Join the chat at beagleboard.org/chat
IRC Log for 2008-09-17
Timestamps are in UTC.
- [00:02:08] * Openfree (n=df@218.82.126.34) has joined #beagle
- [00:06:13] * Openfree (n=df@218.82.126.34) Quit (Remote closed the connection)
- [00:06:28] * rsalveti (n=salveti@189.70.219.182) Quit (Remote closed the connection)
- [00:24:00] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
- [00:24:24] * likewise (n=chatzill@82-171-51-231.ip.telfort.nl) Quit ("ChatZilla 0.9.83 [Firefox 3.0/2008061015]")
- [00:35:10] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit (Remote closed the connection)
- [00:35:55] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
- [01:13:58] * khasim (n=a0393720@203.101.61.10) has joined #beagle
- [01:31:33] * artyom1 (n=artyomt@198.78.191.90.dyn.estpak.ee) Quit (Read error: 110 (Connection timed out))
- [01:32:59] * artyomt (n=artyomt@198.78.191.90.dyn.estpak.ee) has joined #beagle
- [01:43:42] <pbrook>
Is there somewhere to list hardware that does not work with the BeagleBoard?
- [01:44:37] <pbrook>
I have an Edimax/MosChip "USB2 ethernet adapter" that only actually works when connected to a USB1 port.
- [01:45:28] <ds2>
what does it do?
- [01:46:17] <pbrook>
It completely craps out when you try to talk to it at high speed. Doesn't even get as far as device enumeration.
- [01:46:58] <ds2>
desc craps out
- [01:47:07] <ds2>
oops? freeze?
- [01:47:23] <pbrook>
usb 5-1: device descriptor read/64, error -71
- [01:47:31] <pbrook>
usb 5-1: device not accepting address 88, error -71
- [01:47:42] <pbrook>
<kernel stops talking to device>
- [01:47:59] <ds2>
otg or host only?
- [01:48:31] <pbrook>
The problem isn't actually specific to the BeagleBoard. Tried it on a windows PC and got "Device has misbehaved and is not recognised".
- [01:48:33] <ds2>
and does it work through a usb2 hub?
- [01:48:56] <ds2>
could be a power thing
- [01:48:58] <pbrook>
It desn't work through a USB2 hub.
- [01:49:40] <pbrook>
It works through a USB1 hub, and wwhen connected directly to a port that can drop down to usb1 (I think the beagle OTG port can do this).
- [01:50:00] <ds2>
yep
- [01:50:09] <pbrook>
However I'd expect the working cases to be fairly rare for the BB.
- [01:51:43] <pbrook>
I've found evidence on the internet that other people have seen the same problem with this device.
- [01:52:20] <pbrook>
Definitely not a power problem.
- [01:53:44] <ds2>
yuck
- [02:06:05] * _AV500_ (n=av500@213.221.98.41) has joined #beagle
- [02:06:05] * AV500 (n=av500@213.221.98.41) Quit (Read error: 104 (Connection reset by peer))
- [02:16:11] * khasim (n=a0393720@203.101.61.10) Quit (Remote closed the connection)
- [02:32:24] * JoeBorn (n=rootmeis@h-68-166-253-118.chcgilgm.dynamic.covad.net) has joined #beagle
- [02:37:24] * JoeBorn (n=rootmeis@h-68-166-253-118.chcgilgm.dynamic.covad.net) Quit (Read error: 104 (Connection reset by peer))
- [02:38:04] * JoeyBorn (n=rootmeis@h-68-166-253-118.chcgilgm.dynamic.covad.net) has joined #beagle
- [02:39:10] * rsalveti (n=salveti@189.70.219.182) has joined #beagle
- [02:42:53] * dcordes (n=dcordes_@unaffiliated/dcordes) has joined #beagle
- [02:52:08] * sakoman_ (n=sakoman@12.154.210.2) has joined #beagle
- [02:56:14] * cmonexaway (i=xy6091@pool-7733.adsl.interware.hu) has joined #beagle
- [02:56:39] * Olipro_ (n=Olipro@uncyclopedia/Olipro) has joined #beagle
- [02:58:02] * JoeXBorn (n=rootmeis@h-68-166-253-102.chcgilgm.dynamic.covad.net) has joined #beagle
- [02:58:54] * JoeyBorn (n=rootmeis@h-68-166-253-118.chcgilgm.dynamic.covad.net) Quit (Read error: 104 (Connection reset by peer))
- [03:02:00] * cmonex (i=xy6091@pool-7733.adsl.interware.hu) Quit (Read error: 60 (Operation timed out))
- [03:02:04] * dcordes (n=dcordes_@unaffiliated/dcordes) Quit ("leaving")
- [03:02:34] * JoeBorn (n=rootmeis@h-68-164-11-248.chcgilgm.dynamic.covad.net) has joined #beagle
- [03:03:29] * JoeXBorn (n=rootmeis@h-68-166-253-102.chcgilgm.dynamic.covad.net) Quit (Read error: 104 (Connection reset by peer))
- [03:08:31] * JoeBorn (n=rootmeis@h-68-164-11-248.chcgilgm.dynamic.covad.net) Quit (Read error: 104 (Connection reset by peer))
- [03:08:49] * JoeBorn (n=rootmeis@h-68-164-11-248.chcgilgm.dynamic.covad.net) has joined #beagle
- [03:10:46] * smackpotato (n=user@70.51.253.31) has joined #beagle
- [03:11:22] <smackpotato>
so what can i run with a beagle board
- [03:12:09] <smackpotato>
can i replace my intel 1.6 pentium
- [03:12:32] * Olipro (n=Olipro@uncyclopedia/Olipro) Quit (Connection timed out)
- [03:18:10] * khasim (n=a0393720@203.101.61.10) has joined #beagle
- [03:22:21] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) Quit ("Trillian (http://www.ceruleanstudios.com")
- [03:30:01] <jkridner>
smackpotato: that depends on what you do with your 1.6 pentium.
- [03:30:14] <jkridner>
you can run FFmpeg to do 720p video.
- [03:30:28] <jkridner>
3D graphics and DSP acceleration are still in the works.
- [03:30:35] <jkridner>
you can play Quake2.
- [03:31:06] <jkridner>
(Quake3 is likely to be ported later this year to reasonable performance once the 3D graphics is out)
- [03:31:16] <jkridner>
you can use it as a MythTV frontend (project on-going)
- [03:31:30] <jkridner>
http://beagleboard.org/project for list of registered projects.
- [03:31:44] * artyom1 (n=artyomt@198.78.191.90.dyn.estpak.ee) has joined #beagle
- [03:31:51] <jkridner>
today, it is more for hobbyists rather than a computer replacement.
- [03:32:24] <jkridner>
I expect for things to get a bit repackaged by the community into something that is more friendly off-the-shelf.
- [03:32:38] <ds2>
keep the entire house heated in the artic winteer ;)
- [03:32:40] <jkridner>
hope that answers your question, smackpotato.
- [03:32:59] <jkridner>
I don't think a Beagle would do that very well at <2W. ;)
- [03:33:14] <smackpotato>
sure
- [03:33:15] * artyomt (n=artyomt@198.78.191.90.dyn.estpak.ee) Quit (Read error: 110 (Connection timed out))
- [03:33:18] <smackpotato>
thanks
- [03:33:33] <jkridner>
I'm sure other people have lots of good examples of what you can do with a Beagle....
- [03:33:46] <jkridner>
the low power means you can put it into a lot of places you wouldn't normally put a computer.
- [03:34:07] <ds2>
too bad ;)
- [03:35:02] <jkridner>
ds2 must be in a good mood this evening. :)
- [03:35:08] <jkridner>
it was a rough day for me.
- [03:35:20] <jkridner>
had to take the cat to the vet. might lose part of her tail.
- [03:35:30] <ds2>
in pdx for lpc
- [03:35:42] <jkridner>
I love pdx.
- [03:35:48] <smackpotato>
take a guess. when will it be able to run a browser and also mplayer
- [03:35:51] <jkridner>
mom grew up there.
- [03:36:00] <jkridner>
both run today.
- [03:36:06] <smackpotato>
well
- [03:36:10] <jkridner>
firefox 3, epiphany, and mplayer.
- [03:36:20] <smackpotato>
cool
- [03:36:21] <jkridner>
l-o git, however, has problems with audio.
- [03:36:30] <ds2>
coming from a 2420 + bt dun
- [03:37:08] <jkridner>
someone needs to debug the ASoC driver, otherwise there are some crashes in mplayer (if you don't turn the audio off)
- [03:37:15] <jkridner>
ds2: n810?
- [03:37:34] <smackpotato>
what is the best monitor to hook up
- [03:37:42] <ds2>
n800
- [03:37:45] <jkridner>
other browsers running include fennec, midori, ... I forget others.
- [03:38:04] <jkridner>
make sure it is a DVI-D monitor.
- [03:38:12] <jkridner>
The Dell's are known to work nicely.
- [03:38:15] <smackpotato>
ok
- [03:38:21] <jkridner>
I have an LG.
- [03:38:36] <jkridner>
unknown model.
- [03:39:07] <jkridner>
you are generally fine with a computer monitor. some HDTVs require specific timings which means you may have to recompile the kernel or even write some code to make it work.
- [03:39:15] <jkridner>
just be sure it is DVI-D.
- [03:39:47] <jkridner>
ah, LG L1933.
- [03:39:51] <jkridner>
that's what I'm running.
- [03:40:12] <jkridner>
for a browser, you'll also need a USB Ethernet or WiFi.
- [03:40:29] <jkridner>
ds2: you getting outside of pdx while you are there?
- [03:40:33] <ds2>
thhe begle could be a dev plat for pandora
- [03:40:54] <jkridner>
ds2: I didn't know you were a pandora fan.
- [03:41:05] <ds2>
yeah, will do tthat
- [03:41:48] <jkridner>
driving up the gorge?
- [03:42:20] <ds2>
if pandora matches the desc, i would be very happy
- [03:42:52] <ds2>
thinking of looping up to sthelens
- [03:43:14] <jkridner>
hopefully you'll be able to get a good view.
- [03:45:06] <ds2>
i am hoping for a small eruption/plume
- [03:54:48] * jkridner|work1 (n=a0321898@nat/ti/x-4f7936984cc2172e) has joined #beagle
- [03:57:11] * Olipro__ (n=Olipro@uncyclopedia/Olipro) has joined #beagle
- [04:00:51] * jkridner|work (n=a0321898@nat/ti/x-51f5e2c496874b75) Quit (Remote closed the connection)
- [04:13:59] * Olipro_ (n=Olipro@uncyclopedia/Olipro) Quit (Read error: 110 (Connection timed out))
- [04:20:54] * shiv (n=root@59.160.172.220) has joined #beagle
- [04:36:07] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
- [04:36:14] * artyom1 (n=artyomt@198.78.191.90.dyn.estpak.ee) Quit (Read error: 110 (Connection timed out))
- [04:37:33] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
- [04:46:11] * Viral_Sachde (n=Viral_Sa@122.167.151.117) has joined #beagle
- [04:51:15] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
- [04:53:09] * JoeBorn (n=rootmeis@h-68-164-11-248.chcgilgm.dynamic.covad.net) Quit ("on the move")
- [04:56:43] * JoeyBorn (n=jborn@h-68-164-11-248.chcgilgm.dynamic.covad.net) has joined #beagle
- [05:02:49] * jess (n=jess@59.160.172.220) has joined #beagle
- [05:19:39] * _AV500_ is now known as AV500
- [05:22:30] * smackpotato (n=user@70.51.253.31) Quit (Read error: 110 (Connection timed out))
- [05:37:56] * JoeyBorn (n=jborn@h-68-164-11-248.chcgilgm.dynamic.covad.net) Quit (Read error: 60 (Operation timed out))
- [05:56:10] * cmonexaway is now known as cmonex
- [06:27:59] * lgentili (n=lgentili@host164.190-224-241.telecom.net.ar) has joined #beagle
- [06:49:22] * methril|gone is now known as methril
- [06:56:36] * _apkII (n=kows@85-18-136-98.fastres.net) has joined #beagle
- [06:57:22] * _apk (n=kows@85-18-136-98.fastres.net) Quit (Read error: 104 (Connection reset by peer))
- [06:59:51] * lgentili (n=lgentili@host164.190-224-241.telecom.net.ar) Quit (Remote closed the connection)
- [07:24:41] * khasim (n=a0393720@203.101.61.10) Quit (Remote closed the connection)
- [07:25:17] * khasim (n=a0393720@203.101.61.10) has joined #beagle
- [07:41:16] * tomeu (n=tomeu@89-24-41-113.i4g.tmcz.cz) has joined #beagle
- [08:06:12] * Viral_Sachde (n=Viral_Sa@122.167.151.117) Quit ("Leaving.")
- [08:07:29] * zedstar (n=john@82-44-200-69.cable.ubr08.haye.blueyonder.co.uk) has joined #beagle
- [08:07:36] * beagle|ibc (n=root@s55917625.adsl.wanadoo.nl) has joined #beagle
- [08:07:42] * beagle|ibc (n=root@s55917625.adsl.wanadoo.nl) has left #beagle
- [08:08:43] * koen|beagle (n=root@s55917625.adsl.wanadoo.nl) has joined #beagle
- [08:24:00] * artyomt (n=artyomt@194.126.108.2) has joined #beagle
- [08:35:10] * montamer (n=vijay@203.199.213.3) has joined #beagle
- [08:47:00] <ldesnogu>
koen: the libc link fixed abiword for me too :)
- [08:48:03] <koen>
ldesnogu: great!
- [08:48:40] <ldesnogu>
now I just have to load a few apps for the demo early october...
- [08:48:46] <koen>
it seems that having an empty 'libc.so' file during toolchain build has some weird effect
- [08:49:13] <ldesnogu>
you mean the cross toolchain build ?
- [08:50:20] <koen>
yes
- [08:50:42] <koen>
that was fixed weeks ago, but there are still some packages online wanting to link to libc.so
- [08:50:44] <ldesnogu>
it's not really empty in my build area, it contains an ld script
- [08:52:14] <koen>
heh, the company across the road is using OE
- [08:52:27] <koen>
they were wondering how close by I leave to drop by for some advice
- [08:52:36] <koen>
s/leave/live/
- [08:53:32] <ldesnogu>
make them pay you... at least some beers :)
- [08:54:10] * koen wonders how travelling expenses work for walking 200 meters
- [08:54:43] <chakie_work>
take a taxi
- [08:57:03] * cmonexaway (i=xy6091@pool-1352.adsl.interware.hu) has joined #beagle
- [08:58:15] * Olipro (n=Olipro@uncyclopedia/Olipro) has joined #beagle
- [08:58:17] * cmonex (i=xy6091@pool-7733.adsl.interware.hu) Quit (Nick collision from services.)
- [08:58:19] * cmonexaway is now known as cmonex
- [09:00:25] * RogerMonk (n=a0740758@nat/ti/x-21f69a7e2aca3637) has joined #beagle
- [09:01:00] <koen>
RogerMonk: good morning!
- [09:01:10] <RogerMonk>
gm sir!
- [09:02:06] <RogerMonk>
koen - great demos / beagle show at IBC - nice one sir!
- [09:03:50] <artyomt>
buy a ticket trough singapure :)
- [09:09:53] <koen>
Timing buffered disk reads: 28 MB in 3.10 seconds = 9.04 MB/sec
- [09:10:03] <koen>
sweet, my old 1G sd card does 9MB/s
- [09:13:08] * Olipro__ (n=Olipro@uncyclopedia/Olipro) Quit (Read error: 110 (Connection timed out))
- [09:23:40] * cmonexaway (i=xy6091@pool-2488.adsl.interware.hu) has joined #beagle
- [09:23:58] * cmonex (i=xy6091@pool-1352.adsl.interware.hu) Quit (Nick collision from services.)
- [09:24:00] * cmonexaway is now known as cmonex
- [09:24:34] * Olipro_ (n=Olipro@uncyclopedia/Olipro) has joined #beagle
- [09:39:21] * guillaum1 (n=gl@AMontsouris-153-1-59-77.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
- [09:39:43] * guillaum1 (n=gzba4143@AMontsouris-153-1-59-77.w90-2.abo.wanadoo.fr) has joined #beagle
- [09:41:57] * Olipro (n=Olipro@uncyclopedia/Olipro) Quit (No route to host)
- [09:42:48] * t_s_o (n=tso@135.84-49-129.nextgentel.com) has joined #beagle
- [10:06:50] <koen>
Timing buffered disk reads: 32 MB in 3.15 seconds = 10.17 MB/sec
- [10:07:01] <koen>
onenand seems faster than my 1G SD
- [10:14:14] <koen>
something is very wrong with IRQs in .27rc6
- [10:14:37] <koen>
and I think I have a likely culprit
- [10:28:12] <cmonex>
oh..
- [10:52:07] <Crofton|work>
http://mmbtools.crc.ca/content/view/33/58/
- [11:05:04] <Crofton|work>
hmmm, sakoman_ seems to be working on getting better omap3 support in mainline
- [11:22:02] <pbrook>
Are details of the Cortex-A8 r1px NEON bug available?
- [11:22:28] * jess (n=jess@59.160.172.220) has left #beagle
- [11:22:35] <pbrook>
The kenrel fix mentions a document that doesnt seem to be available online, and implkies that userspace code also needs fixing.
- [11:23:52] <pbrook>
However fixing userspace code (or in my case getting the compiler to fix it for you) is kinda tricky when you've no idea what bug you're trying to avoid.
- [11:23:55] * DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) has joined #beagle
- [11:23:55] <Crofton|work>
it is
- [11:24:05] <Crofton|work>
I don't have a link though
- [11:30:31] * dcordes (n=dcordes_@unaffiliated/dcordes) has joined #beagle
- [11:48:00] * artyomt (n=artyomt@194.126.108.2) Quit (Read error: 110 (Connection timed out))
- [11:48:42] * artyomt (n=artyomt@194.126.108.2) has joined #beagle
- [12:02:34] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
- [12:05:36] * jess (n=jess@59.160.172.220) has joined #beagle
- [12:08:59] <sakoman_>
good morning
- [12:10:29] * feig (n=ejf3@86.sub-75-195-32.myvzw.com) has joined #beagle
- [12:13:04] <Crofton|work>
gm
- [12:14:00] <Crofton|work>
any thoughts on why usb host in l-o is broke?
- [12:15:43] * felipec (i=c0647cdb@gateway/web/ajax/mibbit.com/x-4608b5b81a86a743) has joined #beagle
- [12:17:22] * BThompson (n=BThompso@nat/ti/x-f14615a0d70cfc7b) has joined #beagle
- [12:20:51] * Olipro_ is now known as Olipro
- [12:21:15] <jkridner>
good morning all.
- [12:22:14] <jkridner>
good question Crofton|work. I've been busy with higher functions required for presentations.
- [12:22:56] <Crofton|work>
yeah, I'm puzzling through verilog and the expansion conenctor
- [12:23:11] <Crofton|work>
and some guy is asking me about doing a few works of work in CA
- [12:23:38] <jkridner>
getting work is a good thing.
- [12:23:48] <jkridner>
hopefully it is with OMAP. :)
- [12:24:37] <Crofton|work>
heh
- [12:24:39] <Crofton|work>
dunno
- [12:25:45] <jkridner>
It isn't exactly what I'd want, and I didn't follow all of sakoman_'s good advice, but I have dd image of a bootable SD card with Angstrom at http://s3.amazonaws.com/beagle/angstrom-demo-20080916.img.gz
- [12:27:02] <felipec>
is anyone here compiling their own system manually?
- [12:27:08] <jkridner>
requires a 4GB card.
- [12:27:38] <jkridner>
felipec: I've done it in the past, but haven't gotten back there since my HDD went bad.
- [12:28:44] <felipec>
jkridner: ah, I just want a very basic system... like TI's base rootfs
- [12:29:01] <Crofton|work>
minimal-image?
- [12:29:10] <felipec>
using CS toolchain... so I guess I just need to compile busybox and glibc
- [12:29:18] <felipec>
Crofton|work: yeap
- [12:29:26] <Crofton|work>
bitbake minimal-image :)
- [12:30:10] <felipec>
Crofton|work: I already tried OE stuff, it's too complicated
- [12:30:37] <felipec>
I had to patch some ipkg stuff, and then the image never really booted
- [12:30:54] <Crofton|work>
weird
- [12:31:43] <felipec>
Crofton|work: yeah, and I feel I would spend more time properly finding the issue and fixing the stuff rather than doing it myself
- [12:32:49] <sakoman_>
Crofton|work: musb host mode on l-o works fine on Overo (with mans "fix something" patch)
- [12:32:56] <Crofton|work>
hmm
- [12:33:10] <Crofton|work>
I wonder if we have that in the linux-omap recipe
- [12:33:11] <sakoman_>
maybe some other patch you are applying is breaking it?
- [12:33:36] <sakoman_>
I have only 2 patches required on l-o
- [12:34:24] <sakoman_>
It was fun to see the guys at the kernel summit playing with their overo
- [12:34:38] <Crofton|work>
how many have you handed out?
- [12:34:47] <sakoman_>
Everyone got one
- [12:35:02] <sakoman_>
so around a hundred
- [12:36:16] <Crofton|work>
http://gitweb.openembedded.net/?p=org.openembedded.dev.git;a=blob;f=packages/linux/linux-omap_git.bb;h=f4a064f595a76139eef6556370c4382d76cad84e;hb=HEAD
- [12:38:36] <Crofton|work>
your patches are twl4030 rtc and overo_gpio_bt?
- [12:39:00] <Crofton|work>
anymore press releases on overo?
- [12:39:01] * shriram (i=7aa60de8@gateway/web/ajax/mibbit.com/x-5dce1f345268bf9a) has joined #beagle
- [12:39:44] <sakoman_>
Crofton: here is the diff between l-o upstream and my overo branch:
- [12:39:46] <sakoman_>
http://www.sakoman.net/cgi-bin/gitweb.cgi?p=org.openembedded.dev-omap3.git;a=blob;f=packages/linux/linux-omap3-git/overo/overo.patch;h=5e9b696b46298f7dd02f6b4532104468214e8b40;hb=651a87c8c38ba630138b6ebd56c5ba2c81e613da
- [12:39:58] * methril is now known as methril|gone
- [12:40:15] <sakoman_>
No, no press releases. Just a quiet marketing thing.
- [12:47:31] <sakoman_>
Crofton|work: here's the kernel recipe I use:
- [12:47:34] <sakoman_>
http://www.sakoman.net/cgi-bin/gitweb.cgi?p=org.openembedded.dev-omap3.git;a=blob;f=packages/linux/linux-omap3_git.bb;h=5d62dbc572bbeb5615775ae3d345cd593ede1cfd;hb=651a87c8c38ba630138b6ebd56c5ba2c81e613da
- [12:49:57] <jkridner>
Any chance that someone knows why 'cc1' is no longer included in 'task-native-sdk'?
- [12:50:16] <jkridner>
better yet, does anyone still have a copy of the binary image for cc1?
- [12:51:36] <dcordes>
jkridner, did you check the openembedded bugtracker?
- [12:52:15] <dcordes>
no longer implies it was there at some point in the same package?
- [12:52:38] <jkridner>
it was.
- [12:53:14] <dcordes>
koen, could probably help, otherwise I would make an OE bug
- [12:53:20] <Crofton|work>
this sounds familiar
- [12:53:22] <dcordes>
http://bugs.openembedded.net
- [12:53:28] <jkridner>
no bugs filed against it.
- [12:53:37] <Crofton|work>
hmm
- [12:53:40] <dcordes>
I can make one if you like
- [12:53:45] <jkridner>
k. I'll make an account and file a bug.
- [12:53:50] <dcordes>
or that
- [12:54:44] <jkridner>
I waste enough people's time on my struggles already. I'd always rather be taught how to fish.
- [12:56:24] <Crofton|work>
toolchain stuff is difficult fishing
- [12:59:28] * Openfree (n=df@218.82.126.34) has joined #beagle
- [13:01:34] <artyomt>
is there eny specific reason http://www.beagleboard.org/uploads/2.6_kernel_revb-v2.tar.gz doesnt contain USB Video class support included ?
- [13:03:10] * beagle|angstrom (n=angstrom@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
- [13:03:13] <keesj>
GeneralAntilles: I see you already created the next generation environement friendlier casings for the omap
- [13:03:43] <jkridner>
dcordes: bug filed.
- [13:04:19] <jkridner>
artyomt: because it is based on an older kernel and is only meant to verify the hardware.
- [13:04:57] <GeneralAntilles>
keesj, ;)
- [13:05:32] <keesj>
care to share the design :p
- [13:07:01] <GeneralAntilles>
First, find yourself a cardboard box. ;)
- [13:20:11] <koen>
artyomt: the specific reason is that the TI kernel is using the obsolete 2.6.22 as base
- [13:26:05] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit (Read error: 110 (Connection timed out))
- [13:32:04] * bgmarete (n=user@41.206.52.203) has joined #beagle
- [13:32:09] <bgmarete>
Hello
- [13:33:26] * dcordes (n=dcordes_@unaffiliated/dcordes) Quit ("leaving")
- [13:34:29] * trickie (n=trickie@hgaulton.xs4all.nl) has joined #beagle
- [13:37:21] * dcramer_ (n=davec@dcdsl.ebox.com) has joined #beagle
- [13:38:03] * dcramer (n=davec@dcdsl.ebox.com) Quit (Read error: 113 (No route to host))
- [13:42:40] * rsalveti (n=salveti@189.70.219.182) Quit (Remote closed the connection)
- [13:43:12] * jrmuizel (n=jrmuizel@mozca02.ca.mozilla.com) has joined #beagle
- [13:45:49] <jess>
koen: why beagleboard is tagged as OMAP34XX type in the kernel?
- [13:47:30] <koen>
I suspect because there weren't any 35xx tags j[D[3~yet
- [13:48:08] <koen>
and afaik 34xx vs 35xx is purely a marketing thing
- [13:48:27] <jess>
yeah.. that could be the reason..
- [13:48:41] <GeneralAntilles>
Aren't there a fewer interfaces on the OMAP35xx?
- [13:48:56] <pbrook>
Isn't 35xxx basically a larger pitch package?
- [13:49:06] * JuanG (n=Juan@nat/ti/x-0ef23fe30bbd1819) has joined #beagle
- [13:50:22] <koen>
GeneralAntilles: e.g. 3503 has no dsp
- [13:50:22] * JuanG (n=Juan@nat/ti/x-0ef23fe30bbd1819) has left #beagle
- [13:51:01] <GeneralAntilles>
OMAP3410 or OMAP3420 are the same deal, though.
- [13:54:03] <jkridner>
OMAP3503 has no GFX or DSP. 3515 has no DSP. 3525 has no SGX.
- [13:54:26] <jkridner>
all OMAP35x are missing some phone specific or proprietary features.
- [13:54:52] <jkridner>
for example. there is a serial camera bus that I believe is missing on OMAP35x.
- [13:54:58] <koen>
the nokia mark of evil you mean
- [13:55:00] * koen hides
- [13:55:15] * Viral_Sachde (n=Viral_Sa@122.167.151.117) has joined #beagle
- [13:55:22] <jkridner>
you don't hide very well koen.
- [13:55:49] <bgmarete>
Hello all. Just got a Beagle board and I am trying to assemble the peripherals needed. Pray, what might I need a mini-A to mini-B cable for?
- [13:56:01] <jkridner>
I believe that Nokia is happy to see the developer community increase with the availability of OMAP35x.
- [13:56:25] <koen>
I just wish nokia wasn't pushing bridge so much
- [13:56:32] <jkridner>
if you want to us the USB OTG as a host, you need to connect mini-A to the Beagle.
- [13:56:33] <pbrook>
bgmarete: You need mini-a if you want to use teh OTG port in host mode.
- [13:57:00] <jkridner>
mini-B could connect to a powered hub with a mini-B connection.
- [13:57:04] <pbrook>
A minia-mini-b will allow you to connect to a device with a mini-b port (e.g. a USB hub).
- [13:57:52] <jkridner>
koen: do you dislike bridge?
- [13:58:04] <jkridner>
or are you just being nice to me, since I push Link. :)
- [13:58:55] <jkridner>
btw, the userspace portions of Link will be licensed under BSD for the next (1.6) release (about 1 month), unless I misunderstood or something changes.
- [13:59:15] <bgmarete>
jkridner: Thanks.
- [13:59:24] <bgmarete>
pbrook: Thanks
- [13:59:54] <jkridner>
I really don't care which one is community accepted, I just hope that one is. I've made some use of both. Both could be massively trimmed down.
- [14:00:28] <jkridner>
there is some belief that Link is more general, but my view is that is only really true at the lower layers. (and both have too many layers)
- [14:00:44] <koen>
jkridner: from what I've heard at IBC is that bridge won't get much support from TI
- [14:00:59] * rsalveti (n=salveti@200.184.118.132) has joined #beagle
- [14:01:07] <koen>
if you go link you'll get more and better support
- [14:01:33] <koen>
although nobody was sure why the omapzoom guys went with bridge
- [14:01:39] <bgmarete>
I just connected my Nokia 3310c mini-? to standard cable to the board and PC respectively and it appears to have powered the board... Anyone tried this? What kind of cable is that? mini-B to standard-A or mini-A to standard-A?
- [14:02:31] <pbrook>
bgmarete: A is the host end of the cable, and B is the device end.
- [14:02:41] * jkridner admits to playing a role in the confusion.
- [14:03:12] <jkridner>
The Catalog FAE team is more familiar with Link. That is why you heard that view at IBC (you were hanging with Catalog guys like me).
- [14:03:38] * artyomt (n=artyomt@194.126.108.2) Quit (Read error: 110 (Connection timed out))
- [14:03:43] <jkridner>
the 2 solutions solve slightly different problems.
- [14:04:11] <jkridner>
eventually they are likely to merge.
- [14:04:27] <pbrook>
If you connect a mini-b connector to the beagle OTG connector it'll work in device mode (e.g. for doing USB networking to a PC) and power itself off the PC.
- [14:04:55] <adj_>
bgmarete: mini-A to standard-A cable should not even exist, so your cable must be mini-B to std-A
- [14:05:30] <jkridner>
if you want easy-to-use without understanding the details of the DSP, then bridge seems to be better, especially with the OMX IL implementation.
- [14:05:52] <felipec>
jkridner: the minifs on omapzoom doesn't have libdl... I think that's a bit too mini
- [14:06:02] <jkridner>
if you want easy-to-use talking to the DSP to your own algorithms, then Link is probably better for you.
- [14:06:38] <jkridner>
felipec: is libdl the omx *il*?
- [14:07:43] * cmonex (i=xy6091@pool-2488.adsl.interware.hu) Quit (":.there must be life beyond infinity.:")
- [14:08:31] <felipec>
jkridner: dynamic loader
- [14:08:51] <felipec>
jkridner: for dlopen, and such
- [14:09:28] <jkridner>
is that part of bridge? (doesn't match a part of bridge that I used about 4-5 years ago).
- [14:09:44] <felipec>
koen: if link has more support in TI it must be on the engineering level, I don't think the management is pushing it, except on davnci
- [14:10:17] <bgmarete>
Thanks a lot guys. I now understand completely :)
- [14:10:28] <felipec>
jkridner: no, everrying that loads libraries at run-time uses it: glib, gstreamer, etc.
- [14:11:50] <felipec>
and omx too
- [14:12:38] <jkridner>
ah. so a general file system issue.
- [14:12:48] * JoeyBorn (n=jborn@nmd.sbx02711.skokiil.wayport.net) has joined #beagle
- [14:13:04] <jkridner>
Be sure to note that on the omapzoom.org forums.
- [14:13:12] <koen>
felipec: from what I've heard is that the omapzoom team is pushing it aggressively and that the link people are still in davinci-land
- [14:13:13] <jkridner>
I have a Zoom, but I've only used it for demos.
- [14:13:39] <jkridner>
For Beagle, we've only provided a hardware test file system, not anything for reuse in applications.
- [14:14:09] <jkridner>
Beagle goals are to push software upstream and then consume it, without living on a development leaf.
- [14:14:29] <jkridner>
Angstrom is a nice upstream place for entire distro.
- [14:15:01] <jkridner>
my goals are to push useful components, like Link, into distros and l-o.
- [14:15:02] <felipec>
omapzoom forums are a joke
- [14:15:26] <koen>
opensource forums in general are a joke
- [14:15:50] <felipec>
mostly... there should be an omapzoom mailing list
- [14:17:41] <koen>
I still advocate using a mailinglist as forum backend
- [14:17:58] <koen>
so developers won't need to check a zillion topics
- [14:18:54] <bgmarete>
Hey: Sorry to come back :) Am I then to understand that A Standard-A to Mini-A Adapter connects the board in host mode to a hub that may not have a mini-B port? Or what use might such a cable be? :)
- [14:18:54] <felipec>
and the deliverables in omapzoom don't have release numbers, nor SCM
- [14:19:09] <felipec>
I don't see how the community is expected to collaborate
- [14:19:53] * Olipro_ (i=Olipro@uncyclopedia/Olipro) has joined #beagle
- [14:20:31] * Olipro (n=Olipro@uncyclopedia/Olipro) Quit (Nick collision from services.)
- [14:21:13] * Olipro_ is now known as Olipro
- [14:25:53] * feig1 (n=ejf3@141.sub-75-192-6.myvzw.com) has joined #beagle
- [14:26:13] <felipec>
crap, pthread is also missing
- [14:27:15] <jess>
regarding dsp on davinci, has any one tried the TI evaluation codecs successfully on a DM6446 ??
- [14:28:58] <bgmarete>
Hello: I am getting a mini-A to mini-B USB cable. Any reason why I would also need a Stanard-A to Mini-A adapter? Anyone?
- [14:29:43] <beagle|angstrom>
depends if you have a hub with mini-B on it.
- [14:30:55] <pbrook>
bgmarete: As adj_ said, mini-A to standard-A cables don't/shouldn't exist.
- [14:31:15] <bgmarete>
Ah, I see.
- [14:31:44] <bgmarete>
pbrook: Its on the Beagle Page :) http://beagleboard.org/hardware
- [14:32:42] <bgmarete>
But I understand. A is host and B is device. Got that.
- [14:32:43] <pbrook>
Ah, that's a mini-a plug to std-a socket.
- [14:33:40] <beagle|angstrom>
many more things have standard A cables and mini-A to standard-A-socket allows you to connect to any standard A cable.
- [14:33:54] <pbrook>
So then you can use any device with a standard plug on it.
- [14:34:07] <bgmarete>
So as beagle|angstrom said, I need it if I do not have a hub with a mini-B receptacle, yes?
- [14:34:29] <pbrook>
If your hub only has a standard A plug, yes.
- [14:35:04] <pbrook>
My hub has a std-B connector, so I'm using a mini-a to std-b cable.
- [14:35:12] <bgmarete>
pbrook: Thanks
- [14:35:29] <bgmarete>
beagle|amstrong: Thanks.
- [14:36:07] <beagle|angstrom>
pbrook: where did you find a mini-A to std-B cable? I have not seen one of those.
- [14:36:08] * JoeyBorn (n=jborn@nmd.sbx02711.skokiil.wayport.net) Quit (Read error: 113 (No route to host))
- [14:36:37] <pbrook>
beagle|angstrom: http://www.google.co.uk/products?q=mini-a+usb&btnG=Search+Products&show=dd
- [14:36:43] <suihkulokki>
koen: AFAIK it's WTBU rather than Nokia that's pushing for dspbridge.
- [14:37:26] <koen>
yeah
- [14:39:48] <koen>
hmmm
- [14:39:55] <koen>
broken usb on the evm as well
- [14:40:33] <felipec>
and librt too
- [14:47:16] * feig (n=ejf3@86.sub-75-195-32.myvzw.com) Quit (Read error: 110 (Connection timed out))
- [14:50:15] <koen>
vlad_: fennec 1.0a1pre is a lot faster than 0.7
- [14:50:36] <vlad_>
yeah, doesn't even have all the patches in
- [14:50:42] <koen>
vlad_: http://scap.linuxtogo.org/files/6a5c06e16bd1f7255f7a631c2cbfb148.png
- [14:50:43] <vlad_>
like, the repaint-every-100ms hack is still in
- [14:50:57] <koen>
I also RTFM'ed to see how to get to the tab switch bar
- [14:51:11] <vlad_>
haha
- [14:51:17] <vlad_>
hey, it's still an alpha :)
- [14:51:31] <koen>
the problem is that I'm not using a touchscreen
- [14:51:40] <vlad_>
yeah
- [14:51:45] <vlad_>
do you have the pref set to show the mouse?
- [14:51:47] <vlad_>
makes it a little easier
- [14:51:50] <koen>
yes
- [14:52:12] <koen>
vlad_: any idea why fennec is using ~70MB of ram?
- [14:52:39] <koen>
ram/swap was at 65/0 and now at 80/63
- [14:52:41] <vlad_>
nope, but i'd guess untuned caches and stuff
- [14:52:45] <vlad_>
nod
- [14:53:08] <koen>
I suspect the cpu hogging is due to the repainting 10 times a second
- [14:53:14] <vlad_>
yep
- [14:53:42] <koen>
so FF3.1 will probably be a lot faster than FF3 on beagle
- [14:53:59] <vlad_>
well, ff3 doesn't have the repainting thing in there
- [14:54:03] <Crofton|work>
koen, I'm going to try git with no patches
- [14:54:08] <vlad_>
but it's getting a lot of the gfx love
- [14:54:56] <koen>
1 GB of ram on the beagle would rock
- [14:55:06] <vlad_>
yeah, wish it had more ram
- [14:55:25] <koen>
webkit is using a lot less ram on beagle
- [14:55:38] <vlad_>
we've got a few perf tests we want to run that are really memory hungry, to avoid IO issues
- [14:55:50] * chivakker (n=chivakke@nat/ti/x-f8eb52467c20d135) Quit ("Leaving")
- [15:00:55] <beagle|angstrom>
purely hypothetical, how much more would you pay for a Beagle with 256MB of RAM?
- [15:02:16] <GeneralAntilles>
Is that a question to the consumers?
- [15:05:19] <beagle|angstrom>
yeah.
- [15:05:33] <beagle|angstrom>
what would it be worth?
- [15:06:19] <vlad_>
dunno, $40-$50 more?
- [15:06:27] <koen>
I'd pay at leat $20 more
- [15:06:40] <vlad_>
probably $150 or so more for a 1gb variant?
- [15:07:24] <koen>
for CMEM you want to have at least 256MB in total
- [15:07:42] <vlad_>
(CMEM?)
- [15:07:44] <pbrook>
You mean at most 256Mb?
- [15:07:56] <beagle|angstrom>
I'm hoping that greed can get it done. :)
- [15:08:02] <beagle|angstrom>
I really want one too.
- [15:08:25] <beagle|angstrom>
256MB is easy with same Beagle board, since Micron makes a device that is a drop-in replacement.
- [15:08:45] * cbrake (n=cbrake@oh-69-34-21-229.sta.embarqhsd.net) Quit (Read error: 113 (No route to host))
- [15:09:14] <pbrook>
That'd be nice. I didn't realise anyone was making larger 35xx packages.
- [15:09:41] <koen>
vlad_: cmem is a kernel modules that reserves a portion of the memory for the DSP
- [15:09:45] <beagle|angstrom>
pbrook: no change to the OMAP35x package, just the memory device.
- [15:09:56] <pbrook>
beagle|angstrom: Right.
- [15:10:00] <koen>
vlad_: reserving 40MB is fine on 256MB ram, but on 128 it hurts
- [15:10:08] <vlad_>
koen: nod
- [15:10:14] <vlad_>
yeah, 256mb would be a huge improvement
- [15:10:20] <vlad_>
anything bigger would just be icing
- [15:10:50] <koen>
with 1GB of ram one could run the complete demo image from a ramdisk
- [15:11:31] <koen>
beagle|angstrom: is micron already producing those? (256mb)
- [15:12:23] <beagle|angstrom>
yes
- [15:13:41] * koen gets a screwdriver to pry of the 128MB package
- [15:13:53] <pbrook>
Then again, you should probably be asking why you're allocating a 40Mb chunk.
- [15:13:58] <GeneralAntilles>
256MB would be great. My N800 has 128MB, doesn't feel like much of an upgrade. ;)
- [15:14:11] <pbrook>
Don't both the GPU and DSP effectively have MMUs?
- [15:14:23] <koen>
pbrook: they have
- [15:14:36] <koen>
they can see the same ram
- [15:15:00] <pbrook>
So the only thing you really need a big chunk of ram for is the framebuffer, right?
- [15:15:19] <vlad_>
grr.. my zoom is officially dead until I get back to the office next week where I have the power adapter
- [15:15:23] <suihkulokki>
Considering omap3 can boot from mmc, I'd replace the NAND with more RAM, and perhaps add a second SD slot.
- [15:15:35] <khasim>
I have connected pico projector to beagle DVI - it just rocks!!!
- [15:15:52] <khasim>
Big buck bunny is playing cool on DLP
- [15:16:10] <vlad_>
big buck bunny is terrifying
- [15:20:27] <beagle|angstrom>
I agree. It is not a happy movie. Graphics were nice though.
- [15:22:09] <beagle|angstrom>
koen: any hints to start sugar?
- [15:22:28] * beagle|angstrom is now known as jkridner|beagle
- [15:22:52] * dirk2 (n=dirk@F312c.f.strato-dslnet.de) has joined #beagle
- [15:23:22] <jkridner|beagle>
I'm finding some small tasks there, like bringing in squeek (sp?).
- [15:24:19] * Viral_Sachde (n=Viral_Sa@122.167.151.117) has left #beagle
- [15:25:45] * shiv (n=root@59.160.172.220) Quit ("Leaving.")
- [15:27:33] * jess (n=jess@59.160.172.220) has left #beagle
- [15:28:01] <koen>
jkridner|beagle: Xephyr :2 & DISPLAY=:2 sugar
- [15:33:29] <Crofton|work>
git does not make usb work, now to add some patches in
- [15:33:38] <Crofton|work>
after lunch that is, bbl
- [15:34:22] <jkridner|beagle>
hmmm... opkg install xserver-kdrive-xephyr isn't happy
- [15:34:46] <koen>
jkridner|beagle: -force-overwrite, it's on the X security file
- [15:34:46] <jkridner|beagle>
wants to try to overwrite my xserver-xorg
- [15:35:04] <koen>
one of the things on my todo list
- [15:35:04] <jkridner|beagle>
k. won't hurt my xorg?
- [15:35:08] <koen>
no
- [15:35:16] * koen has Xephyr installed as well
- [15:35:59] <jkridner|beagle>
I just don't want to go back to kdrive. xorg has worked so much nicer for me (no funny characters, scroll wheels working, etc.)
- [15:36:09] <koen>
indeed
- [15:37:42] <jkridner|beagle>
I get a python symbol lookup error
- [15:37:56] <jkridner|beagle>
libgtkinput.so doesn't have gtk_window_get_type
- [15:39:02] * t_s_o (n=tso@135.84-49-129.nextgentel.com) Quit ("Konversation terminated!")
- [15:42:31] * khasim (n=a0393720@203.101.61.10) Quit (Remote closed the connection)
- [15:42:42] <koen>
opkg -force-depends remove it
- [15:43:14] <jkridner|beagle>
remove libgtkinput.so?
- [15:43:22] * shriram (i=7aa60de8@gateway/web/ajax/mibbit.com/x-5dce1f345268bf9a) Quit ("http://www.mibbit.com ajax IRC Client")
- [15:44:14] <koen>
that would work as well
- [15:44:20] <kulve>
if Xorg is so much better why do everybody use kdrive still..
- [15:44:39] <kulve>
in these embedded devices..
- [15:44:40] <vlad_>
is it?
- [15:44:50] * bgat (n=bgat@adsl-75-23-86-252.dsl.peoril.sbcglobal.net) has joined #beagle
- [15:45:01] <vlad_>
I'm not sure what advantages it has other than the plugin/extension architecture
- [15:45:13] <koen>
kulve: kdrive doesn't need xorg.conf and uses less memory
- [15:45:16] <bgat>
anyone know if I can plug a RealView ICE JTAG directly into beagleboard?
- [15:45:20] <kulve>
well, we think so and so does jkridner (see above)
- [15:45:29] <kulve>
koen: Xorg doesn't need xorg.conf
- [15:45:38] <dirk2>
overo: http://tservice.net.ru/~s0mbre/blog/2008/09/17/
- [15:45:43] <jkridner|beagle>
bgat: need a TI to ARM JTAG pinout conversion
- [15:45:52] <bgat>
hi jason! :)
- [15:45:53] <koen>
kulve: it's just that recent kdrives lack a keymap and generally only work well on internet tablets from a certain scandinavian company
- [15:46:13] <bgat>
jkridner: is the pinout in the system reference manual? what about voltage levels?
- [15:46:15] <koen>
kulve: yes it does, I tried multiple times now to just remove xorg.conf, and it doesn't work
- [15:46:21] <jkridner|beagle>
it is.
- [15:46:24] <vlad_>
koen: erm? they've got their crazy Xomap thing which isn't even upstream
- [15:46:31] <jkridner|beagle>
voltage levels are 1.8V
- [15:46:32] <vlad_>
Xfbdev seems to work great
- [15:46:34] <koen>
vlad_: Xomap is kdrive
- [15:46:34] <bgat>
jkridner: looking for the pinout in the manual now. :)
- [15:46:41] <kulve>
vlad_: it's heavily kdrive based
- [15:46:45] <vlad_>
koen: yeah, it is
- [15:46:52] <vlad_>
but so are Xfbdev and others :)
- [15:47:12] <koen>
vlad_: and iirc Xomap is xfbdev with the manual dispc page flipping enabled
- [15:47:18] <vlad_>
yep
- [15:47:32] <vlad_>
though Xomap is more like a "random git snapshot of Xfbdev"
- [15:47:42] <vlad_>
I'm pushing them to just get the patch into xorg proper
- [15:48:06] <vlad_>
but I just didn't understand why kdrive-based servers generally only work well on their tablets :)
- [15:48:53] <koen>
vlad_: upstream breakage done by nokia X dude
- [15:49:10] <vlad_>
ugh, really? what happened?
- [15:49:11] <koen>
vlad_: they don't seem to care for other devices anymore
- [15:49:27] <koen>
but then again, kdrive was unmaintained before that
- [15:49:29] <vlad_>
that I can believe, btu they don't maintain kdrive, afaik
- [15:49:32] <vlad_>
nod
- [15:49:37] * koen filed many bugs that went to /dev/null
- [15:49:38] <vlad_>
keithp kinda abandoned it
- [15:49:52] <koen>
vlad_: mr stone now works for nokia
- [15:50:14] <vlad_>
ds?
- [15:51:06] <koen>
yes
- [15:51:39] <jkridner|beagle>
koen: I'm confused. libgtkinput.so seems to be required.
- [15:52:19] * jkridner|beagle isn't familiar with the RealView ICE JTAG hardware and doesn't know if it supports 1.8V.
- [15:52:20] <koen>
jkridner|beagle: opkg remove libgtkinput -force-depends, restart x
- [15:52:25] <bgat>
jkridner: is the 14p header the standard, "Multi-ICE" pinout? My RVI came with a "Multi-ICE Interface" which, suspiciously, has 14 pins on one end, and 20 on the other... :)
- [15:52:46] * bgmarete (n=user@41.206.52.203) Quit ("Leaving.")
- [15:52:51] <jkridner|beagle>
no, TI has a different standard JTAG pinout than ARM.
- [15:52:54] <koen>
jkridner|beagle: libgtkinput is a 3rd party gtk module to pop up the virtual keyboard
- [15:53:13] <bgat>
jkridner: cra^K^K^Kdarn.
- [15:54:03] * jkridner|beagle is restarting X
- [15:54:08] * jkridner|beagle (n=angstrom@c-76-31-18-64.hsd1.tx.comcast.net) Quit ("Powered by OE: www.openembedded.org")
- [15:54:24] * cbrake (n=cbrake@oh-69-34-21-229.sta.embarqhsd.net) has joined #beagle
- [15:55:36] <jkridner>
arg. can't get past the login screen now. power cycling the Beagle.
- [15:57:26] <jkridner>
rats. one of those steps broke my ability to log into an X session.
- [16:00:56] <jkridner>
"opkg install -force-reinstall -force-overwrite xserver-xorg" did *not* fix it.
- [16:02:57] <jkridner>
bgat: http://www.arm.com/support/faqdev/4511.html
- [16:05:42] <bgat>
jkridner: that circuit sure looks like it's going to feed 3.3v to the TI JTAG connector...
- [16:06:09] <jkridner>
I know spectrum digital makes some voltage adapters.
- [16:06:40] <jkridner>
http://emulators.spectrumdigital.com/utilities/
- [16:08:09] <bgat>
hmmm, I take that back... Skimming the RVI manual, it looks like maybe the JTAG interface in the ICE unit uses feedback from the JTAG header to set the voltage levels...
- [16:08:10] <jkridner>
http://emulators.spectrumdigital.com/utilities/adapters/cti20_lvrtck_a.pdf might have the right connection for your 20-pin.
- [16:08:41] <jkridner>
that has become common. just depends how low the drivers will go.
- [16:08:54] <jkridner>
er, receivers.
- [16:09:23] <bgat>
jkridner: ugh, that's not a circuit I'm going to build up from my junk box! :(
- [16:09:52] <jkridner>
probably not, but might be able to get something shipped quick without a lot of expense.
- [16:10:13] <jkridner>
I think that adapter board I just pointed to goes the wrong way.
- [16:10:36] <jkridner>
Spectrum Digital emulators use the TI 14-pin configuration.
- [16:10:47] <dirk2>
bgat: http://tincantools.com/product.php?productid=16144
- [16:10:53] <jkridner>
TI started to move to a 20-pin, but Beagle still uses the 14-pin.
- [16:11:28] <jkridner>
thanks dirk2.
- [16:12:05] <jkridner>
I think that is just a pin swivel, not level adjustment. hopefully the emulator supports the 1.8V. Flyswatter does, but I don't believe the Flyswatter supports RVDS.
- [16:12:37] <dirk2>
RVDS: Most probably not ;)
- [16:15:14] <bgat>
jkridner: the schematic you referred to in FAQ 4511 is do-able. I don't have any transistors to boost the TRST signal, I'll probably dispense with that for the moment, and use my finger on the reset button instead. :)
- [16:15:56] <bgat>
dirk2: the flyswatter adapter swivels from one 14-pin connector to another, neither of which is the 20-pin connector that I need. :)
- [16:16:25] <bgat>
dirk2: of course, if I buy the flyswatter....... but it would be a shame to not use the RVI I have right here, at least at the moment.
- [16:19:41] <jkridner>
kulve: I missed what it was I agreed with. context was a bit lost in the scroll-back buffer.
- [16:19:47] <dirk2>
bgat: k.
- [16:20:14] <jkridner>
was it a bridge vs. link issue?
- [16:22:28] * jkridner cannot seem to find the step that broke my ability to log into X.
- [16:22:42] * jkridner tries removing Xephyr
- [16:22:42] <koen>
jkridner: no clue in the logs?
- [16:22:59] <kulve>
jkridner: it doesn't matter. Just say "yes yes" ;)
- [16:23:16] <kulve>
jkridner: (it was about the xorg vs. kdrive)
- [16:23:42] * jkridner_ (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) has joined #beagle
- [16:23:44] <jkridner|work1>
hmmm. wonder why I was kicked.
- [16:24:38] <jkridner_>
ping
- [16:25:16] * ReMixx (n=me@lawn-128-61-26-92.lawn.gatech.edu) has joined #beagle
- [16:25:31] * jkridner|work (n=a0321898@nat/ti/x-2ea87819f8f8ded9) has joined #beagle
- [16:25:50] <jkridner_>
which logs? not in the Xorg.0.log
- [16:26:40] * Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) Quit (Remote closed the connection)
- [16:26:44] <ReMixx>
Hi can anyone help me get my serial connection working with my new beagleboard, please? :)
- [16:27:07] <jkridner|work>
I can give it a shot.
- [16:27:57] <ReMixx>
Ok great, thanks. So I have everything hooked up right I believe. When I connect via minicom or hyperterminal, I get a bunch of garbage instead of what I'm expecting to receive when I first connect to it.
- [16:28:12] <ReMixx>
?o? ??ww ?????m????
- [16:28:17] <ReMixx>
is what I'm getting in hyperterminal now
- [16:28:55] <jkridner|work>
have you set the proper baud?
- [16:29:07] <ReMixx>
115200 8N1 is what I have set
- [16:29:07] <jkridner|work>
115200n81?
- [16:29:34] <jkridner|work>
grounding?
- [16:29:39] <ReMixx>
what do you mean by grounding?
- [16:29:55] <jkridner|work>
how are you supplying power to the Beagle and to your computer?
- [16:29:57] <koen>
jkridner: check /etc/X11/Xsession.d for files referencing the input plugin
- [16:29:59] <jkridner|work>
are the out of the same outlet?
- [16:30:22] <ReMixx>
I'm supplying power via usb from my computer which is grounded through the outlet
- [16:30:27] <pbrook>
And check pin 5 is conneted properly.
- [16:31:28] * jkridner|work1 (n=a0321898@nat/ti/x-4f7936984cc2172e) Quit (Remote closed the connection)
- [16:32:32] * montamer (n=vijay@203.199.213.3) Quit ("Leaving")
- [16:32:35] <ReMixx>
oh no, he left :( Can anyone else assist me with my serial connection?
- [16:32:59] <jkridner|work>
I'm still here.
- [16:33:16] <jkridner|work>
that was an alias of me that got abandoned by a bad net connection.
- [16:33:40] <jkridner|work>
I'm here trying to solve my input issue.
- [16:33:44] <ReMixx>
ohhh I'm sorry
- [16:34:50] * bgat (n=bgat@adsl-75-23-86-252.dsl.peoril.sbcglobal.net) has left #beagle
- [16:39:04] * jkridner (n=jason@c-76-31-18-64.hsd1.tx.comcast.net) Quit (Connection timed out)
- [16:39:13] * jkridner_ is now known as jkridner
- [16:42:06] <jkridner|work>
ReMixx: you checking your pin 5 connection?
- [16:42:14] <ReMixx>
pin 5?
- [16:42:38] <jkridner|work>
per pbrook recommendation above.
- [16:42:50] <ReMixx>
Oh i didn't know that was directed at me
- [16:43:01] <jkridner|work>
yes, pin 5 is the ground pin.
- [16:43:18] <jkridner|work>
you need 3 signals connected between your computer's serial port and the Beagle.
- [16:43:43] <jkridner|work>
Pin 2 on your computer should connect to pin 3 on the Beagle. 3 to 2 and 5 to 5.
- [16:43:44] <ReMixx>
so pin 5 on the beagle's idc10 connection?
- [16:44:01] <jkridner|work>
yes, and on your computer's DB9.
- [16:44:10] <ReMixx>
i'm using a usb->serial connection
- [16:44:46] <jkridner|work>
then the DB9 on our converter.
- [16:44:51] <ReMixx>
right
- [16:45:58] <ReMixx>
it looks to be all connect right
- [16:46:13] <ReMixx>
so if it wasn't grounded, that would cause a problem similar to mine?
- [16:46:26] * khasim (n=a0393720@203.101.61.7) has joined #beagle
- [16:46:31] * bef0rd (n=befr0d@190.68.13.134) has joined #beagle
- [16:46:39] <jkridner|work>
yes, if the ground connection was bad it could cause a problem similar.
- [16:46:44] <ReMixx>
ok
- [16:47:30] <ReMixx>
on the IDC10-DB9, the red part of the ribbon should be on the 1-2 pin side of the IDC10 connector on the Beagle, correct?
- [16:52:48] <ReMixx>
I switched over to power directly from an outlet into the Beagle and into my laptop, the garbage is slightly different, but still garbage
- [16:52:58] <ReMixx>
?m ? ??} o} ?? ? ? { ?o?o?? w{??}? o ? ] ? ? ??
- [16:54:53] <jkridner|work>
ReMixx: correct on the red stripe being on the side of pin 1.
- [16:55:11] <jkridner|work>
powering directly from the laptop would provide best grounding (same ground source).
- [16:55:39] <jkridner|work>
how did you verify the pinout?
- [16:56:26] <jkridner|work>
one test that has found some people's problems is to short pins 2 and 3 on the IDC header (while not connected to the Beagle) and to simply see that it loops back text.
- [16:57:11] <ReMixx>
i verified the pinout based on some websites showing pictures of it, not sure which websites
- [16:57:36] <ReMixx>
so I can short those pins with just like a paperclip I assume?
- [16:59:41] <jkridner|work>
paperclip should conduct sufficiently, as long as it isn't coated.
- [17:00:08] <jkridner|work>
correct pinout is at http://www.pccables.com/07120.htm
- [17:00:35] * bazbell (n=a0192809@nat/ti/x-afc8532efd931b4a) Quit (Remote closed the connection)
- [17:00:39] <jkridner|work>
but, that *must* be used in conjunction with a null-modem cable, not a straight-through cable.
- [17:00:40] <ReMixx>
i think i've been to that website before, it's the same pinout i used
- [17:00:48] <ReMixx>
i have a null modem cable as well
- [17:01:59] * gregoiregentil (n=zonbu@adsl-71-135-103-46.dsl.pltn13.pacbell.net) has joined #beagle
- [17:02:12] * gregoiregentil (n=zonbu@adsl-71-135-103-46.dsl.pltn13.pacbell.net) has left #beagle
- [17:03:04] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
- [17:03:27] <jkridner|work>
let me know the result of the shorting.
- [17:03:46] <jkridner|work>
you could also take a shot at using tera term, but it isn't likely to work with your USB-to-serial adapter.
- [17:04:07] <ReMixx>
forgive me for being new with the terminal, should I be able to type plain text characters in it and with the short it should echo back what I typed?
- [17:04:18] <ReMixx>
I tried teraterm but it wouldn't go to COM27 for my adapter
- [17:04:23] <ReMixx>
only COM1-4
- [17:05:04] <jkridner|work>
right, I saw the same issue with teraterm.
- [17:05:25] <jkridner|work>
yes, you should be able to type plain text characters and they should echo back.
- [17:05:36] <ReMixx>
ok i'm not able to type plain text characters in hypterterminal
- [17:05:47] <ReMixx>
ww{}w}w?w
- [17:05:52] <jkridner|work>
that rules Beagle out.
- [17:05:55] <ReMixx>
is what i got when trying to type 'asdf'
- [17:06:00] <ReMixx>
makes sense
- [17:06:20] <jkridner|work>
what make/model is your usb-to-serial adapter?
- [17:06:28] <ReMixx>
airlink 101
- [17:06:32] <ReMixx>
let me find the website for you
- [17:06:57] <ReMixx>
http://www.airlink101.com/products/ac-usbs.php
- [17:09:20] * khasim (n=a0393720@203.101.61.7) Quit (Remote closed the connection)
- [17:13:31] <jkridner|work>
I know some terminal programs have had problems with USB-to-serial adapters.
- [17:15:47] <vlad_>
ReMixx: could try a win32 kermit
- [17:15:50] <jkridner|work>
I don't really have a next guess, but it is good to have the adapter recorded, in case someone else experiences something similar.
- [17:16:08] <jkridner|work>
trying kermit might be helpful, if it supports COM27.
- [17:16:45] <jkridner|work>
must be over 10 years since I've used kermit!
- [17:16:53] <vlad_>
heh
- [17:17:04] <ReMixx>
i tried kermit 95 earlier, couldn't get it to even display the garbage haha
- [17:17:22] <vlad_>
I end up going back to it whenever I need a serial connection, since it's one of the few that feels like I have control over what it's doing under the hood
- [17:17:25] <vlad_>
ReMixx: http://kermit.wwarthen.com/ ?
- [17:17:38] * valhalla (n=valhalla@81-174-22-89.dynamic.ngi.it) has joined #beagle
- [17:18:18] <ReMixx>
I will give it a try, thanks vlad
- [17:20:18] <ReMixx>
vlad_ it only goes from com1-4, my usb-serial adapter is on com27
- [17:20:24] <vlad_>
weak
- [17:20:46] <vlad_>
with the garbage that ht was giving you, all the comm settings were correct, right?
- [17:20:52] <vlad_>
115200/8n1, no flow control?
- [17:21:27] <ReMixx>
correct
- [17:21:39] <ReMixx>
for both hyperterminal and minicom
- [17:22:50] <adj_>
can't the com port number be changed?
- [17:23:08] * guillaum1 (n=gzba4143@AMontsouris-153-1-59-77.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
- [17:23:41] * guillaum1 (n=gl@AMontsouris-153-1-59-77.w90-2.abo.wanadoo.fr) has joined #beagle
- [17:24:09] <ReMixx>
can it?
- [17:24:50] * jkridner|work steps away from the computer for lunch.
- [17:26:11] <adj_>
i'd say it depends on the adapter and the driver, at least i can change it with my ft2xx or prolic based serial adapter
- [17:26:19] * artyomt (n=Artyom@198.78.191.90.dyn.estpak.ee) has joined #beagle
- [17:26:50] <ReMixx>
i am able to change it through device manager
- [17:26:53] <jkridner|work>
well, right after noticing: root@beagleboard:~# DISPLAY=:0 gpe-confd
- [17:26:53] <jkridner|work>
X connection to :0.0 broken (explicit kill or server shutdown).
- [17:27:01] <ReMixx>
and it was set to a different baud as well through the device's properties
- [17:29:21] <adj_>
ReMixx: yeah, but a think it really doesn't matter what the device damager says about the speed of the port. For my prolific it seems to be always 9600 baud even if i'm really using the port at some totally different speed
- [17:29:30] <ReMixx>
oh ok
- [17:29:41] <ReMixx>
i'm going to try kermit now that i'm on com2
- [17:31:44] <ReMixx>
kermit gives me garbage too
- [17:31:55] <ReMixx>
just holding down backspace gives me a string of random garbage in kermit
- [17:32:51] * artyomt (n=Artyom@198.78.191.90.dyn.estpak.ee) Quit ("Leaving.")
- [17:43:13] * artyomt (n=Artyom@198.78.191.90.dyn.estpak.ee) has joined #beagle
- [17:46:19] <felipec>
ReMixx: do you have a proper rs-232 cable?
- [17:46:35] * dcordes (n=dcordes_@unaffiliated/dcordes) has joined #beagle
- [17:47:23] <felipec>
ah, you already answered that
- [17:47:40] <ReMixx>
yup
- [17:47:53] <ReMixx>
I'm not sure what it is, I guess I'll have to try my setup on a different computer
- [17:47:58] <ReMixx>
it could just be the usb-serial adapter
- [17:48:13] <ReMixx>
thanks guys, I need to run
- [17:48:24] * ReMixx (n=me@lawn-128-61-26-92.lawn.gatech.edu) Quit ()
- [17:49:03] * bazbell (n=a0192809@nat/ti/x-58349d18e12caefb) has joined #beagle
- [17:49:07] <koen>
artyomt: v4l and uvc are both supported in the angstrom kernel
- [18:01:10] <Crofton|work>
ok, building git + fix-instal and fix-something patches
- [18:01:23] <Crofton|work>
which is what sakoman_ claims works on overo
- [18:03:01] <dirk2>
news of the day: Maemo 5 with OMAP3: http://www.internettablettalk.com/2008/09/17/dr-ari-jaaksi-on-maemo-5/
- [18:03:17] <dirk2>
and in case you missed it above: overo: http://tservice.net.ru/~s0mbre/blog/2008/09/17/
- [18:03:31] * Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) has joined #beagle
- [18:16:11] <jkridner|work>
koen: enlightenment: symbol lookup error: enlightenment: undefined symbol: edje_scale_set
- [18:18:00] <jkridner|work>
nice news dirk2
- [18:19:03] <Crofton|work>
good to see a photo of overo
- [18:25:51] * guillaum1 (n=gl@AMontsouris-153-1-59-77.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
- [18:26:17] * jrmuizel (n=jrmuizel@mozca02.ca.mozilla.com) Quit (Remote closed the connection)
- [18:26:19] <vlad_>
overo looks cool, hopefully there will be a daughterboard with on-board ethernet
- [18:26:29] * jrmuizel (n=jrmuizel@mozca02.ca.mozilla.com) has joined #beagle
- [18:26:43] <jkridner>
does have wifi though.
- [18:26:51] <vlad_>
yeah
- [18:26:53] * jkridner agrees that isn't a complete substitute.
- [18:27:14] <vlad_>
we've just been looking for a good solution for automated testing machines
- [18:27:27] <vlad_>
right now i'm thinking to suggest beagle + usb ethernet, along with some kind of remote reset
- [18:27:43] <jkridner>
you could make a new base board without having to deal with all the tight pitch stuff.
- [18:27:56] <jkridner>
and put 50 Overo's down on that new base board. :)
- [18:28:06] <vlad_>
hah yes
- [18:28:10] <koen>
hmmm
- [18:28:11] <vlad_>
that's an interesting idea
- [18:28:15] <koen>
distcc :)
- [18:28:19] <jkridner>
:)
- [18:28:21] <vlad_>
maybe 10 or so instead of 50, but yes :)
- [18:28:54] * Crofton is wondering if you could multiple overos and pass data between them
- [18:28:59] <vlad_>
i'm also under the impression that microSD tends to be slower than regular-sized SD -- is that necessarily true?
- [18:29:22] <koen>
isn't it just a different connector?
- [18:29:28] <thomasg>
well, it depends
- [18:29:32] * felipec (i=c0647cdb@gateway/web/ajax/mibbit.com/x-4608b5b81a86a743) Quit ("http://www.mibbit.com ajax IRC Client")
- [18:29:41] <thomasg>
normally the "standard" SDs use multiple chips at once
- [18:29:46] <thomasg>
and they usually have the better controller
- [18:30:41] <thomasg>
so I think that you can always assume the microSDs to be slower, even if it woulnd't be necessary in every situation
- [18:30:57] <vlad_>
ah, that makes sense
- [18:31:23] <vlad_>
it's been my experience that that's true but I wasn't sure if that was just because I was buying wrong speeds or something
- [18:31:32] <geist>
nah, I call BS on that
- [18:31:37] <jkridner>
hopefully we'll get some 256MB versions of Overo.
- [18:31:40] <geist>
my experience has been that microsd is the same thing
- [18:31:55] * guillaum1 (n=Guillaum@AMontsouris-153-1-59-77.w90-2.abo.wanadoo.fr) has joined #beagle
- [18:31:56] <geist>
in fact, the fastest SD card i've ever worked with is a microsd
- [18:32:46] <geist>
it stands to reason that sd card manufacturers dont waste time making seperate chips + controllers for micro and regular, and that a regular sd is the same guts
- [18:32:51] <thomasg>
geist: well, the easiest example would be to find a microSD that is as fast as one of the fastest SDs. I'm pretty sure you won't.
- [18:32:56] * artyomt (n=Artyom@198.78.191.90.dyn.estpak.ee) has left #beagle
- [18:33:02] * artyomt (n=Artyom@198.78.191.90.dyn.estpak.ee) has joined #beagle
- [18:33:08] <geist>
I still call bs on that
- [18:33:23] <geist>
microsd isn't that small, all in all
- [18:33:27] <vlad_>
either way!
- [18:33:32] <geist>
if you get the raw chips they easily fit in that package
- [18:33:38] <vlad_>
something would be helpful would be a little registry of fast SD/microSD
- [18:33:46] <vlad_>
so that people know what cards are good ones to get
- [18:33:48] <geist>
got one right here. lemme se...
- [18:33:53] <jkridner>
better to see what the fastest card can be found.
- [18:34:03] <geist>
sandisk 8GB, microsd
- [18:34:08] <geist>
this guy is a quick little puppy
- [18:34:19] * TAK2004 (n=Administ@dslb-088-074-054-137.pools.arcor-ip.net) has joined #beagle
- [18:34:28] <vlad_>
yeah?
- [18:34:36] <vlad_>
SDSDQR-8192-E11M ?
- [18:34:48] <jkridner>
we should probably define a test program on the Beagle to characterize the cards and register the data on a wiki.
- [18:34:49] <vlad_>
http://www.amazon.com/Sandisk-MicroSDHC-MobileMate-SDSDQR-8192-E11M-Package/dp/B000WH6H1M/ to be exact
- [18:34:52] <artyomt>
koen: 9/11 version doesnt seems to have it
- [18:35:19] <geist>
the other key is you need a good reader
- [18:35:29] <geist>
but usually the card comes with a fast enough reader to keep up
- [18:35:52] <geist>
i've clocked this particular microsd in the high 20MB/sec read range, writes are about 20
- [18:36:15] <geist>
no doubt you can get stuff faster, but it seems to stand out from all the other microsd cards i've seen
- [18:36:19] <koen>
root@beagleboard:~# zcat /proc/config.gz | grep -e USB_VIDEO -e V4L2
- [18:36:19] <koen>
CONFIG_VIDEO_V4L2_COMMON=m
- [18:36:19] <koen>
CONFIG_VIDEO_V4L2=m
- [18:36:19] <koen>
CONFIG_USB_VIDEO_CLASS=m
- [18:37:08] <Crofton|work>
weird, adding mru's fix musb patch and I hang after "booting the kernel"
- [18:37:18] <geist>
of course the mmc/sd controller on the beagle itself running linux isn't good enough to really saturate that, and SD is just 4 bits wide, so that's about as fast as you can ever get
- [18:37:28] <geist>
you need 8 bit mmc to get much faster
- [18:37:43] <geist>
25MB/sec or so is theoretical bandwidth speed on SD
- [18:40:15] * dirk2 (n=dirk@F312c.f.strato-dslnet.de) has left #beagle
- [18:40:40] <jkridner>
delays in card can still be compared, even if Beagle interface is the main performance bottleneck.
- [18:41:46] <mru>
hi all
- [18:42:29] <geist>
right, guess my point is that I have a microsd right here that runs at max theoretical speed of SD interface
- [18:42:38] <geist>
thus you can't really find any faster, unti you go do 8 bit mmc
- [18:42:46] <geist>
which is pretty much impossible to find, as far as I can tell
- [18:43:03] <mru>
got my rev C today
- [18:44:16] <jkridner>
rev C proto. not a true rev C.
- [18:44:24] <geist>
ah, reran the test: got 22MB/sec read, 15MB/sec write
- [18:44:25] <Crofton|work>
rev c-1
- [18:44:40] <geist>
that's more like it. I was calling my own BS as I said high 20s, since sd can't go that fast
- [18:44:40] <jkridner>
likely to change prior to release. issues already found.
- [18:45:01] <Crofton|work>
c-10
- [18:45:04] <geist>
jkridner: is it a new ES revision?
- [18:45:09] <mru>
yes, proto
- [18:45:22] <jkridner>
yes. It uses an ES3.0
- [18:45:43] <geist>
jkridner: can you ask internally how ES revisions on 3530s line up with ES on 3430?
- [18:45:55] <jkridner>
we are nowhere close to shipping. mru got an early proto thanks to his work on FFmpeg.
- [18:46:01] <geist>
seeing as 3430 is just entering ES3 as well, it stands to reason they are based on similar schedule
- [18:46:06] <jkridner>
they line up the same.
- [18:46:21] <geist>
so based on that it stands to reason that the errata should be similar?
- [18:46:30] <mru>
I wouldn't be surprised if they're made from the same masks
- [18:46:41] * bef0rd (n=befr0d@190.68.13.134) Quit (Remote closed the connection)
- [18:46:46] <geist>
if so, why does the 3530 errata not include nearly as much detail as 3430?
- [18:46:57] <geist>
ie, the msusb dma problem, the mmc dma problem seem to not be mentioned
- [18:47:00] <mru>
same for the trm to some extent
- [18:50:01] <geist>
also, I think there may be a bug in the 3530 TRM. the hawkeye revision that the cpu reports seems to line up with what 3430s report (same chip id, seem to be using the same ES revision numbers), which is out of sync with the 3530 manual
- [18:50:31] <geist>
ho9pefull that's the case because then it's meaningful, at least
- [18:50:48] <geist>
if it's not the case, then it's busted, because the 3530 reports itself as a 3430, and then reports the wrong revision
- [18:51:42] <koen>
jkridner: http://www.angstrom-distribution.org/demo/beagleboard/Angstrom-Beagleboard-demo-image-glibc-ipk-2008.1-test-20080917-beagleboard.rootfs.tar.bz2 has the bugfix for parallel installable Xephyr and Xorg
- [18:52:21] * ciskje (n=francesc@host76-84-dynamic.16-87-r.retail.telecomitalia.it) has joined #beagle
- [18:55:27] * lukketto (n=luca@host211-134-dynamic.57-82-r.retail.telecomitalia.it) has joined #beagle
- [18:56:11] <koen>
artyomt: the kernel sources are mentioned in the kernel package
- [18:56:22] <koen>
artyomt: and it's built from OE
- [18:56:25] * Beagle4 (n=Beagle4@78.52.200.252) has joined #beagle
- [18:56:33] <koen>
artyomt: no need for PMs to ask such questions
- [18:56:38] <jkridner>
koen: does it have a working gcc?
- [18:56:46] * TAKI (n=Administ@dslb-088-074-033-148.pools.arcor-ip.net) has joined #beagle
- [18:56:54] <koen>
jkridner: as long as you don't need limits.h, yes
- [18:57:29] <koen>
jkridner: please post your error messages from gcc to the oe-devel list, Khem Raj is waiting for them to fix the problem :)
- [18:58:04] <jkridner>
k. maybe I didn't see the mail. I only posted the bug and am not aware of the mail list.
- [18:58:20] * Beagle4 (n=Beagle4@78.52.200.252) Quit (Client Quit)
- [18:58:22] * Beagle1 (n=Beagle1@78.52.200.252) has joined #beagle
- [18:58:24] <Crofton|work>
I can ping khem with the bug info
- [18:58:29] * valhalla (n=valhalla@81-174-22-89.dynamic.ngi.it) Quit ("Leaving")
- [18:58:30] * Beagle0 (n=Beagle0@78.52.200.252) has joined #beagle
- [18:58:59] * Beagle3 (n=Beagle3@78.52.200.252) has joined #beagle
- [19:06:15] * bazbell (n=a0192809@nat/ti/x-58349d18e12caefb) Quit ("Leaving.")
- [19:06:23] * hli (i=chaton@vig91-2-82-232-97-149.fbx.proxad.net) has joined #beagle
- [19:08:36] * artyomt1 (n=Artyom@198.78.191.90.dyn.estpak.ee) has joined #beagle
- [19:11:59] <denix>
koen: I thought limits.h was fixed
- [19:12:57] <koen>
denix: for cross and sdk, but not for gcc (target)
- [19:12:59] <Crofton|work>
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=260284613485
- [19:13:09] <koen>
denix: but I haven't tried since last week
- [19:13:20] * Crofton|work is also known is k4ep
- [19:14:10] <koen>
Crofton|work: I visited the tek booth at IBC
- [19:14:18] <koen>
some mighty fine toys there
- [19:14:27] * TAK2004 (n=Administ@dslb-088-074-054-137.pools.arcor-ip.net) Quit (Read error: 110 (Connection timed out))
- [19:14:36] <denix>
koen: I see, I only tested sdk and cross, not sure about gcc target
- [19:14:39] <Crofton|work>
I'm sure they have some really neat stuff
- [19:14:49] <Crofton|work>
that I can't afford :)
- [19:14:55] <Crofton|work>
but would really like to have
- [19:15:18] * koen notices that the new EOS5D2 can record 1080p h264
- [19:15:36] <koen>
funny specs for a *photo*camera
- [19:15:40] <denix>
Crofton|work: good luck winning it :)
- [19:16:15] <Crofton|work>
looks like I won it, now I need to get it, and have it work :)
- [19:17:29] * fer_luck (n=fer_luck@189.73.253.198) has joined #beagle
- [19:17:39] <denix>
oh, didn't see it has ended... anyway, good luck with that too :)
- [19:17:52] * fer_luck (n=fer_luck@189.73.253.198) Quit (Client Quit)
- [19:18:05] * fer_luck (n=fer_luck@189.73.253.198) has joined #beagle
- [19:22:43] * Olipro_ (i=Olipro@uncyclopedia/Olipro) has joined #beagle
- [19:24:41] * ReMixx (n=ReMixx@r58h135.res.gatech.edu) has joined #beagle
- [19:26:26] * Olipro__ (n=Olipro@uncyclopedia/Olipro) has joined #beagle
- [19:27:04] * Olipro (i=Olipro@uncyclopedia/Olipro) Quit (Nick collision from services.)
- [19:27:06] * Olipro__ is now known as Olipro
- [19:27:11] * artyomt (n=Artyom@198.78.191.90.dyn.estpak.ee) Quit (Read error: 110 (Connection timed out))
- [19:42:17] * lukketto (n=luca@host211-134-dynamic.57-82-r.retail.telecomitalia.it) Quit ("Leaving.")
- [19:42:47] * Olipro_ (i=Olipro@uncyclopedia/Olipro) Quit (Connection timed out)
- [19:51:17] * bazbell (n=a0192809@nat/ti/x-3a5e6a976adeaee0) has joined #beagle
- [19:51:55] * flo_lap (n=fuchs@dslb-088-073-027-051.pools.arcor-ip.net) has joined #beagle
- [19:56:29] <jkridner|work>
any nice embedded solution for something like lvm?
- [19:57:11] <mru>
what's wrong with lvm?
- [20:01:24] * Beagle3 (n=Beagle3@78.52.200.252) Quit ()
- [20:01:40] * tomeu (n=tomeu@89-24-41-113.i4g.tmcz.cz) Quit ("Ex-Chat")
- [20:06:51] * dcordes (n=dcordes_@unaffiliated/dcordes) Quit ("leaving")
- [20:07:00] <jkridner|work>
I don't know. sounds a bit heavy for an embedded system.
- [20:07:04] <jkridner|work>
shall I just try lvm?
- [20:07:41] <jkridner|work>
I want to try to make a default image for a 512MB SD card and allow it to be extended for larger SD cards.
- [20:08:36] <vlad_>
jkridner|work: shouldn't need to use lvm for that
- [20:08:49] <mru>
I was about to say the same
- [20:08:54] <vlad_>
if they partition and dd the image, or just dd an image with a partition map, they can just grow the partition
- [20:09:07] <mru>
a simple resize2fs or whatever it's called should do
- [20:11:48] <koen>
or gparted if that's your thing
- [20:12:03] <koen>
or is parted the console thing?
- [20:12:19] <mru>
never got on good terms with that one
- [20:12:28] <mru>
fdisk has always done the job for me
- [20:12:39] <mru>
it supports bsd disklabels too
- [20:12:58] <vlad_>
yeah, i'm a fdisk fan
- [20:13:53] * koen hugs fdisk
- [20:14:43] <koen>
I've used parted once and only to see if it cross-compiled cleanly :)
- [20:14:54] <mru>
I bet it didn't
- [20:16:23] <koen>
nope
- [20:16:38] <koen>
broken autofoo, missing #include lines, etc
- [20:18:11] <mru>
"broken" is redundant beside autofoo
- [20:18:21] <artyomt1>
http://beagleboard.org/irclogs requires login but http://www.beagleboard.org/irclogs works fine
- [20:19:02] <mru>
anyhow, the 3530 ES3.0 has a revision number of 3
- [20:19:41] <mru>
and the ARM id register says 411fc083
- [20:20:09] <koen>
mru: how much 'performance' do the dmbs eat up?
- [20:20:20] <mru>
haven't got that far
- [20:20:39] <jkridner|work>
k. I'll look to use resize.
- [20:20:39] <mru>
I'm trying to work out precisely what I'm dealing with first
- [20:21:09] <mru>
jkridner|work: which cortex-a8 revision would you say is in ES3.0?
- [20:21:28] <jkridner|work>
artyomt1: the login page is a bit like a 404 on beagleboard.org.
- [20:21:41] <jkridner|work>
site assumes you want to create the page if it doesn't exist. I need to fix that some day.
- [20:22:18] <jkridner|work>
mru: I already forgot. can look it up for you.
- [20:22:49] <artyomt1>
ok
- [20:22:51] <mru>
sprz278b says it should be r2p1
- [20:22:55] <mru>
the silicon disagrees
- [20:23:00] <artyomt1>
just my 2 cents :)
- [20:23:10] <mru>
by extrapolating documented values, it would be an r1p3
- [20:23:19] <mru>
but such a revision isn't mentioned in any arm docs
- [20:23:27] <koen>
vlad_: you did a bit of pixman hacking, right?
- [20:23:36] <jkridner|work>
I didn't think r1p3 existed.
- [20:23:47] <vlad_>
koen: a bit, jrmuizel did more
- [20:23:50] <vlad_>
koen: why?
- [20:24:14] <jkridner|work>
in some other junk right now. will look up later.
- [20:24:37] <koen>
vlad_: mru has made a fast yuv422->yuv420 functions, how hard would it be to put that in pixman so Xorg can use it?
- [20:24:40] <koen>
vlad_: http://git.mansr.com/?p=omapfbplay;a=blob;f=omapfbplay.c;h=0d44fbb45757e5fb51b44dc1d476311cc6875fa8;hb=d8ec4c5d3d55557c321170c94dd5cb2dea22a227
- [20:24:55] <vlad_>
mmm
- [20:24:56] <koen>
" a .. functions", my grammar is getting worse and worse
- [20:25:12] <mru>
first it would need to be made to work with non-mod16 dimensions
- [20:25:15] <vlad_>
I wonder if the yuv stuff is in the core yet
- [20:25:26] <vlad_>
mru: that can be a prereq for taking that path
- [20:25:31] <vlad_>
koen: but not hard
- [20:25:55] <mru>
and I should see if it can be optimised further
- [20:26:08] <mru>
maybe a few more % can be squeezed out of it
- [20:26:37] <vlad_>
koen: you really want to talk with ssp
- [20:26:39] <koen>
how much work would a neon yuv420->rgb be ?
- [20:26:54] <vlad_>
he's not around atm, but he's the one that would be able to tell you
- [20:27:14] <vlad_>
pixman currently has YUY2 and YV12, and I can never keep those straight
- [20:27:34] <vlad_>
so I don't know how those are related to yuv420 and yuv422
- [20:27:41] * koen doesn't know anything about colourspaces
- [20:28:06] <vlad_>
or rather, I don't get why yuv420 isn't yuv411
- [20:28:20] <mru>
because 411 is something else
- [20:28:41] <vlad_>
that is so lame
- [20:29:06] <mru>
why would 411 be more logical than 420?
- [20:32:20] <vlad_>
well, 420 means that for every 4 luma samples there is one Cr and one Cb, right?
- [20:32:41] <mru>
I'm not sure the numbers are supposed to mean
- [20:32:54] <koen>
vlad_: which part is doing the 100ms painting? m-c or m-b?
- [20:33:06] <vlad_>
koen: fennec only, m-b
- [20:33:12] <mru>
but in 420 chroma is subsampled at half resolution horizontally and vertically
- [20:34:56] * prpplague^2 (n=dave123_@ppp-70-244-162-73.dsl.rcsntx.swbell.net) Quit ()
- [20:39:13] * trickie (n=trickie@hgaulton.xs4all.nl) Quit ("Leaving")
- [20:40:52] * artyomt1 (n=Artyom@198.78.191.90.dyn.estpak.ee) has left #beagle
- [20:41:18] * ReMixx (n=ReMixx@r58h135.res.gatech.edu) Quit ("Leaving")
- [20:58:47] <zedstar>
if anyone is interested ive made an ipk of a bluetooth to sqlite logger for the beagle
- [20:58:56] <zedstar>
are there any community repos for ipks?
- [21:02:16] * feig1 (n=ejf3@141.sub-75-192-6.myvzw.com) Quit ("Konversation terminated!")
- [21:05:36] <denix>
why not have it in oe?
- [21:06:42] * dcramer_ (n=davec@dcdsl.ebox.com) Quit (Read error: 104 (Connection reset by peer))
- [21:07:44] <zedstar>
yeh i should add to oe once it is a bit more polished
- [21:07:47] * dcramer (n=davec@dcdsl.ebox.com) has joined #beagle
- [21:16:01] <mru>
great, error in the errata document
- [21:27:30] * BThompson (n=BThompso@nat/ti/x-f14615a0d70cfc7b) Quit (Remote closed the connection)
- [21:36:49] <denix>
mru: errata to the errata... GNU is Not Unix... :)
- [21:39:42] * guillaum1 (n=Guillaum@AMontsouris-153-1-59-77.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
- [21:39:49] <mru>
turns out there is in fact a cortex-a8 r1p3
- [21:39:50] * guillaum1 (n=gl@AMontsouris-153-1-59-77.w90-2.abo.wanadoo.fr) has joined #beagle
- [21:43:29] * john280z (n=johnm@user-0c2h5av.cable.mindspring.com) has joined #beagle
- [21:45:49] * fer_luck (n=fer_luck@189.73.253.198) Quit ()
- [21:47:35] <Crofton>
jkridner, regarding your cc1 issue, can you try "opkg install cpp" and see if this solves the problem
- [21:48:47] <geist>
mru: hmm, hawkeye register says rev 3
- [21:48:58] <geist>
i forget if that is according to 3530 docs or not
- [21:49:43] <mru>
the document I have only goes up to ES2.1
- [21:49:51] <geist>
what ws the rev for 2.1?
- [21:49:57] <geist>
i think it was 1, right?
- [21:50:04] <geist>
which in 3430 is rev 2.0
- [21:50:28] <geist>
which is why i think the 3530 docs are wrong, and the numbering convention follows the 3430
- [21:50:36] <geist>
1 == 2.0, 2 = 2.1, 3 = 3.0
- [21:52:25] <mru>
my supposedly ES2.1 has 2
- [21:52:34] <mru>
as per 3430 docs
- [21:52:44] <mru>
3530 docs are obviously wrong
- [21:53:07] <geist>
yeah
- [21:53:16] <geist>
guess that's fine, since it reports the same core id as 3430
- [21:53:33] <geist>
so any code expected to run on both assumes they are revision compatible
- [22:06:23] * rsalveti (n=salveti@200.184.118.132) Quit (Remote closed the connection)
- [22:06:41] * zedstar (n=john@82-44-200-69.cable.ubr08.haye.blueyonder.co.uk) Quit (Remote closed the connection)
- [22:09:26] * flo_lap (n=fuchs@dslb-088-073-027-051.pools.arcor-ip.net) Quit ("Verlassend")
- [22:11:00] * dcordes (n=dcordes_@unaffiliated/dcordes) has joined #beagle
- [22:11:52] * TAKI (n=Administ@dslb-088-074-033-148.pools.arcor-ip.net) Quit ("Verlassend")
- [22:28:57] * dcordes (n=dcordes_@unaffiliated/dcordes) Quit (Remote closed the connection)
- [22:37:33] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
- [22:48:06] * Beagle0 (n=Beagle0@78.52.200.252) Quit (Read error: 104 (Connection reset by peer))
- [22:48:06] * Beagle1 (n=Beagle1@78.52.200.252) Quit (Read error: 104 (Connection reset by peer))
- [22:55:28] * hli (i=chaton@vig91-2-82-232-97-149.fbx.proxad.net) Quit ("ChatZilla 0.9.83 [Firefox 3.0.1/2008070208]")
- [23:00:48] * JoeBorn (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) has joined #beagle
- [23:03:56] * j24 (n=j24@ns32433.ovh.net) Quit (Read error: 110 (Connection timed out))
- [23:06:52] * ciskje (n=francesc@host76-84-dynamic.16-87-r.retail.telecomitalia.it) Quit (Remote closed the connection)
- [23:20:52] <jkridner|work>
geist: from a code compatibility standpoint, that is a safe assumption. we still need to work on the 03/15/25/30 detection, however.
- [23:21:53] <jkridner|work>
3430 vs. 3530 differences will show up in the I/O, but should not show up in code. board-level assumptions should dominate the kernel differences.
- [23:26:02] * jrmuizel (n=jrmuizel@mozca02.ca.mozilla.com) Quit ()
- [23:46:04] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) has joined #beagle
- [23:51:43] <geist>
jkridner|work: what about errata?
- [23:51:55] <geist>
usually the differences show up in how you handle errata
- [23:52:51] <geist>
if the 3530 and 3430 have in any way different errata that have to be handled differently, then the current strategy of not having a seperate chip id will bit it in the ass
- [23:54:33] * mru suspects the main difference between those chips is the label on the package
- [23:54:49] <geist>
yeah, that's my guess, though the actual errata documents from TI tell me otherwise
- [23:55:03] <mru>
the usb dma bug is in the latest 3530 errata list
- [23:55:18] <mru>
http://www.ti.com/litv/pdf/sprz278b
- [23:57:21] <geist>
what about the mmc dma bug?
- [23:58:14] <geist>
hmm, I dont see the usb dma bug in it
- [23:58:20] <mru>
chip variants are often made with the same masks, and some fuses are burned to disable various parts
- [23:58:34] <mru>
which version of the document do you have
- [23:58:42] <mru>
there's a new one on the web page
- [23:58:48] <mru>
dated august 2008
- [23:59:25] <mru>
"Advisory 3.0.1.130 USB DMA Cannot Handle Concurrent Channels"
- [23:59:59] <geist>
oh hmm, missed it.
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.