Join the chat at beagleboard.org/chat
IRC Log for 2008-07-21
Timestamps are in UTC.
- [00:00:56] * dannyBlue|DDd (n=chatzill@89-180-95-229.net.novis.pt) Quit ("ChatZilla 0.9.83 [Firefox 3.0.1/2008070208]")
- [00:13:22] <jkridner>
mru: how was the show?
- [00:14:28] * lgentili (i=bee0f6b0@gateway/web/ajax/mibbit.com/x-abf5cf028c4e8297) has joined #beagle
- [00:17:21] * lgentili (i=bee0f6b0@gateway/web/ajax/mibbit.com/x-abf5cf028c4e8297) Quit (Client Quit)
- [00:42:21] * mru (n=mru@thrashbarg.mansr.com) Quit ("leaving")
- [00:42:43] * mru (n=mru@thrashbarg.mansr.com) has joined #beagle
- [00:48:16] * oelewapperke (i=34d4599f@85-158-215-1.powered-by.benesol.be) Quit (kubrick.freenode.net irc.freenode.net)
- [00:48:16] * kulve (i=ce80e2f2@emperor.pingu.fi) Quit (kubrick.freenode.net irc.freenode.net)
- [00:53:39] * oelewapperke (i=a619fded@85-158-215-1.powered-by.benesol.be) has joined #beagle
- [00:54:03] * kulve (i=fc32a27c@emperor.pingu.fi) has joined #beagle
- [01:00:15] * MattJ (n=MattJ@88-111-173-221.dynamic.dsl.as9105.com) Quit (Read error: 110 (Connection timed out))
- [01:01:47] * Olipro (i=Olipro@unaffiliated/olipro) Quit (Connection timed out)
- [01:58:03] * jkridner|work (n=a0321898@nat/ti/x-e42a27a9d0753103) Quit (Remote closed the connection)
- [02:03:35] * jkridner_ (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
- [02:11:22] * jkridner (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) Quit (Read error: 110 (Connection timed out))
- [02:16:15] * NishanthM (n=Nishanth@cpe-24-27-74-89.tx.res.rr.com) Quit ("Aloha")
- [02:53:12] * jkridner_ (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) Quit ()
- [02:54:52] * NishanthM (n=Nishanth@cpe-24-27-74-89.tx.res.rr.com) has joined #beagle
- [02:55:45] * jkridner (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
- [02:56:39] * shankar (n=shankar@63.99.26.3) has joined #beagle
- [03:06:12] * shankar (n=shankar@63.99.26.3) Quit ("Leaving")
- [03:11:15] * shankar (n=shankar@63.99.26.3) has joined #beagle
- [03:11:42] * shankar (n=shankar@63.99.26.3) Quit (Client Quit)
- [03:36:59] * NishanthM (n=Nishanth@cpe-24-27-74-89.tx.res.rr.com) Quit ("Aloha")
- [03:48:52] * jkridner|work (n=a0321898@nat/ti/x-733cd08fe581acd3) has joined #beagle
- [04:20:14] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) Quit ("Trillian (http://www.ceruleanstudios.com")
- [04:33:33] * jkridner|work1 (n=a0321898@nat/ti/x-32b3972da559d196) has joined #beagle
- [04:37:29] * jkridner|work (n=a0321898@nat/ti/x-733cd08fe581acd3) Quit (Remote closed the connection)
- [04:39:03] * Thanatos (n=rwvens@a83-132-76-156.cpe.netcabo.pt) has joined #beagle
- [04:40:55] <Thanatos>
hello everyone
- [04:46:03] * Thanatos (n=rwvens@a83-132-76-156.cpe.netcabo.pt) has left #beagle
- [04:46:56] * Beagle2 (n=Beagle2@a83-132-76-156.cpe.netcabo.pt) has joined #beagle
- [04:48:01] <jkridner>
hello
- [04:49:00] <Beagle2>
alright, the guys over at #ioq3 need to know if OMAP3 can run OpenGL
- [04:49:51] <Beagle2>
or if sdl compiles on it
- [04:52:28] <calculus>
I think the drivers are more of an issue
- [04:57:14] <Beagle2>
can you tell me more?
- [04:57:40] <jkridner>
it can, but the drivers won't be out until later this year.
- [04:57:56] <Beagle2>
alright
- [04:58:07] <Beagle2>
it wont work immediately anyway
- [04:58:08] <jkridner>
OpenGL ES 1.1 and 2.0. You need a software translation layer to use OpenGL (without the ES).
- [04:58:21] <jkridner>
nope, not for a few months.
- [04:58:44] <jkridner>
those who want 3D graphics out-of-the-box should probably wait.
- [04:59:03] <Beagle2>
yeah, but I meant the entire system would take a while to get working
- [04:59:46] <Beagle2>
if ill be hiring programmers to get Ioq3 working on the beagle board they have to know what they are working with
- [05:01:10] <jkridner>
it is possible to get Beta drivers via more official channels, NDAs, etc.
- [05:01:22] <jkridner>
Beagle has a different support model than TI products.
- [05:05:36] * bazbell (n=a0192809@nat/ti/x-81323fa19cda767f) Quit ("Leaving.")
- [06:00:09] <ldesnogu>
Beagle2: a possible approach to get some early port of ioq3 would be by using OpenGL ES 2.0 imgtec "emulation" layer
- [06:00:35] <ldesnogu>
however I am not sure the ioq3 core team will like that :)
- [06:22:21] * Viral_Sachde (n=Viral_Sa@122.167.159.183) has joined #beagle
- [06:29:07] * jkridner|work (n=a0321898@nat/ti/x-8fe0d1d7d3406dfd) has joined #beagle
- [06:33:08] * jkridner|work1 (n=a0321898@nat/ti/x-32b3972da559d196) Quit (Remote closed the connection)
- [06:59:16] * trickie (n=trickie@basesoft.xs4all.nl) has joined #beagle
- [08:16:48] * [denix] (n=denys@nat/ti/x-292ca2ca61732e6c) has joined #beagle
- [08:31:36] * denix (n=denys@nat/ti/x-a7216ca9abac8a21) Quit (Remote closed the connection)
- [09:05:20] * ciskje (n=francesc@131.114.33.209) has joined #beagle
- [09:06:51] <ciskje>
Hi
- [09:11:33] * Viral_Sachde (n=Viral_Sa@122.167.159.183) Quit (Read error: 110 (Connection timed out))
- [09:11:52] * Olipro (i=Olipro@unaffiliated/olipro) has joined #beagle
- [09:14:12] * jkridner|work1 (n=a0321898@nat/ti/x-718527b3c8a6347a) has joined #beagle
- [09:17:21] <ciskje>
I'm trying to subscribe to the mailing list but I'm experiencing problems, cannot subscribe. Anyone experiencing the same problems?
- [09:21:29] * jkridner|work (n=a0321898@nat/ti/x-8fe0d1d7d3406dfd) Quit (Remote closed the connection)
- [10:33:35] * MattJ (n=MattJ@88-111-173-221.dynamic.dsl.as9105.com) has joined #beagle
- [10:41:16] * esden`away is now known as esden
- [11:08:04] * ciskje (n=francesc@131.114.33.209) Quit (Remote closed the connection)
- [11:08:27] * ciskje (n=francesc@131.114.33.209) has joined #beagle
- [11:08:46] * ciskje (n=francesc@131.114.33.209) Quit (Remote closed the connection)
- [11:30:06] * Viral_Sachde (n=Viral_Sa@122.167.154.2) has joined #beagle
- [11:31:03] <koen>
good morning all
- [11:31:08] * koen is at BHX atm
- [11:31:17] <MattJ>
Good morning koen :)
- [11:32:00] <Crofton|work>
gm
- [11:32:06] <Crofton|work>
how was the show?
- [11:32:12] <koen>
pretty nice
- [11:32:34] <koen>
the only anguish was coming from trying to cross-compile mythtv
- [11:33:39] <koen>
it still tries to run arm binaries on x86 if you specify --cross-compile
- [11:33:45] <koen>
*%(#*$%#( piece of shit
- [11:33:45] <Crofton|work>
urg
- [11:34:03] <Crofton|work>
I need to run make check on the gnuradio code
- [11:34:10] <Crofton|work>
which runs arm binaries
- [11:34:30] <Crofton|work>
fortunately I finally defeated the firewall on the box I do beagle buils on
- [11:34:35] <koen>
even worse, instead of doing gcc -v, it tries to run an arm binary to get the gcc version
- [11:34:47] <Crofton|work>
so I can nfs mount the build direcotry on the beagle
- [11:34:53] <Crofton|work>
urg
- [11:35:03] <Crofton|work>
I won't complain about boost then
- [11:35:24] <MattJ>
the beagleboard doesn't have ethernet does it? or does it?
- [11:35:32] <koen>
no ethernet
- [11:35:34] <Crofton|work>
usb dongles work
- [11:35:35] <koen>
only usb :)
- [11:35:40] <MattJ>
ok :)
- [11:45:58] * RogerMonk (n=a0740758@nat/ti/session) has joined #beagle
- [11:47:04] <koen>
RogerMonk: good morning!
- [11:48:54] <Crofton|work>
gm
- [11:49:02] <RogerMonk>
Hi all!
- [11:49:14] <RogerMonk>
koen, mru - great job at the show!!
- [11:53:46] <Crofton|work>
RogerMonk, dsplink?
- [11:57:03] <RogerMonk>
Crofton, yep, let's get that nailed this week!
- [11:57:12] <Crofton>
great
- [11:57:20] <Crofton>
do you need anything else from me?
- [11:57:37] <Crofton>
I can push the 128M patch
- [11:57:41] <RogerMonk>
maybe when I get my head into it
- [11:57:44] <Crofton>
ok
- [11:57:58] <Crofton>
I'll be nagging
- [12:03:22] <koen>
RogerMonk: invoice in euros or gbp?
- [12:03:40] <RogerMonk>
either
- [12:04:52] <Crofton|work>
koen, http://www.celinux.org/elc_europe08/
- [12:05:31] <koen>
Crofton|work: I've seen that
- [12:05:41] <koen>
Crofton|work: I might be attending
- [12:05:56] <Crofton|work>
I looked at airfare
- [12:06:11] <Crofton|work>
annoyingly, the off season drop is not as much as I would like
- [12:07:55] * koen wonders why myththv has both a configure and a qmake buildsystem
- [12:08:04] <koen>
(not autotools, a configure shell script)
- [12:08:49] <janneg>
koen: the gcc version check is fixed in trunk
- [12:09:07] <koen>
janneg: and the other braindead checks?
- [12:09:20] <koen>
my x86 cpu can't execute arm code
- [12:09:28] <koen>
nor does uname return arm
- [12:09:34] <janneg>
the configure is mainly used for the internal libav* copies
- [12:09:42] <koen>
has anyone actually *tested* the crosscompile stuff?
- [12:10:04] <koen>
janneg: ffmpeg doesn't have those checks in its configure
- [12:10:17] <janneg>
I don't think so, it is inherited by ffmpeg
- [12:10:31] <janneg>
and the braindead checks are added by us
- [12:12:06] * koen brings town the airport wifi with a videoconference
- [12:15:40] <janneg>
koen: I'll review mythtv's configure in regard of cross-compiling but it will take a couple of days
- [12:17:56] * JuanG (n=Juan@nat/ti/x-2fded60377212239) has joined #beagle
- [12:19:52] * JuanG (n=Juan@nat/ti/x-2fded60377212239) has left #beagle
- [12:20:43] * JuanG (n=Juan@nat/ti/x-fc0735b0c5ef76bc) has joined #beagle
- [12:21:13] * JuanG (n=Juan@nat/ti/x-fc0735b0c5ef76bc) has left #beagle
- [12:21:50] <koen>
drat, my internet time is up
- [12:27:08] <Crofton|work>
bye :)
- [12:30:00] * dschaeffer (n=daniel@timesys-gw0.cust.expedient.net) has joined #beagle
- [12:33:53] <jkridner>
internet time is up?
- [12:34:02] * royerfa (n=x0091539@nat/ti/x-851ca6b7bc7bf900) has joined #beagle
- [12:34:18] <Crofton>
I suspect he paid by the hour in an airport?
- [12:35:42] * BThompson (n=BThompso@nat/ti/x-b0b585118dbe1f97) has joined #beagle
- [12:35:58] <royerfa>
hello everyone
- [12:35:59] <jkridner>
well, I hope this was a positive experiment
- [12:36:05] <jkridner>
hello royerfa.
- [12:36:45] <Crofton>
which experiment?
- [12:41:01] * bazbell (n=a0192809@nat/ti/x-780b948dc34e2e7f) has joined #beagle
- [12:42:02] <jkridner>
having mru and koen go to LRL representing BeagleBoard.org
- [12:43:23] <royerfa>
I am running some SDL program on Angstrom on the EVM
- [12:43:41] <royerfa>
the more simple one ar eworking fine
- [12:43:46] <Crofton|work>
jkridner, thought so
- [12:44:10] <royerfa>
But when I try some games such as alienBlaste which I compiled
- [12:44:18] <royerfa>
I got Segmentation Fault ...
- [12:44:19] <jkridner>
royerfa: great. Do those leave out the 3D portion of SDL?
- [12:45:03] <royerfa>
NO I have only some 2D demo such as testsprite or others
- [12:45:33] <jkridner>
any ideas on how to get your work upstream for everyone to enjoy?
- [12:45:38] <royerfa>
alienBlaster is an opensource Game quite old in 2D
- [12:45:54] <jkridner>
getting it into Angstrom might be the easiest path.
- [12:46:13] <jkridner>
if other people can experiment and see the failures, they might then be able to comment.
- [12:46:14] <royerfa>
I can put the libraries and the executable file on a network
- [12:46:31] <jkridner>
random program that seg faults is difficult to comment upon.
- [12:46:32] <royerfa>
ja
- [12:46:45] <royerfa>
for sure lol
- [12:47:04] <jkridner>
do you have hosting to share the content?
- [12:47:21] <jkridner>
I recommend you create a project on SF or code.google.com.
- [12:47:54] <jkridner>
I'm getting a bit of help and working on adding a project registration process to BeagleBoard.org
- [12:48:07] <jkridner>
see http://beagleboard.org/project
- [12:48:09] <royerfa>
ok
- [12:48:47] <Crofton|work>
http://www.schwardtnet.de/alienblaster/
- [12:49:31] <royerfa>
exatly
- [12:50:01] <Crofton>
I don't see a bb file for it :(
- [12:50:11] <Crofton>
I'd add one, but I am suffering from a lack of motivation
- [12:50:28] <Crofton>
and that would be procrastinating some things I really need to do :)
- [12:51:23] <royerfa>
I wil put my work on google code
- [12:51:53] <royerfa>
How to add some attached file to the google code
- [12:57:03] <ldesnogu>
royerfa: could that be an alignment fault in alienblaster? x86 programmers (that is most of them) don't know what alignment is :)
- [12:59:16] <royerfa>
I don't know either what is alignments :-(
- [13:01:06] <Crofton|work>
ldesnogu, is there a good explanation of alignment anywhere?
- [13:01:52] <ldesnogu>
Crofton|work: hum good question :P
- [13:02:04] * docelic (n=docelic@78.134.192.57) has joined #beagle
- [13:02:30] <jkridner>
does http://www.ibm.com/developerworks/library/pa-dalign/ cover it?
- [13:02:51] <bjdooks>
the general case for arm is that the data should be aligned to the size of the data, so chars on 1 byte alignment, 16bit to 2 byte alignment and 32bit to 4 byte alignment
- [13:03:45] <ldesnogu>
jkridner: yes but it's on the heavy side :)
- [13:04:01] <ldesnogu>
I like bjdooks explanation more
- [13:04:31] <bjdooks>
depending if you have the alignment traps on or not, you get a free shift in the data or the correct action at the expense of taking an abort
- [13:05:13] <ldesnogu>
this alignment trap should be killed, it's only good to make programs slow :-(
- [13:05:29] <bjdooks>
you can turn it on per-process, iirc
- [13:07:38] <ldesnogu>
sure? IIRC it's a global setting available through a proc file
- [13:07:41] <jkridner>
a thought: I doubt it is an x86 vs. ARM alignment issue if it is running on the GP2X. I'd be more inclined to think it was a tool chain or library issue.
- [13:08:02] * Olipro_ (i=Olipro@unaffiliated/olipro) has joined #beagle
- [13:08:06] <ldesnogu>
jkridner: that's what I found after looking a little bit more
- [13:08:30] <ldesnogu>
royerfa: what compiler did you use and what SDL version?
- [13:08:43] <royerfa>
I used the toolchain arm-2008q1
- [13:08:58] <royerfa>
and SDL-1.2.13
- [13:09:01] <ldesnogu>
hum that's bad :)
- [13:09:06] <ldesnogu>
I mean 2008q1
- [13:09:15] <ldesnogu>
could you give 2007q3 a try?
- [13:09:50] <royerfa>
I will try with the 2007q3 and I will tell
- [13:10:24] <jkridner>
2008q1 could explain all.
- [13:15:53] * Olipro__ (i=Olipro@unaffiliated/olipro) has joined #beagle
- [13:23:39] * ciskje (n=francesc@131.114.33.209) has joined #beagle
- [13:24:09] * Olipro (i=Olipro@unaffiliated/olipro) Quit (Nick collision from services.)
- [13:24:12] * Olipro__ is now known as Olipro
- [13:24:19] * Olipro_ (i=Olipro@unaffiliated/olipro) Quit (Nick collision from services.)
- [13:26:02] * dannyBlue|DDd (n=chatzill@89.181.68.180) has joined #beagle
- [13:26:02] <ldesnogu>
royerfa: it would also be nice to get a gdb backtrace of where the segfault occurs; to do that launch the executable this way: "gdb program_name" ; then type "run" ; and when it crashes type "bt"
- [13:26:18] * rsalveti (n=salveti@189.70.143.203) Quit (Read error: 110 (Connection timed out))
- [13:27:56] <ciskje>
Sorry guys for bothering, I'm trying to register to the mailing list but without success. Anyone around there with the same problem?
- [13:32:10] <dannyBlue|DDd>
not me ciskje it worked without problem
- [13:32:27] <ciskje>
:-/
- [13:32:38] <dannyBlue|DDd>
do you use a gmail account?
- [13:32:38] <ciskje>
Thanks anyway
- [13:32:43] <ciskje>
Yes
- [13:32:54] <dannyBlue|DDd>
hum what error do you get?
- [13:33:32] <ciskje>
I don't rember exactly, give me few seconds to give it another try ...
- [13:35:18] * rsalveti (n=salveti@200.184.118.132) has joined #beagle
- [13:41:18] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
- [13:42:26] <ciskje>
The error: "An error occurred while processing your request..." then "Please try again in a few moments".
- [13:42:40] * DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) has joined #beagle
- [13:49:12] <ciskje>
The same error in few retries
- [13:58:05] <Crofton|work>
sakoman, do you need a special bitbake to fetch from a git branch?
- [13:59:02] <sakoman_>
Crofton: practically speaking yes
- [13:59:09] <sakoman_>
I use the svn head now
- [13:59:32] <Crofton|work>
I am on 1.8
- [13:59:38] <sakoman_>
otherwise you need to clean out the source cache every time your change a SRCREV
- [14:00:03] <sakoman_>
try what you have and see if you can live with it
- [14:00:34] <Crofton|work>
I'll ask RP if he plans to move this into 1.8 also
- [14:00:52] <Crofton|work>
I need to clean up mlo/u-boot/kernel versions
- [14:16:46] * sakoman_ (n=sakoman@static-74-41-60-154.dsl1.pco.ca.frontiernet.net) Quit (Read error: 104 (Connection reset by peer))
- [14:30:15] <sakoman>
4
- [14:31:30] * sakoman_ (n=sakoman@static-74-41-60-154.dsl1.pco.ca.frontiernet.net) has joined #beagle
- [14:45:06] * trickie is now known as trickie|away
- [14:48:29] * ciskje (n=francesc@131.114.33.209) Quit ("using sirc version 2.211+KSIRC/1.3.12")
- [14:52:23] * dannyBlue|DDd (n=chatzill@89.181.68.180) Quit ("ChatZilla 0.9.83 [Firefox 3.0.1/2008070208]")
- [14:52:32] <shoragan>
jkridner?
- [14:52:49] <jkridner>
?
- [15:06:40] * Viral_Sachde (n=Viral_Sa@122.167.154.2) Quit (Read error: 110 (Connection timed out))
- [15:34:29] * JoeBorn__ (n=jborn@adsl-75-3-8-191.dsl.chcgil.sbcglobal.net) Quit ("open.neurostechnology.com")
- [15:35:43] * mugenkun (n=sddd@201.37.105.27) has joined #beagle
- [15:36:04] * JuanG_ (n=Juan@nat/ti/x-b385b72e2a866620) has joined #beagle
- [15:37:45] * dirk2 (n=dirk@F3380.f.strato-dslnet.de) has joined #beagle
- [15:56:06] <koen>
re
- [15:56:12] * koen is in a dutch train now
- [15:56:27] * JuanG_ (n=Juan@nat/ti/x-b385b72e2a866620) has left #beagle
- [16:01:45] <Crofton>
koen, this afternoon I'd like to convert the OE recipe to use sakomans git branch
- [16:02:04] <Crofton>
The kernel I built with Paul's patch didn't work so good
- [16:02:14] <koen>
is that wise?
- [16:02:25] <koen>
iirc that doesn't have all patches we use in oe
- [16:02:26] <Crofton>
we also need to work out what mlo/xloader/u-boot combo to use with Angstrom
- [16:02:52] <Crofton>
His seems stable :)
- [16:03:04] <Crofton>
at least until Tony gets the backlog down
- [16:03:16] <koen>
I see no real benifit from it
- [16:03:28] <koen>
and it meeans everyone needs to download 250MB of sakomans server
- [16:03:38] <Crofton>
well, what I built the other day wasn't working so good
- [16:03:46] <Crofton>
good point
- [16:03:52] <Crofton>
that is also annoying
- [16:04:15] <Crofton>
I'm also wondering if my problems relate to mlo/u-boot
- [16:05:08] <dirk2>
Crofton: Which mlo/uboot do you use? I think the MLO/xload from top of http://code.google.com/p/beagleboard/wiki/BeagleSourceCode are okay. And uboot built from sakoman's git should be okay, too
- [16:05:27] <Crofton>
heh, I'd need to check md5sums at this point :)
- [16:06:24] <Crofton>
getting this sorted out is this afternoon's project
- [16:06:41] <Crofton>
after checking the kernel I jsut built for davinci-sffsdr
- [16:08:07] <Crofton>
but first, lunch
- [16:08:12] <koen>
RogerMonk: I'll upoad some pictures when I get back home
- [16:09:17] <dirk2>
jkridner or anybody else familiar with google wiki: Is there any way to see *what* changed in a page (page history)? Looking at http://code.google.com/p/beagleboard/w/list Gerald changed something, but what??
- [16:11:25] <dirk2>
pictures: Anybody saw anything from Khasim's OSSCamp http://osscamp.in/index.php/OSScamp_Bengaluru_Mobile_2008 ?
- [16:19:37] <koen>
bbl
- [16:37:50] * Beagle2 (n=Beagle2@a83-132-76-156.cpe.netcabo.pt) Quit ()
- [16:54:53] * [denix] is now known as denix
- [16:57:37] * bazbell (n=a0192809@nat/ti/x-780b948dc34e2e7f) Quit ("Leaving.")
- [17:06:06] * bazbell (n=a0192809@nat/ti/x-24f79122fc63d4d3) has joined #beagle
- [17:14:05] <royerfa>
Can someone point me to a good way to debug on the Target
- [17:14:38] <Crofton|work>
ipkg install gdb?
- [17:14:45] <royerfa>
gdb is not in the busybox but is in the tool cahin so how to use it ??
- [17:14:50] <royerfa>
gdb
- [17:15:28] <Crofton|work>
what rootfs are you using?
- [17:16:31] <royerfa>
the one with Angstrom distrib
- [17:16:53] <sakoman_>
Crofton: opkg not ipkg :-)
- [17:16:55] <Crofton|work>
opkg install gdb should get gdb on the rootfs
- [17:16:59] <Crofton|work>
yeah
- [17:17:01] <Crofton|work>
I know
- [17:17:14] <sakoman_>
the fingers don't ;-)
- [17:17:20] <Crofton|work>
sorry, I was talking to someone on #gumstix this AM
- [17:17:21] <sakoman_>
same problem here
- [17:17:49] <Crofton|work>
invariably I type the wrong thing anyway
- [17:18:00] <royerfa>
oki
- [17:22:28] <Crofton|work>
ok, my MLO is ok, but u-boot is not flash_u-boot.bin
- [17:23:13] <royerfa>
OK but to use opkg I need internet connexion on the board no ?
- [17:23:30] <Crofton|work>
or copy ipk's to the fs
- [17:25:09] <dirk2>
What do you mean with "flash_u-boot.bin"? I don't think there is a special uboot version for flash.
- [17:25:29] <Crofton|work>
file name from the url you mentioned
- [17:25:46] <Crofton|work>
http://code.google.com/p/beagleboard/wiki/BeagleSourceCode
- [17:25:48] <dirk2>
ah ;)
- [17:25:51] <Crofton|work>
first u-boot.bin
- [17:25:57] <dirk2>
yes, bad naming ;)
- [17:26:13] <Crofton|work>
yep :)
- [17:26:24] <Crofton|work>
That page should list md5sums ...\
- [17:26:31] <dirk2>
I don't use the uboot from there, only xloader and MLO. Uboot I use is from sakoman's git
- [17:27:39] <dirk2>
For the old page http://code.google.com/p/beagleboard/wiki/DeprecatedBeagleSources we had md5sums, but Khasim missed to add them to this new page
- [17:28:12] <dirk2>
Well, it's a wiki. If you have them already, add them, or send them here and I press the edit button
- [17:29:13] <Crofton|work>
heh
- [17:29:19] <Crofton|work>
I can get the ones I have
- [17:31:41] <dirk2>
Reload the page, I added a new column "md5sum"
- [17:32:32] <jkridner|work1>
added a board registration page.
- [17:33:15] <jkridner|work1>
not exactly sure what we'll use it for, but it is just somewhat helpful to keep track of things. I should have had a "status" field. I think I'll add that now. Then, I'm going to need an edit skin.
- [17:34:07] <Crofton|work>
hmm, so did I :)
- [17:34:26] <dirk2>
:)
- [17:35:17] <dirk2>
and now we need somebody showing me where in Google wiki I can see the page history
- [17:36:48] <jkridner|work1>
page history requires downloading the project via SVN.
- [17:37:02] <jkridner|work1>
instead of using the 'trunk', use 'wiki'.
- [17:39:33] * Crofton|work wonders what gets built from the OE recipe ...
- [17:41:15] * Beagle9 (n=Beagle9@OTWAON23-1168100727.sdsl.bell.ca) has joined #beagle
- [17:43:05] <Beagle9>
IS there an easy way to hook a camera up to the Beagle?
- [17:43:25] <DJWillis>
USB?
- [17:43:25] <dirk2>
jkridner|work1: sakoman's flash procedure is added to the beagle news. In this description there is a small error for uboot > 1.3.3 . The first part, the xloader one, needs 'nand ecc sw', too. Else you will 'brick' your board. I tried it ;) Maybe you like http://elinux.org/BeagleBoardNAND , too
- [17:44:03] <jkridner>
k. I'll pull it off of the beagle news then.
- [17:44:29] <sakoman_>
jkridner: I will fix it right now
- [17:44:55] <jkridner>
I'll add both and really confuse people.
- [17:46:37] <Beagle9>
either a USB camera or any camera or imager really...
- [17:46:54] <dirk2>
and I'm wrong above, too ;) I meant 'nand ecc hw' for xloader...
- [17:47:34] <sakoman>
dirk2: see if it is right now
- [17:48:03] <jkridner>
why don't we get sakoman's edits into the the wiki page?
- [17:48:12] <jkridner>
that would probably be the more "community" thing to do.
- [17:48:16] <jkridner>
no?
- [17:48:35] <dirk2>
sakoman: :)
- [17:48:37] <jkridner>
looks like sakoman's page is linked anywah.
- [17:48:44] <jkridner>
anywah=anyway
- [17:49:44] <sakoman>
44
- [17:49:53] <Crofton|work>
42 you mean
- [17:50:26] <sakoman_>
damn kvm switching isn't working right ;-)
- [17:58:54] <royerfa>
I tried to install gdb with opkg
- [17:59:21] <royerfa>
and the package of gdb provided on this pages http://www.angstrom-distribution.org/feeds/2008/ipk/glibc/armv6/debug/
- [17:59:47] <royerfa>
but the installation doesn't work properly
- [18:00:24] * Beagle9 (n=Beagle9@OTWAON23-1168100727.sdsl.bell.ca) Quit ()
- [18:00:51] * bazbell (n=a0192809@nat/ti/x-24f79122fc63d4d3) Quit (Remote closed the connection)
- [18:01:37] * royerfa (n=x0091539@nat/ti/x-851ca6b7bc7bf900) has left #beagle
- [18:07:22] * bazbell (n=a0192809@nat/ti/x-d8972789c79ec87b) has joined #beagle
- [18:10:20] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
- [18:16:57] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
- [18:39:19] <Crofton|work>
hmm, what is the magic needed to get angstrom to mount a direcory via nfs?
- [18:49:05] * dirk2 (n=dirk@F3380.f.strato-dslnet.de) has left #beagle
- [18:55:49] <mugenkun>
hey guys, I would like to get a board, is there yet any places selling it or something?
- [18:57:54] <mugenkun>
or is there any prevision for the release?
- [19:00:21] * esden is now known as esden`away
- [19:02:31] <DJWillis>
sakoman_: ping, are you about for a really quick question?
- [19:03:14] <DJWillis>
sakoman_: did you have an MMC driver you worked on at some point that supported 2 cards?
- [19:13:29] * jkridner|work1 (n=a0321898@nat/ti/x-718527b3c8a6347a) Quit (Remote closed the connection)
- [19:13:42] * jkridner|work (n=a0321898@nat/ti/x-cc5686a92a74e1da) has joined #beagle
- [19:15:35] <sakoman_>
DJWillis: yeah, I think at one point I did have things set up for 2 cards
- [19:15:45] <sakoman_>
But IIRC I removed it
- [19:15:58] <sakoman_>
And of course didn't save it
- [19:16:05] <DJWillis>
sakoman_: don't suppose you have it still do you, ahhh
- [19:16:14] <sakoman_>
But it should be quite simple to recreate
- [19:17:25] <DJWillis>
ok, that's fine then, don't worry. It was causing me a few problems but I am sure they will work out.
- [19:21:06] <DJWillis>
Thanks however, trying to finish off some of the outstanding on the todo, it just never seems to shrink.
- [19:21:32] <sakoman_>
DJWillis: I know the feeling!
- [19:52:49] * docelic (n=docelic@78.134.192.57) Quit ("http://www.spinlocksolutions.com/")
- [19:58:54] <koen>
re
- [20:04:08] <Crofton|work>
koen, what's the trick to mounting a nfs file system on the beagle
- [20:04:16] <koen>
no trick
- [20:04:19] <Crofton|work>
I feel like I need mount.nfs
- [20:04:28] <koen>
mount -t nfs 8394.45.35.35:/foo bar
- [20:04:38] <Crofton|work>
http://rafb.net/p/zv7Wub34.html
- [20:05:01] <Crofton|work>
same with -t nfs and -t nfs4
- [20:05:29] <koen>
cat /proc/filesystems ?
- [20:05:35] <Crofton|work>
heh
- [20:05:50] <Crofton|work>
root@beagleboard:~# cat /proc/filesystems | grep nfs
- [20:05:50] <Crofton|work>
nodev nfs
- [20:05:50] <Crofton|work>
nodev nfs4
- [20:06:01] <Crofton|work>
needless to say, I am frustrated
- [20:07:11] <Crofton|work>
man mount suggests I need mount.nfs
- [20:07:37] <Crofton|work>
a, part of nfs utilities
- [20:07:59] * jkridner|work (n=a0321898@nat/ti/x-cc5686a92a74e1da) Quit (Remote closed the connection)
- [20:08:15] <banderson>
http://www.linuxdevices.com/news/NS9152590220.html
- [20:08:51] <Crofton|work>
F9 has mount.nfs[4]
- [20:09:43] <banderson>
wonder if they use oe?
- [20:09:50] <banderson>
timesys that is
- [20:10:44] <Crofton|work>
I don't know much about timesys, other than they exist
- [20:11:05] <Crofton|work>
I think I see a timesys addr appear here occasionally
- [20:11:31] <koen>
timesys sends like 3 PR releases to linuxdevices a day
- [20:11:33] <banderson>
ya..just looked and saw a one earlier...don't know much either
- [20:11:37] <koen>
sadly enough, it works
- [20:11:48] <banderson>
didn't know that...interesting
- [20:12:11] <koen>
linuxdevices only publishes stuff you send them
- [20:12:24] <koen>
can't blame timesys for exploiting their laziness :)
- [20:12:56] * Crofton|work thinks OE should send them press releases every time we commit :)
- [20:13:26] <Crofton|work>
I'll talk to RP about adding an "email Linux Devices on commit" hook
- [20:15:55] <jkridner>
well, we didn't send them any Beagle announcement materials when they did the article on Beagle.
- [20:16:10] <jkridner>
so, they are at least a little proactive.
- [20:16:42] <koen>
jkridner: AIUI someone else sent it
- [20:16:49] * ciskje (n=francesc@87.19.200.110) has joined #beagle
- [20:16:51] <jkridner>
probably.
- [20:17:31] <jkridner>
is there a Raphael Studer lurking here?
- [20:18:04] * RogerMonk (n=a0740758@nat/ti/x-230f1101a7db6561) has left #beagle
- [20:19:14] <Crofton|work>
crap, this nfs thing has me stumped
- [20:21:32] * BeagleLogBot (n=PircBot@ec2-75-101-156-174.compute-1.amazonaws.com) Quit (Remote closed the connection)
- [20:21:45] * BeagleLogBot (n=PircBot@ec2-75-101-156-174.compute-1.amazonaws.com) has joined #beagle
- [20:22:09] * BeagleLogBot_ (n=PircBot@ec2-75-101-156-174.compute-1.amazonaws.com) Quit (Remote closed the connection)
- [20:22:26] * BeagleLogBot_ (n=PircBot@ec2-75-101-156-174.compute-1.amazonaws.com) has joined #beagle
- [20:22:26] * Topic is 'Welcome to #Beagle | Discussion about the OMAP3 Beagle Board - http://beagleboard.org | Beagle search tools are on #dashboard at irc.gimp.org, NOT here ;) | Log is at http://www.beagleboard.org/irclogs'
- [20:22:26] * Set by jkridner on Tue Jun 10 00:33:18 CDT 2008
- [20:24:21] <koen>
jkridner: mru has a new theory on beagle crashes: the git kernel is pushing the dispc too hard
- [20:24:45] <koen>
jkridner: playing 480p video is fine, 720p video locks up the beagle after a few minutes
- [20:25:23] <jkridner>
possible. I believe the published spec is still 65MHz pixel clock, though I believe there is strong proof for 75MHz being ok.
- [20:25:42] <jkridner>
I'm not sure what other clocks are involve in generating the pixel clock and if any of those go out of spec.
- [20:26:09] <koen>
we were running it at 72MHz, which seemed ok
- [20:26:30] <koen>
but iirc mru said dss1 was set too high
- [20:28:17] * Olipro_ (i=Olipro@unaffiliated/olipro) has joined #beagle
- [20:30:29] <jkridner>
k. running one of the clocks too high could definitely cause issues.
- [20:35:01] * koen sums up debian on beagle: "So instead of a system that is fast enough for 720p mpeg4 video, you get a system that can hardly play an mp3 without skipping and needs 10 times more space."
- [20:36:10] <koen>
debian/arm is so awesome
- [20:36:14] <koen>
if it were 1995
- [20:36:54] <BThompson>
it must not be built very efficiently, what kind of arm was it built for if it cannot run on a cortex A8?
- [20:37:12] <koen>
debian is still using armv4 hardfpa OABI binaries
- [20:37:37] <BThompson>
ahh
- [20:37:41] <koen>
the unofficial port is using armv4t softvfp
- [20:37:44] <koen>
so still slow
- [20:38:18] <koen>
running debian on arm is like running gentoo with 8MB of ram
- [20:38:59] <koen>
the fedora/arm people are a bit more cluefull
- [20:39:05] <koen>
but more underground
- [20:39:17] <ciskje>
Hi
- [20:39:17] <koen>
the fedora people at LRL didn't even know about the arm port :)
- [20:39:24] <koen>
hey ciskje
- [20:39:30] <ciskje>
Hi guys
- [20:40:01] <ciskje>
I'm new on the channel and new on the project. So ... before lurking I preferred introducing. Hope this doesn't hurt anyone
- [20:41:53] <koen>
anvils hurt, introductions don't :)
- [20:42:04] <ciskje>
;-)
- [20:42:40] <ciskje>
I saw you have performance problem. Maybe I could give my 2 cents. Which is your setup?
- [20:44:35] <jkridner>
koen: gentoo does cross-compile.
- [20:45:03] <jkridner>
maybe not as mature as OE for cross-compile, but I actually had a Gentoo build up first.
- [20:45:14] <jkridner>
(without trying to build on Beagle)
- [20:45:21] <jkridner>
still, OE quickly out-paced it.
- [20:45:52] * dannyBlue|DDd (n=chatzill@89.181.68.180) has joined #beagle
- [20:46:16] * Olipro (i=Olipro@unaffiliated/olipro) Quit (Read error: 110 (Connection timed out))
- [20:46:59] <koen>
jkridner: it wasn't a jab at gentoo, it was more like "try to compile stuff when you have no ram"
- [20:47:38] <jkridner>
understood. just didn't want it left that Gentoo requires big RAM on the target, since you don't have to compile on the target.
- [20:47:45] <koen>
(I cheated and put 256MB ram in my sparc5 running gentoo)
- [20:48:40] <jkridner>
I certainly get that trying to compile in 8MB would be pointless. I'm not sure about Beagle's 128MB, but that is probably too small to handle building the compiler. Small projects can compile on Beagle fine.
- [20:49:35] * koen had a compiler on beagle before switching to a 4gb sd card
- [20:49:55] <koen>
128mb is plenty, provided you steer clear of c++ code
- [20:50:12] <jkridner>
The Ubuntu stuff is bloated, but it isn't too bad. I'd guess Debian would actually be a bit lighter.
- [20:50:34] <jkridner>
still, it is certainly behind OE. must acknowledge that.
- [20:51:28] <jkridner>
ciskje: who was pointing to a performance problem?
- [20:53:10] <ciskje>
No guys, I got half the lines.
- [20:53:50] <ciskje>
And a quarter of the meaning.
- [20:54:00] <ciskje>
Sorry
- [20:54:33] <koen>
jkridner: debian (and hence ubuntu) have a very simplistic view of 'architecture'
- [20:54:47] <koen>
jkridner: their package format only knows 'arm' (yes, that's hardcoded)
- [20:55:05] <koen>
so you get stuck with the lowest common demoninator
- [20:55:20] <koen>
which in EABI is armv4t
- [20:56:08] <jkridner>
what about http://mojo.handhelds.org/frisky-armv6el-vfp ?
- [20:56:12] * bazbell (n=a0192809@nat/ti/x-d8972789c79ec87b) Quit (Remote closed the connection)
- [20:56:15] <koen>
that problem is far more fundamental than their assumption that every device has a spinning harddrive
- [20:56:29] <jkridner>
old ABI?
- [20:56:58] <jkridner>
well, that was a starting premise--they can evolve, can't they?
- [20:56:59] <koen>
jkridner: if you take a look in that armv6 repo, you'll see foo_1.2.5_arm.deb
- [20:57:23] <koen>
jkridner: that relies on people not adding other repositories
- [20:57:50] <koen>
jkridner: and you can see the problem when an "arm" deb doesn't run on your device
- [20:58:08] <jkridner>
right. for *some* people, I'd think that wouldn't be a problem.
- [20:58:28] <jkridner>
as long as you have the one distro for the one platform and that one is yours. :)
- [20:59:17] <jkridner>
not "yours", but whoever is looking for packages.
- [20:59:31] <koen>
we looked at using .deb for angstrom to make it more debian like, but haven't done it because the format is just to dumb
- [20:59:47] <jkridner>
I'm not saying that Debian/Ubuntu/etc. are better solutions, but it isn't like they cannot work at all.
- [20:59:51] <koen>
(set ANGSTROM_PKG_FORMAT = "deb" in local.conf and find out)
- [21:00:28] <jkridner>
is it beyond repair?
- [21:00:43] <jkridner>
can the packaging add the desired comprehension?
- [21:00:49] <koen>
it relies on people not installing packages from "foreign" sources
- [21:01:13] <koen>
experience has proven that people will install anything google digs up
- [21:02:17] * koen spent far too much time fixing stuff caused by people installing random shit and the package manager accepting it
- [21:05:27] <koen>
jkridner: one project I'd like to see targetting cortex-a8 is http://clang.llvm.org/
- [21:06:50] <mugenkun>
anyone know where can i get a beagleboard? =(
- [21:08:28] <jkridner>
think that would build a compiler?
- [21:09:17] <koen>
jkridner: I've been a fan of llvm since bossa last year, it does a better job at compiling code than gcc
- [21:09:20] <jkridner>
mugenkun: you should be able to find out in about 1 week.
- [21:09:37] <jkridner>
er, I meant would it build a Linux kernel.
- [21:09:53] <jkridner>
I'd guess no GCC flags?
- [21:10:07] <jkridner>
oh...
- [21:10:14] <jkridner>
it does say GCC compatibility.
- [21:10:24] <mugenkun>
jkridner: in the website?
- [21:10:50] <jkridner>
yes. first notice of when boards will be available and where will be through the mailing list, so join that.
- [21:11:06] <jkridner>
then the link will go live on the website.
- [21:11:12] <koen>
jkridner: you can use llvm as backend for gcc, clang is the frontend that should replace gcc
- [21:11:18] <mugenkun>
already joined :) thanks!
- [21:12:06] <jkridner>
there is a request list, 'revb-request@beagleboard.o...', but we are getting close to going public, so getting on that queue isn't going to help much now.
- [21:13:34] * suihkulokki notices koen is atleast 2 years behind in debian/arm progress
- [21:13:44] <koen>
jkridner: you can use llvm as, well, a vm and jit as well, but that's quite slow and requires like 8MiB of support code
- [21:14:07] <koen>
suihkulokki: armel is the one and only arm port nowadays??
- [21:14:14] <suihkulokki>
eabi port is now official and scheduled to be released in lenny
- [21:14:23] <suihkulokki>
eabi = armel
- [21:14:31] <mugenkun>
ok
- [21:14:48] <mugenkun>
I'm watching the mailing list then
- [21:14:50] <suihkulokki>
and we do not intend to release oldabi arm for lenny
- [21:14:53] <mugenkun>
thanks
- [21:16:20] <ciskje>
Is there any emulation environment for those who doesn't have board ?
- [21:16:42] <koen>
iirc qemu cvs supports armv7 nowadays
- [21:18:05] <ciskje>
Yes. I need to compile gcc 3.x for qemu.
- [21:18:15] <koen>
right
- [21:19:04] * mugenkun (n=sddd@201.37.105.27) Quit (Remote closed the connection)
- [21:19:56] <ciskje>
Any hints, docs, links for getting (somewhat) your current environment running without bothering people on the channel?
- [21:23:56] <jkridner>
http://elinux.org/BeagleBoard
- [21:24:37] <jkridner>
I'm not aware of any setup using qemu right now, though that would be a VERY welcome project.
- [21:25:06] <ciskje>
Thanks for the link
- [21:25:31] <ciskje>
The qemu approach would be the only one for me right now
- [21:25:52] <ciskje>
I think I'll give it a try
- [21:28:16] * BThompson (n=BThompso@nat/ti/x-b0b585118dbe1f97) Quit ("Trillian (http://www.ceruleanstudios.com")
- [21:28:50] * dschaeffer (n=daniel@timesys-gw0.cust.expedient.net) has left #beagle
- [21:29:56] * Olipro_ is now known as Olipro
- [21:44:20] <koen>
one mythtv hurdle down: http://svn.mythtv.org/trac/log/trunk/mythtv/configure
- [21:57:46] * dannyBlue|DDd (n=chatzill@89.181.68.180) Quit (Read error: 104 (Connection reset by peer))
- [21:58:02] * dannyBlue|DDd (n=chatzill@89.181.68.180) has joined #beagle
- [22:09:27] * JoeBorn_ (n=jborn@65.42.208.133) has joined #beagle
- [22:14:03] * RogerMonk (n=a0740758@nat/ti/x-47f42c88e250a42d) has joined #beagle
- [22:22:46] <Crofton|work>
koen, can you humour me and try mounting a nfs directory on your beagle?
- [22:22:58] <Crofton|work>
the guys in #oe swear it works on their boards
- [22:23:49] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
- [22:23:57] * JoeBorn_ is now known as JoeBorn
- [22:27:08] * mickeyl is now known as mickey|zzZZzz
- [22:31:10] * BeagleLogBot2 (n=PircBot@ec2-75-101-156-174.compute-1.amazonaws.com) Quit (Remote closed the connection)
- [22:33:43] <ciskje>
The whole process (plus the packages that were on the queue) nearly "boiled" my machine
- [22:41:44] * RogerMonk (n=a0740758@nat/ti/x-47f42c88e250a42d) Quit (Remote closed the connection)
- [22:50:33] * Olipro (i=Olipro@unaffiliated/olipro) Quit (Read error: 110 (Connection timed out))
- [22:55:59] * rsalveti (n=salveti@200.184.118.132) Quit (Read error: 113 (No route to host))
- [22:56:34] * rsalveti (n=salveti@189.70.143.203) has joined #beagle
- [22:59:52] * bazbell (n=a0192809@nat/ti/x-718db01e8e9c72c1) has joined #beagle
- [23:01:00] * ciskje (n=francesc@87.19.200.110) Quit ("using sirc version 2.211+KSIRC/1.3.12")
- [23:04:01] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) has joined #beagle
- [23:13:47] * JoeBorn (n=jborn@65.42.208.133) Quit (Read error: 113 (No route to host))
- [23:28:32] * dannyBlue|DDd (n=chatzill@89.181.68.180) Quit ("I was here.")
- [23:46:06] <mru>
does anyone here know how the clock configuration in clock34xx.[ch] is supposed to work?
- [23:54:52] * paul_pwsan (n=paul@utopia.booyaka.com) has joined #beagle
- [23:56:07] <paul_pwsan>
mru: what info are you looking for?
- [23:57:27] <mru>
I'm trying to make sense of it
- [23:57:38] <mru>
mostly in relation to the dss1_alwon_fck clock
- [23:57:38] <paul_pwsan>
sakoman_: i have a revised set of I2C patches here if you (or anyone else for that matter) would like to give them a test before they are posted to the list
These logs were automatically created by BeagleLogBot on
irc.freenode.net
using the Java IRC LogBot.
See http://microformats.org/wiki/mflogbot for more information.