of the pandora (since the hardware is so similar)
  • [17:29:12] <charliex> ok so this week?
  • [17:29:20] <dcordes> cian: so if I flash this http://www.angstrom-distribution.org/demo/beagleboard/u-boot.bin sound should work?
  • [17:29:30] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [17:29:31] <cian> yes
  • [17:29:41] <cian> infact if you just put it on an MMC
  • [17:29:42] <asterick> Where does u-boot sit anyway?
  • [17:29:51] <asterick> is it in the internal OMAP rom? or on the 256MB memory
  • [17:29:56] <cian> and hold down the User button during power on/reset
  • [17:30:05] <cian> asterisk: it's in the nand
  • [17:30:27] <dcordes> cian: I have it on the first paritition of my sd
  • [17:30:31] <asterick> Yea, I guess making the OMAP rom user servicable would lead to a lot more problems than it's worth. xD
  • [17:30:51] <sakoman> cian: I would say that koen's OE stuff is probably the best source of fairly up to date kernels
  • [17:30:52] <cian> dcordes: you'll need the MLO file also
  • [17:31:07] <cian> sakoman: do you have a URL?
  • [17:31:08] <sakoman> cian: mine are a bit more bleeding edge
  • [17:31:20] <sakoman> cian: URL for?
  • [17:31:21] <cian> ok, I'd be interested to play with both...
  • [17:31:35] <cian> url for built kernels
  • [17:31:51] <cian> currently I'm ssh'ing into a box at home to build them, I've only got Macs/BSD/Windows at work
  • [17:32:13] <sakoman> cian: http://www.sakoman.net/feeds/omap3/glibc/images/
  • [17:32:18] <cian> thanks
  • [17:32:25] <charliex> ok placed my order, thanks chaps.
  • [17:32:26] <sakoman> be warned they are all bleeding edge
  • [17:32:41] * guillaum1 (n=gl@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) has joined #beagle
  • [17:32:57] <cian> sakoman: may set my dog on fire, etc, got the message :)
  • [17:33:12] <sakoman> and missing the beagle-demo-image (since recent kernel include file changes break ffmpeg and omapfbplay)
  • [17:33:51] <sakoman> but you could use overo-demo-image (when it finishes building shortly) or omap3evm-demo-imae or consol-image
  • [17:34:11] <dcordes> cian: beagle doesn't report anything about flashing on the serial console when I hold down user while powering on.
  • [17:34:13] <sakoman> if you can spel better than me :-)
  • [17:34:46] <cian> dcordes: holding down user just changes the boot order
  • [17:34:48] * docelic (n=docelic@78.134.193.52) has joined #beagle
  • [17:34:56] <sakoman> dcordes: yes I saw the note about u-boot effect on alsa
  • [17:35:13] <cian> dcordes: bottom of http://code.google.com/p/beagleboard/wiki/BeagleBoardDiagnostics has notes on flashing nand
  • [17:35:46] <cian> dcordes: there is no need to replace x-load, stock works fine with newer u-boot
  • [17:36:11] <cian> sakoman: I will format a new SD card and give some of those images a try, I'll let you know how I get on...
  • [17:36:49] <dcordes> cian: ok thanks
  • [17:36:53] <cian> for now, I'm going insane trying to make epiphany use a proxy server
  • [17:37:38] <sakoman> overo-demo-image has midori browser, perhaps that would work better for you
  • [17:37:47] <cian> will try...
  • [17:38:12] <sakoman> you could just try to opkg install midori to try it with your current image
  • [17:39:18] <cian> For the moment I've just put a custom firewall between my beagle & the network here to transparently redirect through the proxy
  • [17:39:26] <docelic> Hey folks, which cable to I need to connect rev.4 board's USB port to an usb hub? Since the hub has a mini connector and the board has a mini connector, I need mini-to-mini cable, or how?
  • [17:39:50] <mru> you need mini-A to B
  • [17:40:09] <mru> or mini-anything to B if you forego automatic host/gadget mode selection
  • [17:40:13] <cian> mru: is there any alternative to shorting pins 4&5?
  • [17:40:24] <jkridner> mini-A-male to standard-A-female
  • [17:40:57] <cian> thanks
  • [17:40:59] <jkridner> I guess hub could have a female-B.
  • [17:41:27] <dcordes> cian: now that I flashed u-boot, it doesn't power on anymore.
  • [17:41:52] <dcordes> only pwr led lights, not pmu
  • [17:41:57] <cian> dcordes: what output do you get on serial?
  • [17:42:02] <cian> Do you see x-loader?
  • [17:42:03] <dcordes> no errirs
  • [17:42:07] <jkridner> so, options are mini-A-male to standard-A-female (most flexible) or mini-A-male to standard-B-male. Other options require forcing the mode in the kernel.
  • [17:42:55] <dcordes> cian: oh yes I erased but didn't flash
  • [17:43:04] <dcordes> cool did I just brick it?
  • [17:43:20] <cian> dcordes: not permenantly
  • [17:43:59] * esden is now known as esden`away
  • [17:44:13] <cian> dcordes: from http://www.angstrom-distribution.org/demo/beagleboard/ put MLO, u-boot.bin, uImage & x-load.bin.ift on a fat formatted SD card
  • [17:44:25] <cian> then hold down the user button and press reset
  • [17:44:33] <cian> the beagle should boot off the SD card
  • [17:44:33] <docelic> mru, jkridner: my hub has 1 mini-B port, and 6 normal A ports.. and the cable that comes with it is A -> mini-B.. Since mini-B must go into beagle, can I connect A into one of the 6 ports on the hub, or that wouldn't work?
  • [17:44:54] <dcordes> cian: ok
  • [17:45:14] <cian> docelic: that wouldn't work
  • [17:45:21] <jkridner> docelic: in general, you won't want to use a mini-B on the Beagle if using it as a host. It won't trigger the OTG kernel to be in host mode.
  • [17:45:47] <jkridner> also, the a ports on your hub expect to face downstream.
  • [17:46:15] <jkridner> USB has an upstream and a downstream, unlike 1394, and you cannot reverse it.
  • [17:46:30] <mru> koen: omapfbplay header location fixed
  • [17:46:35] <jkridner> oh, if only 1394 had won.
  • [17:46:47] <docelic> jkridner, alright, but what can I use except mini-B on the board for usb? I don't see it
  • [17:46:51] <cian> jkridner: do all mini-A's have 4&5 shorted?
  • [17:47:00] <cian> I've been hacking mini-B cables with a glue gun and a soldering iron
  • [17:47:17] <cian> and using a big-A gender changer
  • [17:48:16] <jkridner> cian: I don't know the exact electrical connection--just know mini-A pulls the ID pin at the right time to the right level.
  • [17:48:41] <jkridner> I expect what you are doing will work (as you have seen).
  • [17:50:09] <cian> jkridner: what I'm doing works, it's just a waste of time, http://en.wikipedia.org/wiki/Universal_Serial_Bus#USB_cables says Type A has 4 pulled to Gnd and Type B doesn't
  • [17:50:20] <khasi1> cian: there is a limit to length of the cable I believe,
  • [17:50:32] <khasi1> cian: keep the length as short as possible
  • [17:50:33] <cian> yup, 30 meters
  • [17:50:40] <cian> I'm well under that
  • [17:50:42] <docelic> jkridner, I see two things related to usb on the board.. one is the mini usb connector and it says "usb otg" beside it, and the other is "usb host / p7", but it's just the title and there's no slot...
  • [17:51:24] <jkridner> docelic: the USB host port is not populated on Rev. Bx as it was a buggy layout.
  • [17:51:26] <dcordes> cian: should it report anything on serial?
  • [17:51:40] <jkridner> so, the only USB option is the OTG.
  • [17:52:09] <khasi1> cian: if you are playing with standard mini B cable then make sure that pin 5 is connected, generally I observed though there were 5 wires the GND wire was not connected to the pin at all
  • [17:52:26] <docelic> jkridner, so in practical terms, on rev. Bx, how do I connect anything usb (primarily, keyboard) to the board? can I find instructions about it on the web?
  • [17:52:37] <khasi1> cian: meaning the 5th pin might be floating
  • [17:52:50] <cian> khasi1: spec seems to allow pin 5 to float on mini B
  • [17:52:56] <cian> my cable had only 4 wires
  • [17:52:59] <jkridner> http://beagleboard.org/support/faq (see item 2)
  • [17:53:28] <cian> so i shorted 4&5 in the plug itself and then glued it back together, I'll order a Mini A in the morning
  • [17:53:29] <jkridner> the Angstrom and diagnostic kernels provide support for USB OTG.
  • [17:53:46] <cian> dcordes: yes, you should see X-Loader and U-Boot output
  • [17:53:52] <cian> they will be running off the card
  • [17:53:57] <khasi1> cian: ok, great
  • [17:54:03] <jkridner> I do not recommend the hardware mods currently being discussed--perform them at your own risk.
  • [17:54:18] <jkridner> if you know what you are doing, then you should know if you can ignore me.
  • [17:54:22] <dcramer> jkridner: I actually realized I didn't respond to that message about bridging
  • [17:54:30] <dcramer> I can't actually see how to respond to it ?
  • [17:54:41] <cian> jkridner: to clarify I'm not modding the beagle, just the mini B plug
  • [17:54:55] <jkridner> cian: understood.
  • [17:55:51] <jkridner> dcramer: do you not have any additional info on your setup and the individual steps? I think you are likely right about the DHCP service requirement.
  • [17:56:00] <cian> incidentally, I found USB data transfer cable with linux support
  • [17:56:04] <docelic> so I'm out of luck.. no A->miniA adapters on ebay
  • [17:56:14] <cian> it shows up as usb1: register 'cdc_subset' at usb-musb_hdrc.0-1.3, ALi M5632, 6a:....
  • [17:56:27] <cian> and allows for networking over USB between two linux machines running in host mode
  • [17:56:29] <dcramer> jkridner: I"m running linux inside parallels was going to try it without in a sec
  • [17:56:38] <jkridner> are these out or not available in your region (http://www.vernier.com/accessories/access.html?usb-mini&template=basic.html)?
  • [17:56:55] <cian> dcordes: you might have a problem with your SD card formatting
  • [17:57:08] <cian> the OMAP is pretty fussy about booting of SD cards
  • [17:57:36] <jkridner> http://www.hama.de/portal/searchSelectedProduct*NO/articleId*93590/action*2563/searchMode*1/bySearch*on+the+go is an unconfirmed option.
  • [17:58:49] <cian> jkridner: what about http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=WM17494-ND
  • [17:59:01] <cian> if the hub has a mini B socket
  • [17:59:22] <jkridner> http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=WM17494-ND
  • [17:59:45] <jkridner> sorry, didn't read before hitting paste.
  • [17:59:50] <cian> lol
  • [17:59:58] <jkridner> for the mini-B socket, the above is an option.
  • [18:00:31] <jkridner> so, that would work fine with a powered hub with a mini B port.
  • [18:02:24] <sakoman> mru: thanks for the header location fix!
  • [18:06:11] <cian> dcordes: did you get it working again?
  • [18:08:23] <dcordes> cian: I have a strange problem, somehow I cannot write MLO to the fat partition. it is saved as mlo
  • [18:08:41] <sakoman> koen: gnome-games is failing for me: http://www.sakoman.net:8000/public/logs/254245.txt
  • [18:08:43] <cian> dcordes: did you format on windows or linux?
  • [18:08:44] <Crofton> that should be ok
  • [18:08:54] <Crofton> fat does not pay attention to case ...
  • [18:09:12] <ds2> for people in the US, a few of the big box stores sell the right adapters as a "universal" cable kit
  • [18:09:14] <dcordes> cian: on linux using parted and mkfs.vfat -F 32
  • [18:09:56] <dcordes> same is with normal FAT size
  • [18:09:57] <cian> ok, that caused trouble for me, I found the OMAP prefers FAT16 formatted on windows, not sure why... YMMV
  • [18:10:57] <cian> my linux formatted FAT partition also caused U-Boot to occasionally say "Invalid FAT entry" when doing fatls mmc 0
  • [18:11:09] <mru> I only got the rom loader working with one mmc card
  • [18:11:41] <khasi1> It should work with all cards, only the formatting tool matters.
  • [18:11:45] <mru> of the three I tried
  • [18:11:53] <mru> I tried to format them the same way
  • [18:12:00] <mru> must have done something differently
  • [18:12:02] <khasi1> for some reason the geometry was not getting populated in a way that ROM Code understands
  • [18:12:05] <ds2> mru: what sizes were the card?
  • [18:12:14] <mru> 64M, 128M, 1G
  • [18:12:18] <mru> the 64M worked
  • [18:12:27] <khasi1> I have tried 32M as well
  • [18:12:31] <cian> mru: I have found that small FAT (not 32) partitons work best
  • [18:12:32] <ds2> hmm I'd think the 1G would be the working one
  • [18:12:38] <dcordes> is the flasher case sensitive?
  • [18:12:49] <ds2> I couldn't get my small card to work
  • [18:12:50] <khasi1> mru: just try with HP Disk formatting utlity
  • [18:13:01] <mru> no windows here
  • [18:13:03] <khasi1> ds2: how small was it
  • [18:13:20] <ds2> let take a look
  • [18:13:33] <mru> iirc it wouldn't work if the partition was too small either
  • [18:13:58] <cian> mru: that may your problem, much as it pains me to say it, windows seems to be the best tool for the job when you're trying to make FAT partitions for the OMAP
  • [18:14:18] * t_s_o (n=tso@173.84-49-129.nextgentel.com) has joined #beagle
  • [18:14:38] <ds2> 32M Lexar SD
  • [18:14:52] <mru> fdisk seemed to do a good enough job
  • [18:14:54] <khasi1> ds2: it should work as well
  • [18:14:56] <ds2> but I was using the Linux instruction using fdisk to rebuild the geometry
  • [18:15:12] <ds2> a 2G card with the same instruction worked prefectly
  • [18:15:32] <khasi1> even I dont recommend Windows for just formatting the card.
  • [18:15:50] <ds2> just to be clear, by work I mean having MLO recogonized; it mostly works using the fatread/fatls commands
  • [18:16:06] <mru> u-boot work with just about anything
  • [18:16:12] <mru> it's the rom loader that's picky
  • [18:16:32] <dcordes> I'm trying with a 90M partition now
  • [18:16:43] <khasi1> actually the ROM code has limited space to use, so the FAT analyzing is based on some logic that is failing for some kind of partitions
  • [18:16:43] <ds2> these 32M cards are convenient since they are pretty much a giveway
  • [18:17:04] * docelic (n=docelic@78.134.193.52) Quit ("http://www.spinlocksolutions.com/")
  • [18:17:04] <cian> khasi1: does it do "real" fat or some kind of cheat?
  • [18:17:13] * Olipro (n=Olipro@unaffiliated/olipro) Quit (Read error: 60 (Operation timed out))
  • [18:17:16] <ds2> any chance the ROM code may be confused if it sees the "wrong" kind of FAT (FAT12 vs FAT16 vs FAT32)?
  • [18:17:36] <khasi1> cian: it is not cheat for sure, but it is not having buffering , etc
  • [18:17:36] <mru> empirical evidence show that it only works with fat32
  • [18:17:48] <cian> mru: FAT16 definitely works sometimes
  • [18:17:53] <khasi1> it works with FAT32 only
  • [18:17:55] <mru> fat12 didn't work
  • [18:18:21] <khasi1> fat16 support is not their in u-boot and x-loader so preferable to go for fat32
  • [18:18:34] <ds2> Ah... formating that to fat32 may have failed
  • [18:18:45] <ds2> bbl... used monitor with DVI inputs is here!
  • [18:18:57] <mru> linux mkdosfs refuses to make a fat32 on small partitions
  • [18:18:59] <dcordes> I'm trying in windows
  • [18:19:12] <cian> Hmm, it might make sense to make a raw disk image with a known working FAT32 partition, MLO etc
  • [18:19:15] <cian> for people trying to unbrick
  • [18:19:25] <cian> just so they can remove one more unknown from the equation
  • [18:19:32] <khasi1> are there any kind of tools available in linux apart from the standard ones
  • [18:19:53] <khasi1> may be modify fdisk to tune the card ??
  • [18:20:27] <ds2> anyone know if mtools will do fat32 thesedays?
  • [18:20:42] <cian> khasi1: could it be something as simple as fragmented files don't work?
  • [18:20:46] * dcordes (n=dcordes@unaffiliated/dcordes) Quit (Remote closed the connection)
  • [18:21:23] <khasi1> I had a tool, that I used to analyze when exactly cards didnt work.
  • [18:21:48] <dirk2> cian: Yes, I think to remember somebody mentioned that files haven't to be fragmented
  • [18:22:19] <khasi1> some times I found that FAT does a sector read with a specific number and for some reason the data retrieved didnt contain any info
  • [18:22:20] <cian> dirk2: cool thanks (and kudos for interesting use of haven't)
  • [18:22:46] <cian> khasi1: out of interest is MLO loaded in SD or SPI/MMC mode?
  • [18:22:54] <khasi1> with the other windows tool, the fat table was created always in a way that gave the correct table required for ROM to look into
  • [18:22:59] <khasi1> same case with u-boot and x-loader
  • [18:23:09] <dirk2> ;)
  • [18:23:47] <khasi1> cian: there is no SPI/MMC mode..
  • [18:23:56] <khasi1> cian: it is in SD mode only
  • [18:24:22] <khasi1> and there is no issue in MMC or FAT FS as such, its only the way partition table or boot record is read.
  • [18:24:39] <cian> ok
  • [18:25:17] <khasi1> all these are pending tasks to work on :(
  • [18:25:35] <cian> koen: opkg install libglib-2.0-0 is seg faulting here, any ideas?
  • [18:25:57] <cian> khasi1: and it's not exactly the most fun pending task to work on either...
  • [18:26:48] * mru avoids glib as much as possible
  • [18:27:08] <khasi1> what does mru use ?
  • [18:27:11] * Olipro (n=Olipro@unaffiliated/olipro) has joined #beagle
  • [18:27:28] <mru> what would you use glib for?
  • [18:27:41] <mru> I'm not talking about glibc (note the c)
  • [18:27:42] <cian> i'm trying to install midori
  • [18:28:00] <mru> for the record, I'd also like to avoid glibc
  • [18:28:04] <khasi1> oh... miss reading...
  • [18:29:29] <koen> sakoman: try rebuilding python, it stages headers during do_compile, which packaged-staging doesn't know about
  • [18:33:38] <dcramer> jkridner: ok, tried on a dedicated machine, still no luck
  • [18:37:03] * Beagle5 (n=Beagle5@dyn.83-228-182-215.dsl.vtx.ch) has joined #beagle
  • [18:37:12] * Beagle5 is now known as sprotch
  • [18:37:12] * khasi1 (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [18:41:42] * feig1 (n=ejf3@57.sub-75-195-45.myvzw.com) has joined #beagle
  • [18:45:19] <jkridner> dcramer: the bridging?
  • [18:45:27] <jkridner> dedicated Ubuntu machine?
  • [18:45:29] <dcramer> yeah
  • [18:45:33] <dcramer> dedidicated ubuntu
  • [18:45:40] <dcramer> course just found this tidbit
  • [18:45:49] <dcramer> bridges don't work well with wifi cards
  • [18:47:57] * asterick (i=lemoncar@svn.rkiq.net) Quit ()
  • [18:48:26] <jkridner> oh. I wasn't expecting that.
  • [18:49:06] <dcramer> couple more things
  • [18:49:19] <dcramer> http://www.linuxfoundation.org/en/Net:Bridge
  • [18:49:20] <jkridner> when I've done it, I haven't used bridging, only added a route.
  • [18:49:24] <dcramer> very good reading at the end
  • [18:49:31] <dcramer> ie set delay
  • [18:49:33] <jkridner> I think I have some links on my delicious account as to how I did it.
  • [18:49:45] <jkridner> I just use Ethernet recently.
  • [18:49:52] * feig1 (n=ejf3@57.sub-75-195-45.myvzw.com) Quit ("Konversation terminated!")
  • [18:49:56] <dcramer> also the iptables
  • [18:50:07] <dcramer> I just got it to get an ip address through the bridge
  • [18:50:34] * feig1 (n=ejf3@57.sub-75-195-45.myvzw.com) has joined #beagle
  • [18:51:10] <Crofton|work> bits to connect fpga board to Beagle have shipped
  • [18:52:36] <dcramer> jkridner: ok, working now
  • [18:52:46] <dcramer> course I don't know which bit was the important bit :(
  • [18:53:37] * chakie wants a board too! whaaa!
  • [18:54:38] * dirk2 (n=dirk@F33af.f.strato-dslnet.de) has left #beagle
  • [18:54:58] <jkridner> that's not good. going to power cycle and try again?
  • [18:55:14] <davr> Crofton|work: sounds interesting, what fpga board is this?
  • [18:56:06] <Crofton> diglent spartan 3 starter board
  • [18:56:07] <dcramer> jkridner: yeah
  • [18:56:54] <dcramer> jkridner: one thing though, it didn't set the udhcpc didn't set the ip due to a missing script ?
  • [18:57:48] * guillaum1 (n=gl@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  • [18:57:52] <sakoman> koen: thanks for the python hint, will try it now
  • [18:58:06] * guillaum1 (n=Guillaum@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) has joined #beagle
  • [18:59:22] * johnx (n=john@p2172-ipbf2302hodogaya.kanagawa.ocn.ne.jp) Quit (Read error: 113 (No route to host))
  • [18:59:28] * johnx (n=john@p2172-ipbf2302hodogaya.kanagawa.ocn.ne.jp) has joined #beagle
  • [19:00:11] * feig (n=ejf3@215.sub-75-222-127.myvzw.com) Quit (Read error: 110 (Connection timed out))
  • [19:00:58] <dcramer> I noticed someone has compiled jamvm for the beagle
  • [19:01:03] <dcramer> is there a download link ?
  • [19:02:39] <jkridner> you running OE?
  • [19:03:00] <jkridner> just do 'opkg install jamvm'... links should be within Angstrom setup.
  • [19:03:23] <jkridner> I've already made use of it running Jetty, Rhino, and Helma.
  • [19:03:30] <jkridner> I'm very pleased.
  • [19:04:41] <jkridner> root@beagleboard:~# file `which java`/usr/bin/java: symbolic link to `/usr/bin/jamvm'
  • [19:04:55] <jkridner> root@beagleboard:~# java -versionjava version "1.5.0"JamVM version 1.5.1Copyright (C) 2003-2008 Robert Lougher <rob@lougher.org.uk>
  • [19:05:19] <jkridner> Boot Library Path: /usr/lib/classpathBoot Class Path: /usr/share/jamvm/classes.zip:/usr/share/classpath/glibj.zi
  • [19:05:21] <dcramer> jkridner: OE ?
  • [19:05:26] <dcramer> sorry I"m very new to this
  • [19:05:27] <jkridner> Open Embedded.
  • [19:05:50] <dcramer> jkridner: no, just the diagnostic distro so far
  • [19:05:56] <jkridner> ah.
  • [19:05:56] <dcramer> is there an howto OE ?
  • [19:06:10] <jkridner> yes, starting from the elinux wiki.
  • [19:06:42] <dcramer> k, thanks
  • [19:06:55] <dcramer> I also use Jetty for most of my embedded stuff
  • [19:07:00] <jkridner> http://elinux.org/BeagleBoard#OpenEmbedded
  • [19:07:36] <dcramer> thanks
  • [19:07:53] <jkridner> http://www.angstrom-distribution.org/blog/koen
  • [19:08:11] <jkridner> scroll down to 2nd article (as of today) for links to the demo images.
  • [19:08:27] <keesj> jetty the webserver?
  • [19:08:40] <jkridner> once you have networking, you can do 'opkg install jamvm'
  • [19:08:44] <jkridner> keesj: yes.
  • [19:09:27] <dcramer> jkridner: ok, confirmed bridging does not like wifi
  • [19:09:31] <dcramer> at least my wifi
  • [19:12:44] <jkridner> keesj: beagleboard.org is running helma as the server, which makes use of jetty and rhino. I was able to run helma just fine on a beagle board. :)
  • [19:13:43] <jkridner> it is part of my contextual web stuff.
  • [19:15:21] <jkridner> google gears and greasemonkey and I'll be just about there.
  • [19:18:46] * likewise (n=chatzill@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [19:19:02] <likewise> good day all
  • [19:19:11] <koen> hey likewise
  • [19:19:21] <likewise> hey koen
  • [19:19:56] * JuanG_ (n=Juan@nat/ti/x-8a30401f3bfd4575) has joined #beagle
  • [19:21:08] <jkridner> given the OE braintrust in the room, I have a dummy question that I hope you can squash quickly.
  • [19:21:20] * JuanG_ (n=Juan@nat/ti/x-8a30401f3bfd4575) has left #beagle
  • [19:21:44] <jkridner> http://tinderbox.openembedded.net/public/logs/775614.txt <-- boost is currently built, but is not showing up in the asio include path.
  • [19:22:01] <jkridner> where in the .bb file is the include path setup?
  • [19:22:06] <jkridner> the .bb file is almost bare.
  • [19:22:29] <likewise> brain hurts already :-) but I will look
  • [19:22:47] <koen> jkridner: include paths shouldn't be territory of OE :)
  • [19:22:59] <jkridner> sorry, this one isn't should be a brain teaser.
  • [19:23:07] <koen> jkridner: sane apps use things pkgconfig for stuff like that
  • [19:23:28] <sakoman> koen: tried rebuilding python and then beagleboard-demo-image
  • [19:23:29] <jkridner> this app uses autotools.
  • [19:23:33] <likewise> configure is called with --libdir=/usr/lib --includedir=/usr/include. Is this intended to be cross-compiled?
  • [19:23:45] <sakoman> koen: still get the same failure: http://www.sakoman.net:8000/public/logs/255236.txt
  • [19:23:50] <jkridner> that is what I thought was odd.
  • [19:24:13] <jkridner> .bb does 'inherit autotools pkgconfig'.
  • [19:24:52] <sakoman> koen: ffmpeg is also failing: http://www.sakoman.net:8000/public/logs/255405.txt
  • [19:24:52] <koen> likewise: libdir and includir dir are for runtime
  • [19:25:20] <koen> likewise: so you app doesn't put its libs in /OE/likewise/asting/lib on your device
  • [19:25:42] <sakoman> koen: my oe.dev is about 13 hours old
  • [19:25:42] <koen> sakoman: I get the ffmpeg error too on one machine :(
  • [19:26:13] <likewise> koen: thanks. Anyone else forget what I said.
  • [19:26:44] <jkridner> oddly, ffmpeg build fine for me at http://tinderbox.openembedded.net/packages/102283/
  • [19:27:31] <jkridner> likewise, koen: understood /usr/lib is fine, since it is meant for the runtime, but I'm not finding where the compile time include paths are given.
  • [19:27:52] <koen> I'm doing an asio build right now to see what's wrong
  • [19:28:15] * koen has fixed far too many boost dependant apps configure scripts
  • [19:30:57] <dcramer> jkridner: is there a link to prebuilt image for oe ?
  • [19:32:20] <jkridner> dcramer: yes, http://www.angstrom-distribution.org/beagleboard-demo-image-available
  • [19:33:06] <dcramer> thanks
  • [19:33:19] <jkridner> note the bootargs in the README.txt at http://www.angstrom-distribution.org/demo/beagleboard/README.txt
  • [19:33:28] <dcramer> yes, reading now
  • [19:33:40] <dcramer> although I don't need a gui
  • [19:33:43] <dcramer> just command line
  • [19:35:54] * cian (n=cian@cian.ws) Quit ()
  • [19:36:35] * valhalla (n=valhalla@81-174-21-16.dynamic.ngi.it) Quit ("Leaving")
  • [19:38:59] * max_fr (i=max@212-198-79-163.rev.numericable.fr) has joined #beagle
  • [19:39:34] <jkridner> there are some older console images built at http://amethyst.openembedded.net/~koen/index.php?path=beagleboard/
  • [19:39:48] <jkridner> of course, once you get setup with OE, you can build the console-image yourself.
  • [19:43:49] * dcordes (n=dcordes@unaffiliated/dcordes) has joined #beagle
  • [19:45:52] <dcordes> hmm the serial console only prints 40T when I powr up with x-loader and uboot sd in
  • [19:45:59] <dcordes> is that a bad sign?
  • [19:46:33] <vlad_> that I think means that uboot is dead and you're getting the fallback serial loader
  • [19:47:16] <koen> it means the rom code didn't load x-load from sd
  • [19:47:23] <Crofton> that's a good sign
  • [19:47:29] <dcordes> so it doesn't like my SD
  • [19:47:36] <dcordes> I formated in windows xp this time
  • [19:47:37] <Crofton> you have a working serial port, cable and software :)
  • [19:47:42] <koen> vlad_: could you forward your mozconfig mail to k.kooi@student.utwente.nl?
  • [19:48:09] <dcordes> Crofton: good so I can chill now. I thought I bricked it
  • [19:48:15] <vlad_> koen: yep, lemme do so now so I don't forget again :)
  • [19:48:16] <max_fr> dcordes : hi, have you tried http://elinux.org/BeagleBoardRecovery
  • [19:48:21] <Crofton> 40T is a good sign
  • [19:48:30] <mru> dcordes: it's hard to brick the board beyond recovery
  • [19:48:38] <Crofton> that shortens the list of things you need to figure out
  • [19:48:53] <koen> vlad_: you wouldn't be vlad Vuki??evi?? by chance?
  • [19:48:58] <Crofton> you can get bad code in flash that leads to no 40T, unless you press the buttons properly
  • [19:49:06] <Crofton> I was there once, and that was scary
  • [19:50:08] <vlad_> koen: I am
  • [19:51:03] <vlad_> koen: email sent, more details this week
  • [19:51:07] <koen> thanks
  • [19:51:25] <koen> got it
  • [19:55:22] * like2wise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [19:56:30] <jkridner> any easy way for me to figure what depends on asio, so that I can finish building everything that doesn't?
  • [19:57:00] <jkridner> also, it looks like I'm building a bunch of packages that conflict and I need to state the preferred provider...
  • [19:57:14] <jkridner> does anyone know the advantages of Cacao vs. JamVM?
  • [19:59:40] <jkridner> seems like JamVM, since it works for me, might be the way to go due to size.
  • [20:01:03] * max_fr (i=max@212-198-79-163.rev.numericable.fr) has left #beagle
  • [20:10:00] <ds2> JamVM is smaller
  • [20:10:27] <ds2> but the problem seems to be classpath, not the VM
  • [20:10:27] <dcordes> mru: where can I find those parted instructions you refered to earlier?
  • [20:12:05] <mru> dcordes: http://code.google.com/p/beagleboard/wiki/LinuxBootDiskFormat
  • [20:12:38] <mru> http://wiki.davincidsp.com/index.php?title=MMC_Boot_Format
  • [20:12:58] <dcordes> great thanks
  • [20:13:31] * Beagle3 (n=Beagle3@63.65.238.105) Quit ()
  • [20:14:29] * sprotch (n=Beagle5@dyn.83-228-182-215.dsl.vtx.ch) Quit ()
  • [20:14:52] <dcordes> mru: they use a 128M card in the ti wiki
  • [20:15:03] <dcordes> mru: I only have 1, 2 and 8 gb
  • [20:15:05] <jkridner> hmmm.... I did a 'bitbake boost', just to try building that one package to see what the dependencies were, and now bitbake started back at the native tools!
  • [20:15:25] <mru> boost evil in condensed form
  • [20:17:20] <dcordes> jkridner: bitbake --dry-run --verbose ?
  • [20:18:42] <dcordes> did somebody succesfully flash with 1gb or more?
  • [20:19:19] <ds2> flash with? you mean try a card with > 1G?
  • [20:25:11] <keesj> prpplague: on my setup pulling emu0 and 1 down while booting simply makes the board not boot
  • [20:26:54] <prpplague> keesj: if jtag is connected, that is what i would expect
  • [20:27:04] <dcordes> ds2: yes >=
  • [20:27:25] <ds2> dcordes: I have it running on a 2G SD card
  • [20:28:00] <dcordes> mru: I did exactly what they say in http://wiki.davincidsp.com/index.php?title=MMC_Boot_Format just using a 2gb card. I put the card in, hold down the user button, power on, and only see 40T .
  • [20:28:35] <ds2> how are you making the card?
  • [20:29:17] <dcordes> ds2: like this http://wiki.davincidsp.com/index.php?title=MMC_Boot_Format
  • [20:29:50] <dcordes> and then I put MLO u-boot.bin uImage x-load.bin.ift from http://www.angstrom-distribution.org/demo/beagleboard/ on it
  • [20:31:13] <ds2> dcordes: yes but I did a split of 128M for VFAT and the rest for ext3
  • [20:31:24] * bazbell (n=a0192809@nat/ti/x-aef9f8042a125da5) Quit ("Leaving.")
  • [20:31:44] <ds2> x-load.bin isn't used in for MMC boot, only MLO, u-boot.bin, and uImage are needed
  • [20:32:24] <dcordes> but I want to flash u-boot from the card
  • [20:32:27] <dcordes> because I bricked it
  • [20:32:41] <dcordes> I destroyed the nand uboot
  • [20:33:15] * t_s_o (n=tso@173.84-49-129.nextgentel.com) Quit (Connection timed out)
  • [20:37:03] * nooomem (n=ravisy@ip68-100-201-32.dc.dc.cox.net) has joined #beagle
  • [20:44:42] <ds2> ah okay
  • [20:45:08] <ds2> sigh... this monitor will not display the beagle logo screen :/ so much for "COTS" being simplier/more standardized
  • [20:47:33] * cian (n=cianh@cian.ws) has joined #beagle
  • [20:47:51] <keesj> prpplague: no , jtag is not connected and but even then why should it stop?
  • [20:48:52] <prpplague> keesj: bacause it usually holds the system in debug mode, or atleast that is how i understand it
  • [20:51:48] * paul_pwsan (n=paul@utopia.booyaka.com) Quit (Remote closed the connection)
  • [20:51:53] * [-ip-] (n=ts@mnhm-590e08cb.pool.einsundeins.de) has left #beagle
  • [20:52:05] * paul_pwsan (n=paul@utopia.booyaka.com) has joined #beagle
  • [20:55:18] * t_s_o (n=tso@173.84-49-129.nextgentel.com) has joined #beagle
  • [20:56:02] <dcramer> jkridner: ok, I got the demo angstrom working, but no usb networking :(
  • [20:56:29] <jkridner> still an issue with the host? I wouldn't expect that to change
  • [20:58:00] <dcramer> no, now I don't even see packets going to the usb0 on the host
  • [20:58:03] <dcramer> nothing goes over
  • [20:58:11] <dcramer> wondering if it works on that version ?
  • [21:00:44] * feig1 (n=ejf3@57.sub-75-195-45.myvzw.com) Quit ("Konversation terminated!")
  • [21:02:02] * dcordes (n=dcordes@unaffiliated/dcordes) Quit (Remote closed the connection)
  • [21:10:01] <dcramer> interesting not a single packet goes over
  • [21:10:11] <dcramer> but the gadget is registered on the host computer
  • [21:17:27] <mru> hmm... usb doesn't work with recent linux-omap git
  • [21:18:45] * BThompson (n=BThompso@nat/ti/x-3e680c4477a965bb) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [21:19:55] <dcramer> it doesn't ?
  • [21:20:26] * dcordes (n=dcordes@unaffiliated/dcordes) has joined #beagle
  • [21:20:43] <dcordes> mru: with a 128M partition, it worked.
  • [21:20:54] <mru> good
  • [21:21:01] <mru> I guess I didn't try hard enough
  • [21:22:51] <sakoman> mru: what is the symptom on usb?
  • [21:23:03] <mru> it's as if it's not there
  • [21:23:49] <sakoman> hmm . . . as of a couple of days ago it worked for me on Overo -- both musb and ehci
  • [21:24:13] <mru> maybe it's something I've patched
  • [21:24:18] <dcramer> jkridner: ok rebooted host now it works
  • [21:24:28] <sakoman> building a kernel with this morning's checkins, I'll let you know if that introduces any problems
  • [21:24:56] <dcordes> cian: I just flashed u-boot.bin and x-loader from the demo folder as described at http://code.google.com/p/beagleboard/wiki/BeagleBoardDiagnostics How to flash x-loader and u-boot from u-boot. now when I power on, I get Texas Instruments X-Loader 1.41sms, but nothing else
  • [21:25:34] <cian> ok, you're half way there
  • [21:25:47] <cian> you have flashed the wrong x-loader
  • [21:26:07] <dcordes> I only had that one on the card
  • [21:26:25] <cian> let me find a link...
  • [21:26:30] <sakoman> mru: omapfbplay is still failing for me: http://www.sakoman.net:8000/public/logs/255663.txt
  • [21:26:34] <mru> sakoman: I have this patch in my tree: http://git.mansr.com/?p=linux-omap;a=commit;h=2e6aa4efb0e14c51ff0427927b1b38136911fa93
  • [21:27:12] <cian> dcordes:
  • [21:27:16] <cian> dcordes: try http://beagleboard.googlecode.com/files/x-load.bin.ift_for_NAND
  • [21:27:24] <sakoman> mru: I'm not using that one
  • [21:27:33] * MPankrat (n=chatzill@wsip-24-249-118-14.ks.ks.cox.net) has joined #beagle
  • [21:28:27] <dcordes> cian: ah so that xloader from the demo is for sd?
  • [21:28:52] <cian> dcordes: no, the MLO file is the x-loader image for the SD
  • [21:29:04] <cian> x-load.bin.ift from the demo should work
  • [21:29:11] <cian> I'm not sure why it doesn't
  • [21:29:30] <cian> I was originally only suggesting flashing U-Boot
  • [21:29:34] <cian> not X-Loader
  • [21:29:59] <dcordes> I should stop blindly copy pasting instructions
  • [21:30:25] <cian> that can cause trouble alright
  • [21:30:37] <mru> sakoman: your -I flags are wrong for omapfbplay
  • [21:31:22] <dcramer> jkridner: coolio we now have jamfm
  • [21:31:24] <dcramer> jamvm
  • [21:31:33] <cian> meh
  • [21:31:50] <cian> ^^^sorry, wrong window^^^
  • [21:32:14] <sakoman> mru: hmm . . . you are right. I wonder if something went wrong with my merge with koen's recipe updates
  • [21:32:19] <dcordes> moo
  • [21:34:23] * RogerMonk (n=a0740758@nat/ti/x-0f2a6cad7b3fdc73) has joined #beagle
  • [21:36:12] <cian> dcordes: did that x-loader work?
  • [21:37:39] <dcordes> cian: one moment
  • [21:38:24] <dcordes> cian: which u-boot.bin and MLO should I take to flash?
  • [21:40:57] <cian> dcordes: you don't flash MLO it just boots the OMAP from the SD card
  • [21:41:21] <cian> dcordes: use the u-boot.bin from www.angstrom-distribution.org/demo/beagleboard
  • [21:41:40] <dcordes> so I only put u-boot.bin and x-loader-nand?
  • [21:44:35] <cian> dcordes: you must put MLO on the SD card
  • [21:44:41] * paul_pwsan (n=paul@utopia.booyaka.com) has left #beagle
  • [21:44:47] <sakoman> mru: usb seems to function for me with latest linux-omap git: http://pastebin.com/m12af3458
  • [21:45:18] <cian> chances are you have flashed the u-boot.bin correctly if you followed the instructions on http://code.google.com/p/beagleboard/wiki/BeagleNANDFlashing
  • [21:45:50] <cian> you must just flash x-loader now, because you have got a broken version in the nand
  • [21:46:22] * robclark (n=robclark@nat/ti/x-5be14ec66906c341) Quit ()
  • [21:50:27] * DDevine (n=devine@123-243-187-85.static.tpgi.com.au) Quit (Read error: 104 (Connection reset by peer))
  • [21:50:57] <mru> sakoman: oh well, I'll try again later
  • [21:51:08] <dcramer> ok, I've looked is there an ftp daemon for openembedded ?
  • [21:53:47] <dcordes> cian: ok I only flashed x-load and got U-Boot 1.3.4-rc2 (Aug 2 2008 - 15:30:02)
  • [21:53:53] <dcordes> now booting from nand again
  • [21:54:02] <cian> dcordes: fantastic
  • [21:54:12] <cian> let me know if audio works for you now
  • [21:54:14] <dcordes> is that the good audio uboot?
  • [21:54:19] <dcordes> ;)
  • [21:57:00] <dcordes> I will make a new kernel with the 720p framebuffer resolution
  • [21:59:57] <dcordes> does somebody know the kconfig for otg host mode force by chance?
  • [22:00:41] <dcordes> I find the need of having a host cable connected while booting annoying
  • [22:05:02] * nooomem (n=ravisy@ip68-100-201-32.dc.dc.cox.net) Quit ()
  • [22:11:49] * DDevine (n=devine@123-243-187-85.static.tpgi.com.au) has joined #beagle
  • [22:12:23] * t_s_o (n=tso@173.84-49-129.nextgentel.com) Quit ("Konversation terminated!")
  • [22:13:03] * beagle-dave (n=tux@adsl-69-226-208-117.dsl.pltn13.pacbell.net) has left #beagle
  • [22:13:05] * Beagle9 (n=Beagle9@128-193-136-124.public.oregonstate.edu) has joined #beagle
  • [22:13:46] * Beagle9 (n=Beagle9@128-193-136-124.public.oregonstate.edu) Quit (Client Quit)
  • [22:17:23] * cbrake is now known as cbrake_away
  • [22:26:04] <turqy> do the beagles from digikey come iwth U-boot preloaded?
  • [22:26:18] <ds2> think so
  • [22:27:51] <NishanthMenon> turqy: check http://code.google.com/p/beagleboard/wiki/outofthebox
  • [22:30:21] <mru> sakoman: http://source.mvista.com/git/gitweb.cgi?p=linux-omap-2.6.git;a=commit;h=ffd60d49c70b31d26430a1098edf0eef5e4dfac8
  • [22:30:25] <mru> that broke it
  • [22:31:28] <ds2> missing code?
  • [22:31:43] <ds2> that commit seem to lack stuff for anything but the n800
  • [22:32:36] <mru> @nokia.com, go figure
  • [22:33:22] <dcordes> to replace the kernel image, I only make uImage and replace the old one with it, right?
  • [22:34:17] <ds2> yep
  • [22:34:31] <ds2> or build a naked and directly jump into it ;)
  • [22:34:59] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
  • [22:37:43] <dcordes> ds2: build a naked?
  • [22:39:12] <dcordes> make uImage doesn't build because it lacks mkimage
  • [22:48:09] <sakoman> mru: interesting! I wonder why it still works on overo
  • [22:50:15] * MPankrat (n=chatzill@wsip-24-249-118-14.ks.ks.cox.net) has left #beagle
  • [22:51:42] * JJS (n=josh@CPE-70-92-19-222.wi.res.rr.com) has joined #beagle
  • [22:53:05] <dcordes> cian: it seems to work
  • [22:53:52] <dcordes> cian: yes, I head the big buckbunnysound
  • [22:55:24] <dcordes> cian: after a few seconds it spit an error and now sound won't work anymore. but it doesn't panic
  • [22:56:00] <mru> that's a known bug
  • [22:56:12] <mru> in the alsa drivers and libs
  • [22:56:22] <dcordes> mru: do you know why uboot affects the panic-on-sound problem?
  • [22:56:24] <mru> you'll have to live without sound for now
  • [22:56:28] * nooomem (n=chatzill@ip68-100-201-32.dc.dc.cox.net) has joined #Beagle
  • [22:56:40] <mru> no, what effect does it have?
  • [22:57:09] <nooomem> hey all
  • [22:57:18] <mru> hi nooomem
  • [22:57:43] <dcordes> with the stock shipped bootloader, I got a kernel panic in koen's image/kernel when playing sound. after flashing the bootloader provided by koen, it doesn't panic any longer
  • [22:58:28] <dcordes> mru: cian reports the same
  • [22:58:35] <nooomem> mru: you do ffmpeg stuff, right?
  • [22:58:41] <mru> why don't you ask koen?
  • [22:58:50] <mru> nooomem: yes, I do ffmpeg stuff
  • [22:59:37] <dcordes> koen: ping?
  • [22:59:40] <nooomem> mru: I want to start with beagleboard by building a simple media player which reads from the memory card.
  • [23:00:20] <nooomem> mru: it could just be - pick random media file on memory card - play it through sound out
  • [23:00:49] <nooomem> I assume some people in here have tried sqlite on this little guy?
  • [23:00:50] <dcordes> nooomem: in linux?
  • [23:01:35] <nooomem> dcordes: no, I want it to be my own program with the help of C libs
  • [23:02:00] <nooomem> dcordes: just a starter project
  • [23:02:17] <dcordes> you could do it with a bash script ran by a udev event
  • [23:02:35] * guillaum1 (n=Guillaum@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  • [23:02:45] <dcordes> that finds all media files and does playback them with mplayer -shuffle
  • [23:02:51] * guillaum1 (n=gl@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) has joined #beagle
  • [23:02:54] * like2wise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  • [23:03:08] <dcordes> nooomem: that's what I did with the zaurus once
  • [23:04:31] * JoeBorn (n=jborn@75.2.240.113) Quit (Read error: 110 (Connection timed out))
  • [23:07:46] <DDevine> nooomem i am going to do a similar thing.
  • [23:07:47] <nooomem> dcordes: cool - what did you use?
  • [23:08:02] <nooomem> DDevine: yeah, I think it's a good/common start project
  • [23:08:24] <DDevine> yeah, im also attaching an LCD to it
  • [23:08:26] <JJS> is anyone familiar with some manufacturers of sub-7" lcd's with dvi input?
  • [23:08:36] <JJS> all i can find are vga :/
  • [23:08:56] <DDevine> so my program has to output for the LCD (the LCD comes with buttons so I will have some functions :)
  • [23:09:32] <DDevine> http://www.crystalfontz.com/products/631/index.html - I want to mod one of these into a case (I will probably put it inside my computer while I work on it.)
  • [23:10:58] <DDevine> ok i really need to get to school.
  • [23:11:25] <nooomem> DDevine: later!
  • [23:12:28] <dcordes> nooomem: udev, bash, find, mplayer
  • [23:12:48] <DDevine> yep thats more or less it
  • [23:12:51] <dcordes> xmms in a different case where I played the mp3s with gui
  • [23:13:15] <dcordes> so the user could interact
  • [23:13:21] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) has joined #beagle
  • [23:13:31] <dcordes> on top of Xfbdev
  • [23:13:37] * turqy (n=turqy@adsl-75-9-35-139.dsl.rcsntx.sbcglobal.net) Quit (Read error: 104 (Connection reset by peer))
  • [23:13:42] * turqy (n=turqy@adsl-75-9-35-139.dsl.rcsntx.sbcglobal.net) has joined #beagle
  • [23:14:05] <nooomem> dcordes: how hard would this be to do without the help of linux?
  • [23:14:16] <dcordes> I don't know
  • [23:14:20] <nooomem> dcordes: hmm, k
  • [23:14:23] <DDevine> what else would you use :p
  • [23:14:27] <dcordes> but harder
  • [23:14:31] <DDevine> and what else is better?
  • [23:14:32] <dcordes> that's for sure
  • [23:14:52] <mru> nooomem: I see no reason to avoid linux here
  • [23:14:58] <mru> other than potential hack value
  • [23:15:24] <nooomem> mru: k, I just want to avoid the overhead
  • [23:15:51] <dcordes> DDevine: you will miss class!
  • [23:16:03] <DDevine> i know.
  • [23:16:30] <ds2> dcoredes: nake image
  • [23:17:09] <nooomem> mru: I guess I could just kick off this thing as a service and it would start blaring random music whenever the startup is done
  • [23:17:22] <dcordes> ds2: ?
  • [23:17:48] <DDevine> noomem you could do what I am going to do with one of the button LCDs from CrystalFontz
  • [23:18:40] <dcordes> DDevine: lol the pcbs of that beast are bigger than whole beagle
  • [23:18:42] <nooomem> DDevine: might just do that - I see they have a simple C sample
  • [23:18:59] <nooomem> dcordes: :)
  • [23:19:41] <DDevine> dcordes: I figure if i get the bracket version i can make it into a case
  • [23:19:45] <DDevine> which would be nice
  • [23:19:50] <DDevine> though yes, it is kinda big
  • [23:20:10] <DDevine> but i want to put a USB hub and batteries in there as well
  • [23:20:17] <DDevine> so it will end up just the right size.
  • [23:20:44] <dcordes> use a small lcd with touchscreen
  • [23:21:54] <DDevine> there is 2.5inch ones that you can get from lilliput
  • [23:22:20] <DDevine> they arent touch screen actually
  • [23:22:33] <JJS> but do they have a dvi input?
  • [23:23:35] <DDevine> nope
  • [23:24:27] <ds2> hmmm there used to be something called a naked image that don't have the uboot headers, seems that options has gone away
  • [23:24:35] * NishanthMenon (n=gnat@nat/ti/x-a7cb5db4763afe61) Quit ("Take my advice. I don't use it anyway")
  • [23:26:03] <DDevine> btw should I go with Debian Slackware or Angstrom?
  • [23:26:16] <DDevine> not slackware, Gentoo
  • [23:26:37] <Crofton|work> Angstrom is easiest :)
  • [23:27:04] <DDevine> thats what I thouht.
  • [23:27:26] * vlad_ is having good success with debian, but I think I'm an outlier
  • [23:27:37] <vlad_> and/or am not doing things The Right Way (tm)
  • [23:28:06] * Olipro_ (n=Olipro@unaffiliated/olipro) has joined #beagle
  • [23:29:10] <DDevine> I think ill try angstrom, then when it works ill back up my stuff and attempt gentoo or debian.
  • [23:32:30] <mru> DDevine: do you run gentoo on a pc?
  • [23:32:48] <vlad_> hrm. Undefined symbol __fortify_fail@GLIBC_PRIVATE ?
  • [23:37:54] <vlad_> oh I see, needs -lrt? that's a little weird, but ok
  • [23:39:25] <dcordes> JJS: maybe don't lookout for a tiny touchscreen, but for a tiny lcd and a touchscreen seperately
  • [23:44:56] * Olipro (n=Olipro@unaffiliated/olipro) Quit (Read error: 110 (Connection timed out))
  • [23:45:04] <dcordes> and can't you connect lcd controllers via i2c?
  • [23:45:11] <dcordes> beagle has i2c on the expansion slot no?
  • [23:45:50] <davr> i've seen small LCDs with serial input
  • [23:47:25] <davr> like this guy: http://www.sparkfun.com/commerce/product_info.php?products_id=8538
  • [23:51:39] <vlad_> huh, that's pretty cool
  • [23:51:55] <vlad_> esp. the onboard microsd