• [00:00:05] * guillaum1 (n=Guillaum@AMontsouris-153-1-70-121.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  • [00:00:26] * guillaum1 (n=gl@AMontsouris-153-1-70-121.w90-2.abo.wanadoo.fr) has joined #beagle
  • [00:01:01] * JuanG (n=Juan@nat/ti/x-41bde6a601a89279) has joined #beagle
  • [00:01:07] * JuanG (n=Juan@nat/ti/x-41bde6a601a89279) has left #beagle
  • [00:03:00] * zodttd|away is now known as zodttd
  • [00:04:19] <TangerinePentode> do people use the DSP a lot?
  • [00:04:32] <TangerinePentode> i guess that's vague
  • [00:04:54] <TangerinePentode> are there many applicaions that use the DSP on the beagle board?
  • [00:06:04] <nathanm> Other than some demonstrations with private code I'm not aware of any working DSP apps that can run on the BB right now. I'm hoping someone will jump in and correct me.
  • [00:08:22] <TangerinePentode> even though TI provides free linux dev tools?
  • [00:08:53] <mru> the trouble is getting the code onto the dsp and communicating with it
  • [00:09:18] <mru> they're not making it easy
  • [00:09:45] <nathanm> I've also heard that the free devtools generate code that isn't quite compatible with the DSPlink BIOS/driver we have for the BeagleBoard.
  • [00:09:58] <mru> I don't want to use dsplink
  • [00:10:01] <mru> it's a monster
  • [00:10:20] <TangerinePentode> it's scary, for shure
  • [00:10:23] <TangerinePentode> sure
  • [00:10:23] * odesus (n=Omar@201.165.141.180) has joined #beagle
  • [00:10:49] <nathanm> There's a wiki page claiming you can get OpenMAX for the DSP running on BB. It might be worth a try. That would accelerate multimedia codecs at least: http://elinux.org/BeagleBoard/DSP_Howto
  • [00:11:23] <mru> openmax is a monster in itself
  • [00:11:37] <TangerinePentode> is openmax open to the hardware, or is it built on top of something else?
  • [00:11:50] <nathanm> I know. I am really looking for a hello world style example I can build on. I guess that's what you're working on mru?
  • [00:12:20] <mru> I tried, and failed, to perform the IVA reset sequence
  • [00:12:32] <mru> the last stage never indicated completion
  • [00:13:02] <TangerinePentode> OpenMAX isn't "open" is it?
  • [00:13:15] <mru> more or less
  • [00:13:40] <nathanm> Are there any JTAG tools that let you get at the state of the DSP to see what went wrong?
  • [00:13:55] <mru> I don't have any jtag kit
  • [00:16:46] * Crofton|work dreams of a free dsp interface that is widely used
  • [00:16:52] <Crofton|work> and tools to support it
  • [00:17:09] <TangerinePentode> I feel like it should be easy enough to write a simple task scheduler for the DSO
  • [00:17:11] <TangerinePentode> DSP
  • [00:17:24] <mru> coding for the dsp is not the problem
  • [00:17:42] <mru> it's running it that's a bit of a bitch
  • [00:17:46] <TangerinePentode> :)
  • [00:17:52] <TangerinePentode> hmm
  • [00:19:56] <TangerinePentode> well, I have a copy of the dsp docs printed
  • [00:20:14] <TangerinePentode> trying to figure out how hard it would be to write an OS
  • [00:20:29] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [00:20:37] <mru> have you written an os before?
  • [00:20:45] <TangerinePentode> no
  • [00:20:52] <mru> worked on one at a low level?
  • [00:20:59] <TangerinePentode> yes
  • [00:21:09] <TangerinePentode> bootloader, device drive
  • [00:21:27] <TangerinePentode> more at a higher level
  • [00:21:43] <TangerinePentode> but i've also worked with microCosII
  • [00:22:02] <mru> we use that at work
  • [00:22:15] <mru> it's a nice little os
  • [00:22:23] <TangerinePentode> yeah, I like it too
  • [00:22:32] <TangerinePentode> i want to do something like that for the DSP
  • [00:22:44] <TangerinePentode> slaved to the linux kernel
  • [00:22:58] <TangerinePentode> with dynamically loaded code modules
  • [00:23:06] <mru> are you reading my mind?
  • [00:23:16] <TangerinePentode> nope
  • [00:23:21] <TangerinePentode> that just makes sense
  • [00:23:24] * mru adjusts tin-foil hat
  • [00:23:41] <TangerinePentode> except you'll do imaging with it
  • [00:23:47] <TangerinePentode> i'll do cryptography
  • [00:25:01] <nathanm> All I want is a way to get data from user mode into the DSP, operate on it, and ship the results home to user mode. But DSPlink docs were making my eyes glaze over. Trying them again now.
  • [00:25:25] <TangerinePentode> dsplink is scary
  • [00:25:32] <TangerinePentode> i use it a little at work
  • [00:25:36] <TangerinePentode> scary
  • [00:25:53] <TangerinePentode> too many heavy heavy layers
  • [00:25:54] <nathanm> Stop scaring me. I'm still innocent.
  • [00:26:40] <mru> nathanm: help us write something better instead
  • [00:26:54] <TangerinePentode> yeah, that's the key
  • [00:28:56] <nathanm> I've read the DSP interface section of the OMAP3 TRM and that's scary too. Lots of clocks, interrupts, DMA channels... even an MMU. o.o
  • [00:29:18] <TangerinePentode> that's just processor stuff
  • [00:29:22] <mru> the mmu looks easy
  • [00:29:34] <mru> clocking and power are the hard parts
  • [00:30:07] <TangerinePentode> where are you looking?
  • [00:30:34] <nathanm> So you would try to service the MMU by walking Linux page tables? Or just set it up for physical access?
  • [00:31:07] <TangerinePentode> i would just do physical address
  • [00:31:10] <mru> trying to synchronise it with the linux page tables could get tricky
  • [00:31:24] * odesus (n=Omar@201.165.141.180) has left #beagle
  • [00:32:01] <TangerinePentode> probably just set aside a chunk for communication, and give the DSP its own memory
  • [00:32:21] <mru> I'd try to be a bit more flexible than that
  • [00:32:49] <TangerinePentode> to enable sharing of memory?
  • [00:32:56] <mru> but first I need to get the damn thing out of reset
  • [00:33:02] <TangerinePentode> :)
  • [00:33:12] <TangerinePentode> that just sounds like a register somewhere
  • [00:33:30] <mru> it's an entire collection of registers
  • [00:33:38] <mru> and complicated sequences of operations
  • [00:34:01] <mru> set this to that, wait for this or that, set something else, etc.
  • [00:34:42] <TangerinePentode> ok
  • [00:34:53] <mru> it's all in the TRM
  • [00:34:54] <TangerinePentode> sounds like boot code
  • [00:35:00] <TangerinePentode> TRM?
  • [00:35:13] <nathanm> Reference Manual.
  • [00:35:25] <nathanm> It's very thorough. Like 4000 pages thorough.
  • [00:35:40] <TangerinePentode> where's that at?
  • [00:35:48] <mru> TI webpage
  • [00:35:56] <mru> or google spruf98b.pdf
  • [00:36:51] <TangerinePentode> it's huge
  • [00:37:22] <mru> section 14.4.1 is about booting the dsp
  • [00:37:50] <TangerinePentode> sweet
  • [00:38:34] * TAKI (n=Administ@dslb-088-074-040-034.pools.arcor-ip.net) Quit (Read error: 104 (Connection reset by peer))
  • [00:41:24] <TangerinePentode> Sweet Baby Jesus, it's 3400 pages
  • [00:43:42] <nathanm> Thorough.
  • [00:43:50] <nathanm> Except for the PVR stuff of course.
  • [00:43:58] <TangerinePentode> it's all the manuals in one!
  • [00:44:34] <mru> for extra fun, get the 800-page dsp manual and the 2000-page ARM manual
  • [00:44:51] <geist> yeah, TI loves writin them some docs
  • [00:45:09] <mru> the ARM manuals are written by ARM
  • [00:45:10] <TangerinePentode> i have the 800 page dsp manual
  • [00:45:23] <pbrook> IIRC the TI docs don't actually cover operation of the ARM core itself.
  • [00:45:23] <geist> talkin bout spruf89b
  • [00:45:25] <TangerinePentode> (printed out. i'm old and have bad eys)
  • [00:45:38] <mru> pbrook: correct
  • [00:45:42] <geist> sadly a lot of new stuff is in the new armv7 ARM
  • [00:45:45] <geist> which is terrible
  • [00:45:49] <mru> pbrook: only the omap-specific integration
  • [00:45:58] <mru> the armv7 arm is ok
  • [00:45:59] <geist> they rearranged it to be maximally unreadable
  • [00:46:08] <mru> takes a little getting used to, that's all
  • [00:46:13] <geist> wish they had a pre arm7 and arm7+
  • [00:46:18] <geist> instead of trying to mix it all up
  • [00:46:29] <mru> I prefer having it all in one place
  • [00:46:34] <geist> since half of the doc has hyperlinks to the appendixes that have the real info
  • [00:46:34] <pbrook> geist: No you don't. I was working with exactly that, and it's a real PITA.
  • [00:47:03] <geist> I guess.I still use the armv6 manual a lot until i hit something ineed the v7 one for
  • [00:47:04] <mru> geist: the split is only for the vmsa/pmsa stuff
  • [00:47:20] <geist> and vfp
  • [00:47:28] <pbrook> The main problem with the v7 manual is that it's not publicly available.
  • [00:47:44] <geist> was digging through that the other day, ended up getting clearer inf out of the cortex-a8 manual directly
  • [00:47:47] <Crofton|work> TangerinePentode, I could tell you areold from the pentode reference
  • [00:47:49] <mru> it's not terribly hard to get it
  • [00:47:51] <geist> since it at least didn't cover all the zillion revisions
  • [00:47:59] <Crofton|work> and I am with you in the printed docs
  • [00:48:15] <geist> but anyway, so it goes
  • [00:48:16] <TangerinePentode> not that old
  • [00:48:30] <geist> ARM makes the inexorable march towards complexity
  • [00:48:41] <TangerinePentode> but I studied radio many years ago and I like tubes
  • [00:49:17] <Crofton|work> "Hollow state technology"
  • [00:49:18] <nathanm> I have a little Atmel SAM7/TDMI7 uC running all bare-metal code. About 10 seconds after downloading the OMAP3 TRM I realized my bare metal days are done.
  • [00:49:44] * Crofton|work cries
  • [00:50:06] * Crofton|work bare metal days ended with the M6800
  • [00:50:11] <Crofton|work> note the two zeros
  • [00:50:12] <geist> aww you can bare metal an omap3
  • [00:50:20] <geist> it's just got a crapload more setup to do
  • [00:50:20] <TangerinePentode> tubes are nifty, and an easy way to do 10,000 W
  • [00:50:22] <nathanm> Some can. I hit my limit.
  • [00:50:27] <geist> :)
  • [00:50:42] <mru> nathanm: start with u-boot
  • [00:50:47] <mru> it has all the setup stuff
  • [00:50:52] <Crofton|work> setting up an ompa3 takes more RAM than I had on this
  • [00:50:57] <nathanm> Hehe.
  • [00:51:05] <TangerinePentode> is there a driver for the IPC module?
  • [00:51:06] <geist> i've got a little system set up for bare metal stuff, since i prefer it myself
  • [00:51:19] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit ()
  • [00:51:23] <geist> middle of writing another msusb driver so i can talk to it
  • [00:51:29] <TangerinePentode> bare metal is for 8-bits
  • [00:51:35] * gerrynjr (n=gerrynjr@gentoo/user/gerrynjr) has joined #beagle
  • [00:51:45] <gerrynjr> hello all
  • [00:51:46] <Crofton|work> http://en.wikipedia.org/wiki/MEK6800D2
  • [00:52:07] <gerrynjr> question.. the site says it does opengl... is this supported via X11?
  • [00:52:13] <gerrynjr> ie, I can port xbmc to this bad boy?
  • [00:52:13] <mru> not yet
  • [00:52:19] <gerrynjr> ah, alright
  • [00:52:22] <gerrynjr> thanks 8-)
  • [00:52:38] <gerrynjr> still would make an excellent robot controller as well
  • [00:52:47] <gerrynjr> what sort of power consumption are we looking at?
  • [00:52:52] <mru> 2W
  • [00:52:54] <mru> peak
  • [00:53:04] <TangerinePentode> dinner time
  • [00:53:07] <TangerinePentode> later all
  • [00:53:13] <Crofton|work> they are still working on the pm code ....
  • [00:53:16] <gerrynjr> mru: nuts
  • [00:53:44] <gerrynjr> ah... I just stumbled on the site... is this a recent development/community?
  • [00:53:50] <nathanm> mru: What I'd really like to do is zero copy from USB stack into DSP, so you can get/put data via USB camera/ethernet/etc. TI DSP Bridge Driver prefers to use a fixed physical buffer for I/O though.
  • [00:54:24] <geist> nathanm: that would be even better, since the msusb can't do scatter gather
  • [00:54:38] <geist> you could read direclty into the DSPs set aside buffer
  • [00:55:03] <mru> or point the dsp mmu wherever you please
  • [00:55:13] <nathanm> All the hardware is there to do nice stuff.
  • [00:55:34] <gerrynjr> hmm, awesome!
  • [00:55:53] <gerrynjr> looks like I may be going to the embedded conference... kist need time off from work
  • [00:55:55] <gerrynjr> *just
  • [00:56:17] <gerrynjr> bah... too expensive... nvm
  • [00:56:36] <nathanm> gerrynjr: I have heard from several people using BeagleBoard for robot controllers. It is a popular application due to performance per watt.
  • [00:56:53] <gerrynjr> nathanm: exactly
  • [00:56:58] <gerrynjr> and, you've got i2c...
  • [00:57:02] <gerrynjr> is it functional?
  • [00:57:12] <gerrynjr> lots of embedded controllers promise i2c... few deliver
  • [00:57:21] <nathanm> Yes but it is 1.8V so you need to add some level converters.
  • [00:57:22] <gerrynjr> (the qwerk board comes to mind)
  • [00:57:23] <geist> absolutely
  • [00:57:33] <gerrynjr> hrm
  • [00:57:35] <geist> it's necessary on omap3 since it talks to the pmic via it
  • [00:57:58] <nathanm> prpplague is working on a level converter daughterboard.
  • [00:58:47] <geist> bummer they didn't leave the pullup off the board. is the external i2c interface used by anything on the main board?
  • [00:58:52] <gerrynjr> hrm... add pwm, and it's like heaven 8-)
  • [00:59:01] <gerrynjr> but, i can do that using the serial port and a servo controller
  • [01:00:43] <gerrynjr> damn... quite a bit for $149
  • [01:00:45] <nathanm> gerrynjr: I am using a USB device to provide PWM/ADC/GPIO/etc with 5V tolerance.
  • [01:01:16] <gerrynjr> curious... can the DSP be used for other forms of signal processing?
  • [01:01:19] <gerrynjr> (aside from HD)
  • [01:01:29] <gerrynjr> thinking along the lines of a dedicated sensor controller
  • [01:02:10] <nathanm> Yes. It can run its own realtime OS while the ARM runs Linux.
  • [01:02:16] <gerrynjr> oooo
  • [01:02:23] <gerrynjr> awesome
  • [01:04:14] <gerrynjr> still cannot get over the board's size too, for what it has...
  • [01:05:07] <nathanm> TI makes good chips.
  • [01:05:58] <gerrynjr> heck, this thing could be a HTPC
  • [01:06:06] <gerrynjr> lots of things, to lots of people
  • [01:06:45] * rsalveti (n=salveti@189.70.254.7) Quit (Read error: 110 (Connection timed out))
  • [01:06:52] * rsalveti (n=salveti@189.70.31.46) has joined #beagle
  • [01:09:20] <gerrynjr> thanks for the feedback, I'll probably be on again soon... may order by the end of the week!
  • [01:09:21] <gerrynjr> 8-)
  • [01:10:29] <nathanm> Have fun!
  • [01:10:34] <gerrynjr> thanks!
  • [01:10:45] * gerrynjr (n=gerrynjr@gentoo/user/gerrynjr) Quit ()
  • [01:22:15] * woski (n=dleme@65.182.51.67) Quit (Read error: 104 (Connection reset by peer))
  • [01:22:44] * woski (n=dleme@65.182.51.67) has joined #beagle
  • [01:31:32] * doog (n=doug@cpe-75-80-160-171.san.res.rr.com) Quit ("Leaving")
  • [01:45:27] * spawt (i=spawt@freenet6.org) has joined #beagle
  • [01:57:51] * john280z (n=johnm@user-0c2h5av.cable.mindspring.com) Quit ("Ex-Chat")
  • [02:01:58] * ScratchMonkey (n=ScratchM@98.210.250.133) has joined #beagle
  • [02:02:38] <ScratchMonkey> anyone know of a firewire interface that will work with beagle? I need a tiny PC that will talk to one of these: http://agile-systems.com/Pages/Products/Motion_MicroMax.aspx
  • [02:19:14] * thephilosopher (n=chatzill@modemcable108.47-58-74.mc.videotron.ca) has joined #beagle
  • [02:19:32] * spawt (i=spawt@freenet6.org) Quit ("My damn controlling terminal disappeared!")
  • [02:19:51] * JoeyBorn (n=jborn@h-69-3-16-201.chcgilgm.dynamic.covad.net) has joined #beagle
  • [02:25:53] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [02:27:55] <nathanm> mru: I added TI DSP bridge patches to my .27 kernel and was able to run a short DSP program. After a second or two the kernel crashes with irq -33 errors... every time.
  • [02:35:21] * dcordes_ (n=dcordes_@unaffiliated/dcordes) Quit (Read error: 110 (Connection timed out))
  • [02:36:07] * dcordes_ (n=dcordes_@unaffiliated/dcordes) has joined #beagle
  • [02:37:54] * kozak (n=subbu@117.192.8.188) Quit (Read error: 110 (Connection timed out))
  • [02:38:12] * kozak (n=subbu@117.192.7.43) has joined #beagle
  • [02:46:39] <TangerinePentode> nathanm: what does the DSP bridge do?
  • [02:47:40] <rscott_> I'll jump in here - DSP bridge is another alternative to DSP link, i.e. a way to get code running on the DSP and communicate with it.
  • [02:48:58] <rscott_> I've been trying to get DSP link working all day, without any noticable progress. Ti's stable kernel doesn't have BB support, and kernels that do support BB aren't compatible with Ti's DSP link module.
  • [02:50:06] <rscott_> I tried DSP bridge last week, but was told by Ti that DSP link was really the way to go - but only on their EVM.
  • [02:51:37] <TangerinePentode> what does it do?
  • [02:51:48] <TangerinePentode> are they source patches?
  • [02:55:38] <rscott_> DSP link is built outside of the kernel as a module. When insmodded, it complains about invalid module format. Modinfo indicates that it was built with the correct kernel. I tried putting a printk in the initialization routine, but it never fired. My guess is that there's some kernel call that has been deprecated since Ti wrote it.
  • [02:56:18] <TangerinePentode> man, I hate when people do that
  • [02:57:23] <TangerinePentode> what about dsp bridge
  • [02:57:27] <TangerinePentode> is that open source?
  • [02:57:30] <rscott_> I haven't done any kernel hacking in a long time - and I'm not even sure how to get any info from the kernel when it's loading the module. Nothing shows up in dmesg...
  • [02:58:02] <TangerinePentode> you may have to build a new kernel with a lot of debugging enables
  • [02:58:08] <TangerinePentode> enabled
  • [02:58:13] <rscott_> I tried DSP bridge, but got kernel oopses when it was inserted...
  • [02:59:04] <rscott_> Yup - I remember doing a lot of kernel printks when trying to figure out why my mmu mapping wasn't working :-) Just hoping that things had improved...
  • [03:00:52] * slothlove (n=srussell@user-387ggve.cable.mindspring.com) has joined #beagle
  • [03:02:43] * kozak (n=subbu@117.192.7.43) Quit (Read error: 110 (Connection timed out))
  • [03:02:59] <TangerinePentode> fun fun
  • [03:03:31] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [03:04:35] * gduncan (n=spec@pool-98-116-240-138.nycmny.east.verizon.net) has joined #beagle
  • [03:07:18] * kozak (n=subbu@117.192.7.151) has joined #beagle
  • [03:08:12] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit (Remote closed the connection)
  • [03:08:52] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [03:16:10] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [03:22:07] * JoeyBorn (n=jborn@h-69-3-16-201.chcgilgm.dynamic.covad.net) Quit (Read error: 104 (Connection reset by peer))
  • [03:24:06] * zodttd (n=zodttd@user-142gtg7.cable.mindspring.com) Quit (Read error: 104 (Connection reset by peer))
  • [03:27:01] * kozak (n=subbu@117.192.7.151) Quit (Read error: 110 (Connection timed out))
  • [03:30:37] * kozak (n=subbu@117.192.6.227) has joined #beagle
  • [03:37:49] * JoeyBorn (n=jborn@h-66-167-69-5.chcgilgm.dynamic.covad.net) has joined #beagle
  • [03:38:39] * jrmuizel_ (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [03:39:57] * zodttd (n=zodttd@user-142gtg7.cable.mindspring.com) has joined #beagle
  • [03:41:21] * Olipro_ (n=Olipro@uncyclopedia/Olipro) has joined #beagle
  • [03:41:50] <nathanm> rscott_: Koen says OE will be working with TI in the next couple of weeks to get DSP Link embedded into Angstrom.
  • [03:42:32] * Olipro (n=Olipro@uncyclopedia/Olipro) Quit (Nick collision from services.)
  • [03:42:33] * Olipro_ is now known as Olipro
  • [03:44:09] <rscott_> Thanks for the update - unfortunately, I'm under considerable time pressure to show that OMAP is the way to go on this project. I'm going to see if I can get an EVM and/or get some more help from Ti on this.
  • [03:44:11] <nathanm> rscott_: To get DSP Bridge working, I had to make changes to the kernel config and boot options to set aside some memory, then tell bridgedriver where to find it (also had to use BIOS ddspbase_tiomap3430.dof64P)
  • [03:44:52] <nathanm> Then again, by 'working', I mean runs for a few seconds before crashing. So, you should probably stick to the road you're on.
  • [03:47:43] <rscott_> Well - at least a few seconds is better than no seconds! Is there anything I can do to help debug this? Assuming you've got time/energy to document how to get where you are...
  • [03:47:50] * jrmuizel_ (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit (Remote closed the connection)
  • [03:48:30] * jrmuizel_ (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [03:48:42] * jrmuizel_ (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit (Remote closed the connection)
  • [03:49:19] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) Quit (Read error: 104 (Connection reset by peer))
  • [03:49:51] <nathanm> I started by following these directions, but they're incomplete: http://elinux.org/BeagleBoard/DSP_Howto
  • [03:50:48] <nathanm> They assume you're already cross-compiling the kernel.
  • [03:51:28] <rscott_> Yeah - when I tried that method, my module just oopsed when insmodded. I was using 2.6.26, though - I'll try again with the latest kernel from the omap git tree...
  • [03:51:58] <nathanm> Do you have Dave's i2c patch for your omap git tree? Without that you won't boot.
  • [03:52:18] <rscott_> Dunno - how do I check?
  • [03:52:19] <nathanm> You should load the module like this: modprobe bridgedriver phys_mempool_base=0 base_img=/home/root/dsp/ddspbase_tiomap3430.dof64P
  • [03:52:28] <nathanm> Oh, you would know if you had. Let me dig that up for you.
  • [03:52:39] <nathanm> (Of course, fix the paths in that modprobe...)
  • [03:52:48] <rscott_> Yup...
  • [03:52:58] <nathanm> Here's the patch: http://marc.info/?l=linux-omap&m=122297833314280&w=2
  • [03:53:30] <nathanm> Also, go into your kernel config and turn off EHCI, and of course make sure you have CONFIG_FB_OMAP_CONSISTENT_DMA_SIZE=14
  • [03:54:56] <nathanm> Append mem=122m to CONFIG_CMDLINE. Then, before loading bridgedriver do modprobe phys_mempool_base=0x87a00000 phys_mempool_size=0x600000
  • [03:55:33] <nathanm> At that point, you should be able to run ping.out and see ping messages for a few seconds. Then it's off to irq -33 hell.
  • [03:56:18] <nathanm> I'm stumped on the irq -33 thing. It seems everyone has seen that at random times with all different kernels but nobody knows what to fix.
  • [03:57:24] <rscott_> Humm - has anyone looked at the DSP side? I wonder if it's issuing too many messages or malformed messages?
  • [03:57:36] <nathanm> I don't know how to get in there.
  • [03:58:33] <rscott_> Oh - I haven't gotten to the DSP/BIOS side of things yet. I (naively) assumed that source was available for it...
  • [03:58:56] <TangerinePentode> that stuff is pretty closed
  • [03:59:23] <nathanm> The kernel side is documented, but I couldn't find anything on the DSP side other than binaries.
  • [04:00:29] <rscott_> *sigh* I saw some earlier traffic about a desire for a simple DSP os. I could certainly get behind that!
  • [04:01:06] <rscott_> Wonder how hard it would be? Do we have any DSP sample code for the interprocessor communication?
  • [04:01:31] <TangerinePentode> I want to do that
  • [04:01:41] <TangerinePentode> but I have a boatload of reading to do
  • [04:01:51] <TangerinePentode> and a dayjob
  • [04:01:57] <TangerinePentode> and a girlfriend
  • [04:02:26] <nathanm> Sometimes I doubt your commitment to the Beagleboard.
  • [04:02:57] <TangerinePentode> I guess I'll have to stop sleeping
  • [04:03:12] <rscott_> For my purposes, all I need is a way to do load code, tell the DSP to execute it, and tell me when it's done. Though I imagine there's a lot of hidden complexity with respect to MMU consistency, atomicity, etc, etc
  • [04:03:38] <nathanm> I was hoping DSP Bridge would do that for me. And it kinda does.
  • [04:04:15] <TangerinePentode> for 3-5 seconds
  • [04:04:43] <rscott_> Is there any hope Ti would open it up for us? I don't mind trueing the wheel vs. inventing it...
  • [04:05:16] <TangerinePentode> yeah
  • [04:05:28] <TangerinePentode> I sent an email to someone at Ti
  • [04:06:03] <TangerinePentode> Jason Kridner
  • [04:06:29] <rscott_> I'll put it on my list of things to chat about when I next talk to them.
  • [04:07:10] <nathanm> Koen knows something about this. He has been getting TI to make things available for OE. The DSP is his latest target.
  • [04:07:45] * nooomem (n=chatzill@ip68-100-201-32.dc.dc.cox.net) has joined #Beagle
  • [04:08:24] <TangerinePentode> where is koen at?
  • [04:08:28] <TangerinePentode> europe?
  • [04:08:29] <rscott_> Koen has done some impressive work on OMAP - I'm sure he has Ti's attention...
  • [04:08:53] <nathanm> He might be able to pass you something before it's ready for OE but after it's available for public consumption.
  • [04:09:23] <rscott_> That would be cool...
  • [04:09:38] <nathanm> He's at University of Twente in the Netherlands.
  • [04:11:11] <TangerinePentode> curious about what time zone he's in
  • [04:11:12] <rscott_> Thanks for all the help guys - I
  • [04:11:17] <rscott_> 'm off to bed...
  • [04:12:07] <nathanm> Good luck rscott_.
  • [04:12:17] <TangerinePentode> night
  • [04:12:26] * rscott_ (n=rscott@c-76-121-5-243.hsd1.wa.comcast.net) Quit ("Zzzz")
  • [04:12:33] <nathanm> UTC+1
  • [04:13:40] <TangerinePentode> 9 hours away
  • [04:14:41] * bullwood42 (n=bullwood@adsl-99-171-171-142.dsl.aus2tx.sbcglobal.net) Quit (Read error: 110 (Connection timed out))
  • [04:22:51] * JoeyBorn (n=jborn@h-66-167-69-5.chcgilgm.dynamic.covad.net) Quit (Read error: 104 (Connection reset by peer))
  • [04:24:32] * Jaymus (n=chatzill@S0106000a5e21e0bc.cg.shawcable.net) has joined #beagle
  • [04:25:35] <Jaymus> I'm designing a tablet device using the Beagle. Does anyone know of a relatively cheap LCD display, approx. 9' I can get somewhere?
  • [04:29:40] <ds2> uh... 9' displays are not cheap nor do I think they make 9' single peice displays!
  • [04:30:27] <ds2> it'd be impressive though
  • [04:30:41] * slothlove (n=srussell@user-387ggve.cable.mindspring.com) Quit (Read error: 60 (Operation timed out))
  • [04:32:03] <TangerinePentode> :)
  • [04:32:15] <TangerinePentode> i think he means 9"
  • [04:32:35] * Jaymus (n=chatzill@S0106000a5e21e0bc.cg.shawcable.net) Quit (Remote closed the connection)
  • [04:34:22] <TangerinePentode> although a 9' display would be pretty cool, especially if they could curve it
  • [04:36:26] <ds2> they are doing a projector that can mate with the beagle
  • [04:36:36] <ds2> so a 9' display not completely impossible, just not in LCD technology
  • [04:36:56] <nathanm> Huh, I wonder what rscott's email is. I got DSP Bridge working without irq -33s.
  • [04:37:08] <ds2> the irq -33's are intermittient
  • [04:37:39] * JoeyBorn (n=jborn@h-68-165-56-248.chcgilgm.dynamic.covad.net) has joined #beagle
  • [04:37:51] <nathanm> Not these ones. I reproduced them 5 times in a row within seconds of executing the DSP command. But I get the intermittent kind too, always during boot up.
  • [04:38:13] <ds2> oh Hmmmm what's is the full line?
  • [04:38:13] * Jaymus (n=chatzill@S0106000a5e21e0bc.cg.shawcable.net) has joined #beagle
  • [04:38:44] <nathanm> One moment.
  • [04:39:02] <Jaymus> Sorry if anyone replied - got disconnected. Does anyone know of a cheap 800x480 LCD out there I could buy for a beagle project?
  • [04:39:58] <gduncan> Your easiest route is to gut a monitor.
  • [04:40:25] <nathanm> ds2: irq -33, desc: c0414580, depth: 0, count: 0, unhandled: 0
  • [04:41:08] <gduncan> As for finding and OEM unit, subscribe to EE Times and search the major manufacturers chinese web sites.
  • [04:42:29] <nathanm> ds2: I was triggering it by allowing phys_mempool_base=0 when loading bridgedriver which is not correct. Instead I reserved 6MB by passing mem=122m into the kernel and then starting the mempool at 0x87a00000. No more irq -33s.
  • [04:42:30] <ds2> Hmmm
  • [04:42:52] <ds2> nathanm: I wonder if it is a power management related interrupt
  • [04:42:53] * kozak (n=subbu@117.192.6.227) Quit (Read error: 110 (Connection timed out))
  • [04:43:12] <nathanm> ds2: I haven't been able to find anything on it Googling around, except that everyone seems to get it.
  • [04:43:36] <ds2> nathanm: everyone is using the L-O kernel which has little to no PM support... know if there is any reports on the reference kernel?
  • [04:43:49] <nathanm> Good question. The TI kernel you mean?
  • [04:43:58] <ds2> yes
  • [04:44:17] <ds2> I don't think I have seen it in my hybrid kernel w/PM support
  • [04:45:28] <nathanm> Hm. I wonder how long it will take to get into L-O.
  • [04:47:30] <TangerinePentode> irq -33 isn't in the TRM?
  • [04:48:30] * rsalveti (n=salveti@189.70.31.46) Quit (Remote closed the connection)
  • [04:48:50] * rsalveti (n=salveti@189.70.31.46) has joined #beagle
  • [04:51:44] <nathanm> Smarter people than me have said it might be a dispc problem.
  • [04:51:50] * asmola_ (n=asmola@c-24-62-166-165.hsd1.nh.comcast.net) has joined #beagle
  • [04:52:31] <Jaymus> gduncan: Thanks for the input, I'll give the searching method a whirl. Gutting a desktop monitor sucks as this is for an embedded application.
  • [04:54:15] <gduncan> In quantity, yes. I was thinking for prototypes and demo. Plus you see who the manufacturer of the panel really is.
  • [04:55:45] <gduncan> Are you planning on making lots of these things? Do they have to ruggedized?
  • [04:55:50] * asmola__ (n=asmola@c-24-62-166-165.hsd1.nh.comcast.net) has joined #beagle
  • [04:56:54] * Dan (n=dan@adsl-75-45-97-176.dsl.scrm01.sbcglobal.net) has joined #beagle
  • [04:57:06] * asmola__ (n=asmola@c-24-62-166-165.hsd1.nh.comcast.net) Quit (Client Quit)
  • [04:58:07] <Dan> Is anyone here running Angstrom with USB support?
  • [04:58:52] <nathanm> You mean USB host support?
  • [04:59:04] * soman (n=somnath@61.16.248.242) has joined #beagle
  • [04:59:08] <Dan> yes. I've tried everything and can't get it to work
  • [04:59:26] <nathanm> Oh, that's strange. Last time I tried Angstrom kernel the USB worked okay -- not perfectly but good enough. What device do you have?
  • [05:00:31] <Dan> uImage_OTG from code.google.com logs like USB is working, then the log says it is entering TI power-down mode. The video never turns on. I am using a USB keyboard and mouse
  • [05:01:05] <Dan> That's the farthest I've gotten.
  • [05:01:22] <nathanm> Do you have a powered hub?
  • [05:01:45] <Dan> Yes. Beagle is plugged into the 'host' USB-B connector.
  • [05:02:18] <nathanm> And you have a mini-A cable, not mini-B?
  • [05:02:42] <Dan> What is the difference?
  • [05:02:59] <Dan> is the TX/RX swapped?
  • [05:03:14] <nathanm> Mini-B won't work. Mini-B is also much more common. Usually you will see the letter A on the side plugged into the BeagleBoard.
  • [05:03:25] <nathanm> Mini-A grounds the ID pin.
  • [05:04:29] <Dan> I don't see markings on it. I will assume at this point that it is the incorrect cable. The mouse LED never turns on.
  • [05:05:18] <Dan> Thanks for your help nathanm!
  • [05:05:25] <nathanm> Your hub has a wall wart, right?
  • [05:05:45] <Dan> yes. And the cable is made by ziotek
  • [05:06:01] <nathanm> Huh, weird. My mouse lights up just from being plugged into the hub, but I guess they're all different.
  • [05:06:55] <nathanm> Does lsusb show you anything?
  • [05:07:05] * nooomem (n=chatzill@ip68-100-201-32.dc.dc.cox.net) Quit ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
  • [05:07:18] * nooomem (n=chatzill@ip68-100-201-32.dc.dc.cox.net) has joined #Beagle
  • [05:07:50] <Dan> I can't even get to the prompt. If I boot a different uImage without USB I can get a prompt.
  • [05:08:08] <Dan> <3>Power Management for TI OMAP. <- last thing in the log
  • [05:08:48] <nathanm> Where are you finding these kernels?
  • [05:09:37] <Dan> http://code.google.com/p/beagleboard/downloads/list
  • [05:10:12] <Dan> I do have a mini-B cable. I checked the website for the manufacturer
  • [05:10:32] * asmola (n=asmola@c-24-62-166-165.hsd1.nh.comcast.net) Quit (Read error: 110 (Connection timed out))
  • [05:10:33] <nathanm> Well, if it's any consolation, you are also using outdated kernels.
  • [05:10:53] <nathanm> Use this one: http://www.angstrom-distribution.org/demo/beagleboard/
  • [05:10:54] * JoeBorn__ (n=jborn@h-69-3-100-249.chcgilgm.dynamic.covad.net) has joined #beagle
  • [05:11:15] * JoeyBorn (n=jborn@h-68-165-56-248.chcgilgm.dynamic.covad.net) Quit (Read error: 110 (Connection timed out))
  • [05:11:15] <nathanm> Also, I recommend this guide for getting Angstrom going. It's the most up to date I've found: http://elinux.org/BeagleBoardBeginners
  • [05:13:12] <Dan> Thanks. I tried the angstrom-distribution before and it didn't work - most likely because of my cable problem.
  • [05:13:44] <nathanm> Yeah, those mini-As are kind of hard to find. I suggest you go online. I wasted a day scouring computer shops to no avail.
  • [05:13:46] * asmola_ (n=asmola@c-24-62-166-165.hsd1.nh.comcast.net) Quit (Read error: 110 (Connection timed out))
  • [05:14:28] <Dan> Is there any other setup that needs to be done with that version? I've seen all kinds of weird commands needed to get USB working on other systems.
  • [05:14:33] <nathanm> Somebody told me that there was a 5-in-1 USB adaptor kit at Big Lots that includes the mini-A, so if you have a Big Lots around it might be worth looking.
  • [05:14:55] <nathanm> Nope. With the Angstrom disto kernel it will just look for the ID pin (telling it you have the A cable), and use that to autoconfigure host mode.
  • [05:15:18] <Dan> Big lots is right down the street. I'll check it out tomorrow.
  • [05:16:36] <Dan> Thanks again for the help! I was at wits-end. Normally I use windows and the setup for linux is daunting.
  • [05:16:44] <nathanm> You sound like you're real close now.
  • [05:17:02] * shiv (n=root@59.160.172.220) has joined #beagle
  • [05:17:26] * dirk2 (n=dirk@F3249.f.strato-dslnet.de) has joined #beagle
  • [05:18:41] * Dan (n=dan@adsl-75-45-97-176.dsl.scrm01.sbcglobal.net) has left #beagle
  • [05:35:38] * Olipro (n=Olipro@uncyclopedia/Olipro) Quit (Connection timed out)
  • [05:36:50] * JuanG (n=Juan@nat/ti/x-959fc3545c6fb82b) has joined #beagle
  • [05:36:53] * JuanG (n=Juan@nat/ti/x-959fc3545c6fb82b) has left #beagle
  • [05:38:16] * Jaymus (n=chatzill@S0106000a5e21e0bc.cg.shawcable.net) Quit (Read error: 110 (Connection timed out))
  • [05:49:10] * JoeyBorn (n=jborn@h-64-105-84-220.chcgilgm.dynamic.covad.net) has joined #beagle
  • [05:50:55] * JoeBorn__ (n=jborn@h-69-3-100-249.chcgilgm.dynamic.covad.net) Quit (Read error: 110 (Connection timed out))
  • [05:55:59] * dcordes (n=dcordes@unaffiliated/dcordes) has joined #beagle
  • [05:56:06] * dcordes_ (n=dcordes_@unaffiliated/dcordes) Quit ("leaving")
  • [06:14:35] * nooomem (n=chatzill@ip68-100-201-32.dc.dc.cox.net) Quit ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
  • [06:19:01] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [06:19:31] * Jaymus (n=chatzill@S0106000a5e21e0bc.cg.shawcable.net) has joined #beagle
  • [06:38:04] * JoeyBorn (n=jborn@h-64-105-84-220.chcgilgm.dynamic.covad.net) Quit (Read error: 113 (No route to host))
  • [06:38:09] * JoeBorn_ (n=jborn@h-68-165-57-158.chcgilgm.dynamic.covad.net) has joined #beagle
  • [06:39:14] * _Codeman_ (n=wirtz@arktur.dfn.de) has joined #beagle
  • [06:40:02] * Beagle0 (n=Beagle0@s-inf-pc117.oulu.fi) has joined #beagle
  • [06:42:00] * soman (n=somnath@61.16.248.242) Quit (Read error: 104 (Connection reset by peer))
  • [06:53:33] * JoeBorn_ (n=jborn@h-68-165-57-158.chcgilgm.dynamic.covad.net) Quit (Read error: 104 (Connection reset by peer))
  • [06:57:20] * soman (n=somnath@61.16.248.242) has joined #beagle
  • [06:59:59] * JoeBorn_ (n=jborn@h-68-164-10-149.chcgilgm.dynamic.covad.net) has joined #beagle
  • [07:00:49] * Beagle0 (n=Beagle0@s-inf-pc117.oulu.fi) Quit (Remote closed the connection)
  • [07:03:01] * zodttd is now known as zodttd|away
  • [07:08:36] * cedric_ (n=cedric@vpn.echovox.com) Quit ("Ex-Chat")
  • [07:25:35] * and-ri (n=x0103690@nat/ti/x-7b2f51d55e89d16f) has joined #beagle
  • [07:29:29] <koen> good morning all
  • [07:30:21] <thephilosopher> allo
  • [07:33:13] <dirk2> koen: I keep fingers crossed that DVB patch makes your stick work
  • [07:37:10] * TAK2004 (n=Administ@dslb-088-074-040-034.pools.arcor-ip.net) has joined #beagle
  • [07:47:31] * Beagle9 (n=Beagle9@s-inf-pc117.oulu.fi) has joined #beagle
  • [07:47:41] * Beagle9 is now known as ewerw
  • [07:47:49] <[X]Spot> If i did all the steps here http://www.angstrom-distribution.org/building-%C3%A5ngstr%C3%B6m then how can I build a package (vlc for example) using my cutom configuration settings ?
  • [07:52:41] <koen> dirk2: trying to apply it now :)
  • [07:54:35] * Jaymus (n=chatzill@S0106000a5e21e0bc.cg.shawcable.net) Quit (Read error: 110 (Connection timed out))
  • [07:59:37] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [08:00:11] * Beagle2 (n=Beagle2@c-71-231-159-198.hsd1.or.comcast.net) has joined #beagle
  • [08:00:16] <Beagle2> hello
  • [08:00:45] <Beagle2> does beagleboard run xmame? I saw it is a package for angstrom
  • [08:01:48] <Beagle2> does beagleboard run xmame?
  • [08:02:05] * Beagle2 (n=Beagle2@c-71-231-159-198.hsd1.or.comcast.net) Quit (Client Quit)
  • [08:02:44] * zedstar (n=john@fsf/member/zedstar) has joined #beagle
  • [08:28:26] * roxfan2 (n=dunno@2.214-200-80.adsl-dyn.isp.belgacom.be) has joined #beagle
  • [08:28:56] * zodttd|away is now known as zodttd
  • [08:41:04] * roxfan (n=dunno@2.214-200-80.adsl-dyn.isp.belgacom.be) Quit (Read error: 110 (Connection timed out))
  • [08:43:37] * JoeyBorn (n=jborn@h-68-165-188-143.chcgilgm.dynamic.covad.net) has joined #beagle
  • [08:44:49] * JoeBorn_ (n=jborn@h-68-164-10-149.chcgilgm.dynamic.covad.net) Quit (Read error: 110 (Connection timed out))
  • [08:50:11] * soman (n=somnath@61.16.248.242) Quit (Remote closed the connection)
  • [08:52:20] * soman (n=somnath@61.16.248.242) has joined #beagle
  • [09:05:26] * zodttd is now known as zodttd|away
  • [09:14:12] * ali_as (n=ali_as@ambix.plus.com) Quit (kubrick.freenode.net irc.freenode.net)
  • [09:14:12] * swetland (n=swetland@nat/google/x-6eae5cb1f4c43e79) Quit (kubrick.freenode.net irc.freenode.net)
  • [09:14:12] * Pavlov (n=pavlov@shell.off.net) Quit (kubrick.freenode.net irc.freenode.net)
  • [09:14:12] * _apk (n=kows@85-18-136-98.fastres.net) Quit (kubrick.freenode.net irc.freenode.net)
  • [09:15:07] * ali_as (n=ali_as@ambix.plus.com) has joined #beagle
  • [09:15:07] * swetland (n=swetland@nat/google/x-6eae5cb1f4c43e79) has joined #beagle
  • [09:15:07] * Pavlov (n=pavlov@shell.off.net) has joined #beagle
  • [09:15:07] * _apk (n=kows@85-18-136-98.fastres.net) has joined #beagle
  • [09:16:19] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [09:17:06] * jess (n=jess@59.160.172.220) has joined #beagle
  • [09:23:25] * ScratchMonkey (n=ScratchM@98.210.250.133) Quit ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
  • [09:32:41] <_Codeman_> Hi all!
  • [09:33:15] <_Codeman_> Is anyone around who can help me getting beagle startet with an Angstrom-Image from SD?
  • [09:38:47] * thephilosopher (n=chatzill@modemcable108.47-58-74.mc.videotron.ca) Quit ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
  • [09:44:34] * thephilosopher (n=chatzill@modemcable108.47-58-74.mc.videotron.ca) has joined #beagle
  • [09:48:10] <chakie_work> _Codeman_: i used these docs to format the card: http://wiki.omap.com/index.php?title=MMC_Boot_Format
  • [09:48:17] <chakie_work> of course, the site is down now
  • [09:51:35] <chakie_work> and then repartition according to this: http://code.google.com/p/beagleboard/wiki/LinuxBootDiskFormat
  • [09:52:45] <chakie_work> copy the MLO, u-boot.bin, uImage and x-load.bin.ift to the first FAT partition on the SD card. The MLO file should probably be copied first.
  • [09:53:13] <chakie_work> extract the root filesystem to the second partition (the ext3 one)
  • [09:53:51] <chakie_work> i've used the demo image from here: http://www.angstrom-distribution.org/node/47
  • [10:01:06] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [10:16:37] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  • [10:19:21] * shriram (i=3b6030f5@gateway/web/ajax/mibbit.com/x-3ed6367c2fb6e773) has joined #beagle
  • [10:20:55] <dirk2> chakie_work: http://wiki.omap.com/~davincid/wiki/index.php?title=MMC_Boot_Format works for me
  • [10:21:16] <chakie_work> didn't for me
  • [10:21:28] <chakie_work> does now
  • [10:21:42] <dirk2> _Codeman_: Do you know http://elinux.org/BeagleBoardBeginners ?
  • [10:22:53] <chakie_work> i searched for that one too but didn't find it :)
  • [10:25:35] <dirk2> At http://elinux.org/BeagleBoard (this is easy to remember?) in section "Links" under "Other Beagle wiki pages" there is a list of all other Beagle related eLinux wiki pages ;)
  • [10:27:33] <chakie_work> the beagle suffers from a reversed documentation problem
  • [10:27:42] <chakie_work> there's too much of it all over the place :)
  • [10:31:08] <dirk2> Put a bookmark to eLinux and code.google.com wiki and you should have everything you need ;)
  • [10:31:46] * roxfan (n=dunno@2.214-200-80.adsl-dyn.isp.belgacom.be) has joined #beagle
  • [10:32:39] <chakie_work> yeah, i've been collecting resources i've used and steps i've done into my own wiki
  • [10:33:27] <chakie_work> it's frustrating to know that somewhere was a good page about, say, formatting the memory card but you just can't find it
  • [10:34:37] <_Codeman_> @chakie: I will take a look at this page, thanks!
  • [10:35:25] <_Codeman_> @dirk2: Thanks, I know this page.
  • [10:36:07] <_Codeman_> @chakie,dirk2: The kernel seem sto boot but I have problems with mounting / on the ext3 partition...
  • [10:37:26] * RogerMonk (n=a0740758@nat/ti/x-7de11702bb0c5c06) has joined #beagle
  • [10:37:55] <_Codeman_> I get "No filesystem could mount root, tried: ext3 ext2 msdos vfat". It looks like there is a problem with finding the second partition on the SD. I found some hints on the net but they won't help
  • [10:38:29] <_Codeman_> Perhaps this is a problem because I have a 4 GB SD?
  • [10:38:50] <and-ri> chakie_work: what do zou want to do with the formatting page for the MC? which Os you use?
  • [10:39:19] <chakie_work> and-ri: i use linux
  • [10:39:58] <chakie_work> _Codeman_: no idea about sizes, i happended to have a 2g card (as in the docs) and it worked fine
  • [10:40:23] * shriram (i=3b6030f5@gateway/web/ajax/mibbit.com/x-3ed6367c2fb6e773) Quit ("http://www.mibbit.com ajax IRC Client")
  • [10:41:20] <_Codeman_> Ahhh..... I am trying to mount my rootfs with a card reader on my i386 and this won't work, too....
  • [10:43:03] <chakie_work> i had issues with a tool old card reader that wouldn't even mount my 2g card. didn't work in windows/osx either
  • [10:43:17] <chakie_work> after getting a new reader it worked fine
  • [10:43:56] * roxfan2 (n=dunno@2.214-200-80.adsl-dyn.isp.belgacom.be) Quit (Read error: 110 (Connection timed out))
  • [10:45:26] <_Codeman_> The card reader itself seems to work: Currently I am writing the Angstrom toofs on the SD...
  • [10:48:31] <_Codeman_> Ok. I tried to unmount and mount... seems to work on the i386. Now for the beagle
  • [10:56:04] <and-ri> chakie_work: so i have no idea to format mc under linux. but in windows xfdsik is realy helpful
  • [10:56:04] * JoeyBorn (n=jborn@h-68-165-188-143.chcgilgm.dynamic.covad.net) Quit (Read error: 104 (Connection reset by peer))
  • [10:56:33] <chakie_work> and-ri: the docs listed above contain all you need to know
  • [10:58:24] <_Codeman_> Don't ask me why but now it runs!
  • [10:59:47] <_Codeman_> But I had to boot twice... the first time the beagle won't load the uImage. After restarting it works for the first time. Perhaps my SD card has a problem...
  • [11:00:32] * igor321_ (n=igor@141-241.dsl.iskon.hr) has joined #beagle
  • [11:03:01] <and-ri> Chakie_work: oh i thought you need help with it :-)
  • [11:08:42] <_Codeman_> :-) I thaught, too... But it really seems to be a problem with the SD. Angstrom bootet. The I made a reboot and now the same problem as before: "VFS: Cannot open root device "mmcblk0p2" or unknown-block(2,0)" :-(
  • [11:09:58] <chakie_work> _Codeman_: the first boot takes much longer as it does some unpacking etc
  • [11:10:36] * JoeyBorn (n=jborn@h-69-3-16-130.chcgilgm.dynamic.covad.net) has joined #beagle
  • [11:11:56] <_Codeman_> @chakie: Yep, I read this in the Beginners guide.
  • [11:14:19] <koen> it doesn't do unpacking
  • [11:14:35] <koen> it configures software that can't be configured on the buildmachine
  • [11:22:25] * geist (n=geist@c-98-234-105-90.hsd1.ca.comcast.net) Quit (Read error: 110 (Connection timed out))
  • [11:25:24] <chakie_work> ok, someone has before said that the delay is because of unpacking
  • [11:28:07] * geist (n=geist@c-98-234-105-90.hsd1.ca.comcast.net) has joined #beagle
  • [11:30:39] * dirk2 (n=dirk@F3249.f.strato-dslnet.de) has left #beagle
  • [11:33:19] <ewerw> hi I would like to download some files from my PC to the beagleboard's SD card through the serial connection how could I do that?
  • [11:33:54] <chakie_work> kermit?
  • [11:35:01] * koen would just get a card reader or use a network
  • [11:37:01] <_Codeman_> @chakie: Hmmm, really strange... no it boots up the third time without problem...
  • [11:37:29] <chakie_work> well, i'm definitely no expert, just a n00b that happened to have a few links handy
  • [11:38:18] <ewerw> I would like to test the compiled program on the beagleboard without moving the SD card, or rebooting Angstrom.
  • [11:39:13] <Stskeeps> I have a stupid problem.. i have a revision B5, took got one of these from a mobo, http://www.pccables.com/07120.htm , where i have the red wire down towards the MMC reader in direction, i have tried two different null modem cables, 115200 8N1, no flow control, two-three different machines, minicom and hyperterminal, board boots up with sound and lights, but no serial output.. any clever ideas or did i manage to burn out the serial pins?
  • [11:39:54] <_Codeman_> Hmmmm now the beagle hangs at "Starting OS Bootloader..." I think my SD card is a stupid one :-(
  • [11:39:54] <Stskeeps> on one of them i had a bit of line noise though
  • [11:40:28] <_Codeman_> @Stskeeps: I had to use a crossover serial cable to get minicom to work with the beagle
  • [11:40:42] <Stskeeps> _Codeman_: crossover.. isn't that a nullmodem though?
  • [11:41:47] <_Codeman_> @Stskeeps: Yep, I think so... with nullmodem you might have handshake crossover also but I dont know if this is important for the beagle
  • [11:44:15] <Stskeeps> really has me stumped though - had no problems with null modem to terminal on a server so :p
  • [11:50:27] <dcordes> in the hardware reference manual it is empphasized that other voltages than 5V can do harm to the beagleboard. what happens when the voltage drops beyond 5V?
  • [11:51:41] <dcordes> *below 5V
  • [11:51:41] * dcramer_ (n=davec@dcdsl.ebox.com) Quit (Read error: 104 (Connection reset by peer))
  • [11:55:31] * DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) Quit (Read error: 104 (Connection reset by peer))
  • [11:55:58] * DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) has joined #beagle
  • [11:59:48] * Openfree (n=df@222.65.132.11) has joined #beagle
  • [12:01:02] * dcramer (n=davec@dcdsl.ebox.com) has joined #beagle
  • [12:07:18] * felipec (i=c0647cdb@gateway/web/ajax/mibbit.com/x-1979338ad20040d4) has joined #beagle
  • [12:10:54] * fagius (n=fagius@static-71-111-254-119.rlghnc.dsl-w.verizon.net) has joined #beagle
  • [12:16:34] * JoeyBorn (n=jborn@h-69-3-16-130.chcgilgm.dynamic.covad.net) Quit (Read error: 110 (Connection timed out))
  • [12:16:55] * JoeyBorn (n=jborn@h-68-164-7-206.chcgilgm.dynamic.covad.net) has joined #beagle
  • [12:17:46] * BThompson (n=BThompso@nat/ti/x-8bf4f78ef1e6e68c) has joined #beagle
  • [12:19:21] * feig (n=ejf3@244.sub-75-193-18.myvzw.com) has joined #beagle
  • [12:25:28] * gduncan (n=spec@pool-98-116-240-138.nycmny.east.verizon.net) Quit ()
  • [12:32:12] * soman (n=somnath@61.16.248.242) Quit ("Leaving")
  • [12:47:01] * ewerw (n=Beagle9@s-inf-pc117.oulu.fi) Quit (Remote closed the connection)
  • [12:47:54] * slothlove (n=srussell@user-387ggve.cable.mindspring.com) has joined #beagle
  • [12:50:24] * Beagle9 (n=Beagle9@s-inf-pc117.oulu.fi) has joined #beagle
  • [12:50:43] * Beagle9 is now known as ewerw
  • [13:00:00] * JoeyBorn (n=jborn@h-68-164-7-206.chcgilgm.dynamic.covad.net) Quit (Read error: 110 (Connection timed out))
  • [13:00:14] * shriram (i=3b6030f5@gateway/web/ajax/mibbit.com/x-4fc6c9079a852d29) has joined #beagle
  • [13:01:56] * igor321_ is now known as igor321
  • [13:05:26] * shriram (i=3b6030f5@gateway/web/ajax/mibbit.com/x-4fc6c9079a852d29) Quit ("http://www.mibbit.com ajax IRC Client")
  • [13:10:29] * docelic (n=docelic@78.134.200.56) has joined #beagle
  • [13:20:09] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
  • [13:20:41] * rsalveti (n=salveti@189.70.31.46) Quit (Read error: 110 (Connection timed out))
  • [13:34:41] * robclark (n=robclark@nat/ti/x-af12463ebb0cb0a5) has joined #beagle
  • [13:38:32] * rsalveti (n=salveti@200.184.118.132) has joined #beagle
  • [13:42:16] * rscott_ (n=rscott@c-76-121-5-243.hsd1.wa.comcast.net) has joined #beagle
  • [13:45:44] * mib_r3a7ec (i=3b6030f5@gateway/web/ajax/mibbit.com/x-21190db08d0bf965) has joined #beagle
  • [13:46:27] <mib_r3a7ec> what is T2? is it inside OMAP?
  • [13:46:50] * Olipro (n=Olipro@uncyclopedia/Olipro) has joined #beagle
  • [13:49:42] <roxfan> t2? where's that from?
  • [13:49:51] * jess (n=jess@59.160.172.220) has left #beagle
  • [13:50:05] <mib_r3a7ec> Triton 2 i guess
  • [13:50:42] * jrmuizel (n=jrmuizel@mozca02.ca.mozilla.com) has joined #beagle
  • [13:52:45] * JoeyBorn (n=jborn@nmd.sbx02252.chicail.wayport.net) has joined #beagle
  • [13:52:54] <roxfan> i have no idea what you're talking about...
  • [13:55:13] <_Codeman_> Does anyone know if there is a "Howto use a toolchain for newbies on beagleboards" around?
  • [13:56:14] <koen> _Codeman_: http://elinux.org/BeagleBoardAndOpenEmbeddedGit
  • [13:56:51] <pbrook> _Codeman_: It tends to be specific to whatever distro you're running, rather than the hardware itself.
  • [13:56:56] <koen> mib_r3a7ec: T2 ismost likely the TWL4030, codenamed triton2
  • [13:58:38] <_Codeman_> @koen: Thanks!
  • [13:58:50] * _Codeman_ is back
  • [13:58:59] * _Codeman_ is back
  • [13:59:16] * _Codeman_ is away: until tomorrow
  • [13:59:19] <mib_r3a7ec> not really sure. since in 2430 sdp - they used to say T2 and TWL4030. not sure if they were speaking in contexts
  • [14:01:00] <koen> zuh: http://www.fourcc.org/yuv.php
  • [14:11:39] * shriram (i=3b6030f5@gateway/web/ajax/mibbit.com/x-5967c40d489015ee) has joined #beagle
  • [14:12:20] <slothlove> why is it that opkg says that tk-dev only depends on tk, and it also tells me that tk is installed, but when I try to install tk-dev it complains about dependencies it can't satisfy, including packages like util-macros and a bunch of packages ending in "proto"
  • [14:13:36] <shriram> does the new style twl drivers fixes any bugs?. currently am using 2.6.26 and am in process of applying changes/bug fixes after 2.6.26
  • [14:15:55] <shriram> what do guys think? how should i go about it. after 2.6.26 there are plenty of changes. what is the best way of going about it.
  • [14:24:54] <koen> I'd wait for 2.6.27-omap1 to get tagged
  • [14:25:03] * mib_r3a7ec (i=3b6030f5@gateway/web/ajax/mibbit.com/x-21190db08d0bf965) Quit ("http://www.mibbit.com ajax IRC Client")
  • [14:25:51] <shriram> okay.
  • [14:27:55] <shriram> and then. you compare the patches 2.6.26-omap2 patch and 2.6.27-omap2 patch?
  • [14:30:00] <koen> that won't work
  • [14:30:11] <koen> twl4030 was pretty much rewritten
  • [14:30:17] <koen> and all the arm headers moved
  • [14:32:20] <shriram> how does one go forward. you need to freeze the kernel somewhere and just apply the bug fixes.
  • [14:33:37] * igor321 (n=igor@141-241.dsl.iskon.hr) Quit ("leavin'")
  • [14:34:31] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [14:48:54] * JoeyBorn (n=jborn@nmd.sbx02252.chicail.wayport.net) Quit (Read error: 113 (No route to host))
  • [14:52:00] * TAKI (n=Administ@dslb-088-074-055-212.pools.arcor-ip.net) has joined #beagle
  • [14:59:03] <felipec> _Codeman_: have you tried codesourcery's toolchains?
  • [15:09:23] * TAK2004 (n=Administ@dslb-088-074-040-034.pools.arcor-ip.net) Quit (Read error: 110 (Connection timed out))
  • [15:13:29] * ewerw (n=Beagle9@s-inf-pc117.oulu.fi) Quit (Remote closed the connection)
  • [15:15:53] * gletelli_ (n=chatzill@fw-tnat.cambridge.arm.com) has joined #beagle
  • [15:20:59] * docelic (n=docelic@78.134.200.56) Quit ("http://www.spinlocksolutions.com/")
  • [15:22:07] * Openfree (n=df@222.65.132.11) Quit (Remote closed the connection)
  • [15:22:45] * zedstar (n=john@fsf/member/zedstar) Quit ("Leaving")
  • [15:24:07] * shriram (i=3b6030f5@gateway/web/ajax/mibbit.com/x-5967c40d489015ee) has left #beagle
  • [15:28:07] * shiv (n=root@59.160.172.220) Quit ("Leaving.")
  • [15:44:53] * and-ri (n=x0103690@nat/ti/x-7b2f51d55e89d16f) Quit ()
  • [15:55:32] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [16:09:17] * igor321 (n=igor@141-241.dsl.iskon.hr) has joined #beagle
  • [16:09:53] * doog (n=doug@cpe-75-80-160-171.san.res.rr.com) has joined #beagle
  • [16:23:28] * igor321 (n=igor@141-241.dsl.iskon.hr) Quit ("leavin'")
  • [16:29:07] <TangerinePentode> koen: how much of the linux/dsp code have you seen?
  • [16:32:16] <TangerinePentode> we were trying to figure out where ti might be on opening it up, so that we could write an alternate OS
  • [16:32:53] <felipec> TangerinePentode: an alternative DSP OS?
  • [16:33:18] <TangerinePentode> yes
  • [16:33:31] <TangerinePentode> much much simpler
  • [16:33:42] <TangerinePentode> just a task scheduler really
  • [16:34:02] <felipec> TangerinePentode: like DSP gateway?
  • [16:34:07] <felipec> http://dspgateway.sourceforge.net/
  • [16:34:43] <TangerinePentode> such a thing exists?
  • [16:34:48] <TangerinePentode> cool
  • [16:34:56] <suihkulokki> I think TangerinePentode means alternative OS to run _on_ the dsp, ie instead of the BIOS TI provides
  • [16:36:13] <TangerinePentode> dspgateway just talks to dsp/bios?
  • [16:36:25] <TangerinePentode> yeah, I mean on the DSP
  • [16:36:51] * thephilosopher (n=chatzill@modemcable108.47-58-74.mc.videotron.ca) Quit (Read error: 110 (Connection timed out))
  • [16:37:04] <felipec> suihkulokki: yes, dspgateway has a DSP OS
  • [16:37:17] <TangerinePentode> yeah, it does
  • [16:37:24] <TangerinePentode> reading the webpage
  • [16:37:37] <TangerinePentode> does it run on the 3530?
  • [16:38:31] <TangerinePentode> felipec: thanks
  • [16:39:12] <felipec> TangerinePentode: nope, not 3530
  • [16:39:12] <TangerinePentode> it's a nokia project
  • [16:39:53] <felipec> TangerinePentode: yes, we use it on Nokia N8x0, but now we are moving to TI provided stuff for omap3
  • [16:40:22] <TangerinePentode> how is it?
  • [16:40:46] <felipec> TangerinePentode: what?
  • [16:41:19] <TangerinePentode> dspgateway
  • [16:41:47] <felipec> TangerinePentode: what do you mean?
  • [16:42:42] <TangerinePentode> does it work well, how heavy is it
  • [16:42:51] <TangerinePentode> do you like it
  • [16:43:56] * Jaymus (n=chatzill@S0106000a5e21e0bc.cg.shawcable.net) has joined #beagle
  • [16:44:16] <felipec> TangerinePentode: I don't have much experience with it, but I have a N810 and works well
  • [16:46:37] <TangerinePentode> i don't think it has been ported to omap3, but that's worth looking into
  • [16:47:59] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  • [16:49:15] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [16:54:52] <felipec> TangerinePentode: yeah, it would be interesting if somebody does that
  • [16:56:46] <TangerinePentode> it would be easier than writing my own os
  • [16:56:53] <koen> zuh: I there already sram support for omap in l-o git?
  • [16:57:00] <koen> s/I/is/
  • [16:57:47] <zuh> I don't know :)
  • [16:58:20] <koen> I stumbled upon http://gitweb.pingu.fi/?p=xf86-video-omapfb.git;a=commitdiff;h=16b5d9d232d09545c895c18a74480a536521fd69
  • [16:58:39] <koen> haven't tried it yet
  • [16:58:50] <Stskeeps> koen: kulve's over in #maemo
  • [16:59:06] <zuh> The comment still stands, it doesn't work at all on N800 at least
  • [16:59:12] <felipec> TangerinePentode: I guess so. But bascially I think TI has provided all the tools required
  • [16:59:14] <koen> Stskeeps: he's also over here
  • [16:59:22] <Stskeeps> ah
  • [16:59:32] <Stskeeps> irssi, don't see /names instantly ;)
  • [17:00:10] <TangerinePentode> felipec:the ti dsp is very complicated, and very closed
  • [17:00:32] <zuh> koen: I just found a comment in xomap that stated "we'd want to do this, but a kernel bug doesn't allow it". Thought I'd try it on beagle but haven't had the chance yet...
  • [17:01:20] <TangerinePentode> plus, I think it would be fun
  • [17:02:26] <TangerinePentode> mainly, I think it would be fun
  • [17:02:37] <zuh> koen: and this discussion too: https://wiki.maemo.org/Talk:Drivers_justification#5mbit_SRAM
  • [17:03:17] <felipec> TangerinePentode: I don't think it's so complicated, it's just that TI's sucky tools makes it look complicated
  • [17:06:10] <zuh> koen: btw, thanks for reminding me about fourcc.org, it inspired me to fix the video skew for funny src sizes :)
  • [17:06:10] <TangerinePentode> i've only used the tools for dm355 (davinci) and that is way too complicated
  • [17:09:41] <koen> TangerinePentode: the dm355 doesn't have a dsp
  • [17:10:03] <TangerinePentode> yeah, not exactly
  • [17:10:20] <TangerinePentode> but it still similar things
  • [17:10:43] <TangerinePentode> it still does similar things
  • [17:11:02] <koen> it has hardware engines, but no dsp
  • [17:11:06] * guillaum1 (n=gl@AMontsouris-153-1-70-121.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  • [17:11:07] <TangerinePentode> yeah
  • [17:11:22] * guillaum1 (n=Guillaum@AMontsouris-153-1-70-121.w90-2.abo.wanadoo.fr) has joined #beagle
  • [17:11:54] <TangerinePentode> but the interface is the same at user space
  • [17:11:56] * Jaymus (n=chatzill@S0106000a5e21e0bc.cg.shawcable.net) Quit (Read error: 110 (Connection timed out))
  • [17:13:19] * RogerMonk (n=a0740758@nat/ti/x-7de11702bb0c5c06) has left #beagle
  • [17:13:58] <koen> zuh: do you know of a basic algo to do yv12 -> 422?
  • [17:14:15] <koen> zuh: even having a slow implementation in the xv driver would be nice for testing
  • [17:14:51] <zuh> Yes, indeed. I'm actually looking at one, and thinking of writing something there based on that.
  • [17:15:26] * JoeyBorn (n=jborn@adsl-75-3-4-87.dsl.chcgil.sbcglobal.net) has joined #beagle
  • [17:15:56] <zuh> And then people with the skillz can go wild and do all sorts of arm assembly or whatever for it :)
  • [17:18:16] * valhalla (n=valhalla@81-174-25-222.dynamic.ngi.it) has joined #beagle
  • [17:19:08] * valhalla (n=valhalla@81-174-25-222.dynamic.ngi.it) Quit (Client Quit)
  • [17:21:29] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  • [17:28:17] <ds2> wonder if the N800 is using the SRAM for power management already
  • [17:29:33] <mru> koen: yuv420 to 422 is simple
  • [17:30:06] <mru> omapfbplay has a nice function to do it
  • [17:30:40] <koen> even in its own .S file :)
  • [17:31:56] <ds2> what is yuv420? I assume YUV422 means 4:2:2 ratios of Y:U:V but 0V??
  • [17:32:20] <koen> ds2: http://www.fourcc.org/yuv.php
  • [17:32:32] <nathanm> Broadcast terminology is very weird.
  • [17:32:59] <mru> nathanm: how so?
  • [17:33:13] * mru has been dabbling in broadcast for a while...
  • [17:34:47] <nathanm> Lines of resolution is my favorite. But the 422/420/411 thing too. If you come from the PC world it's weird.
  • [17:35:11] <mru> you mean line 525 lines really being 480?
  • [17:35:42] <mru> nothing weird about that
  • [17:35:52] <ds2> ah... so it does not mean you are tossing out the V component completely
  • [17:36:12] <mru> and it's very rare that anything but 420 is used, so nobody really talks about it
  • [17:36:18] <nathanm> More like a 240x240 image having 180 lines of resolution.
  • [17:36:34] <mru> huh?
  • [17:37:07] * dirk2 (n=dirk@F330a.f.strato-dslnet.de) has joined #beagle
  • [17:37:10] <koen> due to the subsampling?
  • [17:38:31] <ds2> from a math point of view 4-2-0 is a bit alarming... you are mapping 3 dimensions of data into 2 but in reality that is not what is happening
  • [17:38:32] <nathanm> All NTSC signals produce the same number of horizontal lines (525), so when comparing VHS to DVD to LD and so on, broadcast guys started taking the vertical column count instead -- but then converted it to horizontal lines again using the 4/3 aspect ratio. Thus, 640x480 = 480 lines, 320x480 = 240 lines, 240x480 = 180 lines, etc.
  • [17:40:00] <mru> I've never come across that terminology before
  • [17:40:15] <ds2> not all... 262.5 if it is a NI signal
  • [17:41:13] <nathanm> I'll blame video nerds arguing on usenet and not broadcast guys then.
  • [17:59:55] <nathanm> Still can't get my DSP reliable using DSP Bridge. Even after compiling out framebuffer support I still get those irq -33s after a few minutes (up from a few seconds).
  • [18:00:11] * JuanG (n=Juan@nat/ti/x-87c84a01508aec04) has joined #beagle
  • [18:00:16] * JuanG (n=Juan@nat/ti/x-87c84a01508aec04) has left #beagle
  • [18:00:28] <nathanm> I'll wait for Koen to figure it all out. :)
  • [18:02:51] * emeb (n=ericb@ip72-223-84-209.ph.ph.cox.net) has joined #beagle
  • [18:03:16] <TangerinePentode> nathanm: there was just something on the list about it
  • [18:04:49] <TangerinePentode> http://pastebin.com/d3c93ec6a
  • [18:05:45] <mru> strongly ordered means sloooow
  • [18:05:55] <mru> use it only as a last resort
  • [18:06:47] * odesus (n=OMAR@148.223.106.43) has joined #beagle
  • [18:07:05] <nathanm> At least I/O space doesn't get hammered quite as hard as memory.
  • [18:11:23] * JoeyBorn (n=jborn@adsl-75-3-4-87.dsl.chcgil.sbcglobal.net) Quit (Remote closed the connection)
  • [18:12:22] <Crofton|work> figuring out the irq thing would ge breat
  • [18:12:26] * JoeyBorn (n=jborn@adsl-75-3-4-87.dsl.chcgil.sbcglobal.net) has joined #beagle
  • [18:12:26] <Crofton|work> er be great
  • [18:17:54] <nathanm> It looks like it's more complicated to set up than just changing the omap34xx_io_desc, since I can't see how to set Strongly Ordered mode (Region Access Control Register B=0) in mmu.c/create_mapping.
  • [18:18:11] <nathanm> I hope somebody smarter is working on a patch.
  • [18:22:00] * JoeyBorn (n=jborn@adsl-75-3-4-87.dsl.chcgil.sbcglobal.net) Quit ("goin downtown")
  • [18:22:41] <Crofton|work> yeah
  • [18:22:55] * roxfan2 (n=dunno@2.214-200-80.adsl-dyn.isp.belgacom.be) has joined #beagle
  • [18:23:21] <Crofton|work> someone should bring this up on the l-o list ...
  • [18:23:45] <nathanm> I'll volunteer.
  • [18:24:55] <Crofton|work> thanks :)
  • [18:26:12] * gduncan (n=spec@dyn-bway-130-111.dyn.columbia.edu) has joined #beagle
  • [18:30:45] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [18:35:26] * roxfan (n=dunno@2.214-200-80.adsl-dyn.isp.belgacom.be) Quit (Read error: 110 (Connection timed out))
  • [18:36:19] * gcohler (n=demo@pool-72-93-108-183.bstnma.fios.verizon.net) has joined #beagle
  • [18:38:50] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [18:41:15] <nathanm> I'd happily run a miserably slow kernel if it meant I could stay up for more than a minute or two.
  • [18:41:33] <Crofton|work> what are you doing that aggavates the situation?
  • [18:43:37] <nathanm> Using DSP Bridge to ping data back and forth between the DSP and ARM.
  • [18:43:37] * woski (n=dleme@65.182.51.67) Quit (Read error: 104 (Connection reset by peer))
  • [18:44:16] * woski (n=dleme@65.182.51.67) has joined #beagle
  • [18:45:31] <nathanm> Felipe C also says that really hammering the framebuffer produces IRQ -33 pretty quickly.
  • [18:46:09] <gcohler> I've seen that too.
  • [18:47:14] <nathanm> I run for hours just using the ARM side from SSH, no FB or DSP. But as soon as I start the DSP the IRQ -33 pops up within a minute or two.
  • [18:47:22] <Crofton|work> ah
  • [18:47:51] <Crofton|work> I see it very rarely
  • [18:47:52] <gcohler> I don't use the DSP at all. But I can get iRQ-33's by doing mplayer and giving it a movie to play.
  • [18:49:06] <Crofton|work> should I test, or buy more SD cards and a whiteboard to track what card has what SW
  • [18:49:15] <nathanm> I used to get it on startup now and then (maybe due to X?), but I killed the framebuffer support (and X) and that went away.
  • [18:49:27] * gduncan (n=spec@dyn-bway-130-111.dyn.columbia.edu) Quit (Read error: 110 (Connection timed out))
  • [18:49:35] * odesus (n=OMAR@148.223.106.43) has left #beagle
  • [18:50:52] <gcohler> Yes, I think if you get rid of X you likely solve your problem. But if you want X, then we have a flaky situation. :)
  • [18:51:42] * JimDonova1 (n=jimd@c-66-30-221-249.hsd1.ma.comcast.net) has joined #beagle
  • [18:51:43] <felipec> nathanm: are you using the ping test app on the DSP?
  • [18:51:49] <nathanm> felipec: Yes.
  • [18:52:16] <nathanm> felipec: Also happens with the strmcopy test.
  • [18:56:55] * JimDonova1 (n=jimd@c-66-30-221-249.hsd1.ma.comcast.net) Quit ("Powered by OE: www.openembedded.org")
  • [18:58:32] * JimDonova1 (n=jimd@c-66-30-221-249.hsd1.ma.comcast.net) has joined #beagle
  • [19:00:08] <felipec> nathanm: I see, so it seems both the DSP and fb are doing something wrong, but the fb is doing it much worse?
  • [19:00:19] <felipec> er, DSP is doing it much worse
  • [19:02:21] * prpplague (n=dave@mail.americanmicrosystems.com) Quit (Remote closed the connection)
  • [19:02:53] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
  • [19:05:23] <nathanm> felipec: Seems that way. Since it's so reproducible for me now, and there is supposedly a workaround, I'm hoping I can get a patch to burn it in.
  • [19:06:11] <felipec> nathanm: yeap, but it would be better to track it down and fix it properly instead
  • [19:06:23] <felipec> anyway, time to go home
  • [19:06:31] * felipec (i=c0647cdb@gateway/web/ajax/mibbit.com/x-1979338ad20040d4) Quit ("mibbit.com: going home")
  • [19:07:56] * roxfan2 is now known as roxfan
  • [19:08:29] * emeb (n=ericb@ip72-223-84-209.ph.ph.cox.net) has left #beagle
  • [19:08:43] * emeb (n=ericb@ip72-223-84-209.ph.ph.cox.net) has joined #beagle
  • [19:08:56] * emeb (n=ericb@ip72-223-84-209.ph.ph.cox.net) has left #beagle
  • [19:09:32] * emeb (n=ericb@ip72-223-84-209.ph.ph.cox.net) has joined #beagle
  • [19:09:47] * gcohler (n=demo@pool-72-93-108-183.bstnma.fios.verizon.net) Quit ("Powered by OE: www.openembedded.org")
  • [19:11:17] * JoeyBorn (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) has joined #beagle
  • [19:16:27] * JimDonova1 (n=jimd@c-66-30-221-249.hsd1.ma.comcast.net) Quit (Remote closed the connection)
  • [19:17:54] * JimDonova1 (n=jimd@c-66-30-221-249.hsd1.ma.comcast.net) has joined #beagle
  • [19:19:03] * gcohler (n=gcohler@pool-72-93-108-183.bstnma.fios.verizon.net) has joined #beagle
  • [19:23:17] * richardw (n=richardw@nat/ti/x-71a799e2368496c7) has joined #beagle
  • [19:23:31] * [X]Spot (n=stancho@78.90.115.116) Quit (Read error: 113 (No route to host))
  • [19:24:26] * garren (n=garren@dsl-243-99-185.telkomadsl.co.za) has joined #beagle
  • [19:27:49] <garren> lo all
  • [19:31:20] * JimDonova1 (n=jimd@c-66-30-221-249.hsd1.ma.comcast.net) Quit ("Powered by OE: www.openembedded.org")
  • [19:33:51] * JimDonova1 (n=jimd@c-66-30-221-249.hsd1.ma.comcast.net) has joined #beagle
  • [19:33:52] * gcohler (n=gcohler@pool-72-93-108-183.bstnma.fios.verizon.net) Quit ("Powered by OE: www.openembedded.org")
  • [19:34:13] * JimDonova1 (n=jimd@c-66-30-221-249.hsd1.ma.comcast.net) Quit (Client Quit)
  • [19:34:32] * markit (n=marco@88-149-177-66.static.ngi.it) has joined #beagle
  • [19:35:49] * gcohler (n=demo@pool-72-93-108-183.bstnma.fios.verizon.net) has joined #beagle
  • [19:37:14] <doog> is there another env var besides TMPDIR which might stop bitbake builds from putting anything in /tmp?
  • [19:37:16] * JoeyBorn (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) Quit (Remote closed the connection)
  • [19:38:42] * JoeyBorn (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) has joined #beagle
  • [19:38:45] <doog> I'm building a VMware image for bb dev and my small root partition needed to be larger because some .s,.c,.o,.la files showed up in /tmp.
  • [19:39:20] * JimDonova1 (n=jimd@c-66-30-221-249.hsd1.ma.comcast.net) has joined #beagle
  • [19:39:37] <Crofton> doog, are the temporary files from gcc?
  • [19:39:47] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit (Read error: 104 (Connection reset by peer))
  • [19:40:24] <doog> Crofton: "cccrwBy3.o ccEFClQV.o cceus9wx.s" do look like they could be from gcc.
  • [19:40:36] <Crofton> yes
  • [19:41:04] <doog> I just checked gcc and TMPDIR is for gcc temp files so that's not it.
  • [19:42:11] * gcohler (n=demo@pool-72-93-108-183.bstnma.fios.verizon.net) Quit (Remote closed the connection)
  • [19:42:34] * gcohler (n=demo@pool-72-93-108-183.bstnma.fios.verizon.net) has joined #beagle
  • [19:42:34] * JimDonova1 (n=jimd@c-66-30-221-249.hsd1.ma.comcast.net) has left #beagle
  • [19:42:56] * JimDonova1 (n=jimd@c-66-30-221-249.hsd1.ma.comcast.net) has joined #beagle
  • [19:43:40] <Crofton|work> doog, I am 99% certain they are temporary files created by gcc during compilation
  • [19:44:24] * [X]Spot (n=stancho@78.90.115.116) has joined #beagle
  • [19:44:36] * [X]Spot (n=stancho@78.90.115.116) has left #beagle
  • [19:44:42] * [X]Spot (n=stancho@78.90.115.116) has joined #beagle
  • [19:46:11] * hli (i=chaton@vig91-2-82-232-97-149.fbx.proxad.net) has joined #beagle
  • [19:46:12] <doog> Crofton|work: they sure do look like it. export TMPDIR is set in local.conf so I'm not sure why it's not getting seen all the time. I'll look further and see if setting TMP helps since it used for another OS and might have been poorly mixed.
  • [19:48:09] * richardw (n=richardw@nat/ti/x-71a799e2368496c7) Quit (Remote closed the connection)
  • [19:49:52] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [19:51:41] <doog> Crofton|work: it wasn't exported in local.conf just sourced there so I exported TMPDIR in profile.sh. Should do the trick, testing now.
  • [19:55:12] <doog> Crofton|work: nope, still getting them. Will try setting TMP, otherwise I'll have to tie the root partition to the OE_HOME partition as a hack.
  • [20:04:41] * slothlove (n=srussell@user-387ggve.cable.mindspring.com) Quit (Read error: 110 (Connection timed out))
  • [20:04:57] * dirk2 (n=dirk@F330a.f.strato-dslnet.de) has left #beagle
  • [20:06:07] * igor321 (n=igor@141-241.dsl.iskon.hr) has joined #beagle
  • [20:08:12] * gcohler (n=demo@pool-72-93-108-183.bstnma.fios.verizon.net) Quit (Read error: 113 (No route to host))
  • [20:13:25] <nathanm> Crofton|work: Pratheesh is right. Strongly Ordered I/O eliminates my reproducible IRQ -33 problems with the DSP. It is freakishly stable now.
  • [20:15:08] * koen wonders if S-O is a fix or workaround
  • [20:16:49] * gduncan (n=spec@static-71-249-182-7.nycmny.east.verizon.net) has joined #beagle
  • [20:17:12] * mru suspects workaround
  • [20:22:32] * garren (n=garren@dsl-243-99-185.telkomadsl.co.za) Quit ("Ex-Chat")
  • [20:22:34] <nathanm> Pratheesh's post on BeagleBoard list made it sounds like a workaround to me.
  • [20:22:51] <nathanm> I posted the patch on linux-omap in case anyone else wants to try it. I am very happy to have my DSP programs running non-stop for 30 minutes now.
  • [20:29:24] * Administrator__ (n=chatzill@80.121.193.143) has joined #beagle
  • [20:29:42] * Administrator__ is now known as Dafcok
  • [20:31:17] * igor321 (n=igor@141-241.dsl.iskon.hr) Quit ("leavin'")
  • [20:32:45] * JimDonova1 (n=jimd@c-66-30-221-249.hsd1.ma.comcast.net) has left #beagle
  • [20:36:08] * thephilosopher (n=chatzill@modemcable108.47-58-74.mc.videotron.ca) has joined #beagle
  • [20:36:13] * thephilosopher (n=chatzill@modemcable108.47-58-74.mc.videotron.ca) Quit (Client Quit)
  • [20:44:04] * docelic (n=docelic@78.134.200.56) has joined #beagle
  • [20:47:51] * gduncan (n=spec@static-71-249-182-7.nycmny.east.verizon.net) Quit ()
  • [20:48:11] * zodttd|away is now known as zodttd
  • [20:49:10] * gduncan (n=spec@static-71-249-182-7.nycmny.east.verizon.net) has joined #beagle
  • [20:52:02] <doog> Crofton|work: FYI, nothing worked(TMP,TMPDIR,TEMP) so I'll see if I get another lockup as is, otherwise link /tmp to $OE_HOME/tmp
  • [20:52:28] <Crofton|work> doog, the stuff appearing in tmp is not related to OE
  • [20:52:37] <Crofton|work> try looking at the GCC docs
  • [20:53:37] <doog> Crofton|work: I thought it might be bitbake scripts not setting TMPDIR. definately GCC/compiling stuff.
  • [20:55:31] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  • [20:58:10] * feig (n=ejf3@244.sub-75-193-18.myvzw.com) Quit ("Konversation terminated!")
  • [21:00:08] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [21:00:10] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit (Remote closed the connection)
  • [21:00:48] * BThompson (n=BThompso@nat/ti/x-8bf4f78ef1e6e68c) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [21:01:03] * gcohler (n=demo@pool-72-93-108-183.bstnma.fios.verizon.net) has joined #beagle
  • [21:01:49] * gcohler (n=demo@pool-72-93-108-183.bstnma.fios.verizon.net) Quit (Client Quit)
  • [21:02:02] * gduncan (n=spec@static-71-249-182-7.nycmny.east.verizon.net) Quit ()
  • [21:04:51] * Dafcok (n=chatzill@80.121.193.143) Quit (Read error: 104 (Connection reset by peer))
  • [21:12:54] <nathanm> koen: Richard W at TI thinks the memory change has no performance side effects. I haven't noticed anything broken or laggy on my kernel yet, but I haven't even enabled FB.
  • [21:17:33] * zodttd_ (n=zodttd@user-142gtg7.cable.mindspring.com) has joined #beagle
  • [21:22:17] <Crofton|work> The Beagle Intern blog is cracking me up
  • [21:22:33] * markit (n=marco@88-149-177-66.static.ngi.it) Quit (Remote closed the connection)
  • [21:22:42] <Crofton|work> "My boss plays hockey"
  • [21:30:30] <nathanm> Hehe. BeagleIntern assumes 'git' is a mean name-calling word because it shows up on IRC so much.
  • [21:33:41] * zodttd_ is now known as zodttd|away
  • [21:34:58] * cbrake is now known as cbrake_away
  • [21:36:00] * zodttd (n=zodttd@user-142gtg7.cable.mindspring.com) Quit (Read error: 110 (Connection timed out))
  • [21:40:04] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) has joined #beagle
  • [21:43:17] * gduncan (n=spec@static-71-249-182-7.nycmny.east.verizon.net) has joined #beagle
  • [21:44:01] * gduncan (n=spec@static-71-249-182-7.nycmny.east.verizon.net) Quit (Client Quit)
  • [21:45:53] <Crofton|work> what's the new rootdelay setting?
  • [21:46:27] <Crofton|work> ah rootwait
  • [21:47:13] * zodttd|away is now known as zodttd
  • [21:48:06] * zodttd (n=zodttd@user-142gtg7.cable.mindspring.com) Quit ("Bye!")
  • [21:48:18] <Crofton|work> and following the mmc card partitioning page worked for me :)
  • [21:48:25] * zodttd (n=zodttd@user-142gtg7.cable.mindspring.com) has joined #beagle
  • [21:50:55] * guillaum1 (n=Guillaum@AMontsouris-153-1-70-121.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  • [21:51:13] * guillaum1 (n=gl@AMontsouris-153-1-70-121.w90-2.abo.wanadoo.fr) has joined #beagle
  • [21:51:53] * nathanm (n=nathan@adsl-69-106-228-95.dsl.pltn13.pacbell.net) Quit (Read error: 113 (No route to host))
  • [21:52:41] * nathanm (n=nathan@adsl-69-106-228-95.dsl.pltn13.pacbell.net) has joined #beagle
  • [21:57:26] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
  • [22:11:36] * gduncan (n=spec@cpe-67-243-144-218.nyc.res.rr.com) has joined #beagle
  • [22:12:40] * dcramer (n=davec@dcdsl.ebox.com) Quit (Read error: 104 (Connection reset by peer))
  • [22:13:32] * dcramer (n=davec@dcdsl.ebox.com) has joined #beagle
  • [22:14:46] * JoeyBorn (n=jborn@dsl017-022-247.chi1.dsl.speakeasy.net) Quit (Read error: 110 (Connection timed out))
  • [22:19:06] * gcohler (n=gcohler@pool-72-93-108-183.bstnma.fios.verizon.net) has joined #beagle
  • [22:21:08] * docelic (n=docelic@78.134.200.56) Quit ("http://www.spinlocksolutions.com/")
  • [22:24:27] * dcordes_ (n=dcordes@unaffiliated/dcordes) has joined #beagle
  • [22:26:51] * hli (i=chaton@vig91-2-82-232-97-149.fbx.proxad.net) Quit ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
  • [22:30:23] * gduncan (n=spec@cpe-67-243-144-218.nyc.res.rr.com) Quit (Read error: 110 (Connection timed out))
  • [22:36:37] * dcordes (n=dcordes@unaffiliated/dcordes) Quit (Read error: 110 (Connection timed out))
  • [22:43:10] * TAKI (n=Administ@dslb-088-074-055-212.pools.arcor-ip.net) Quit (Read error: 104 (Connection reset by peer))
  • [22:45:43] <TangerinePentode> so, it sounds like, according to the list, the irq -33 problem has a patch and dspbridge works
  • [22:46:57] * N208L1 (n=danson@rrcs-24-153-204-174.sw.biz.rr.com) Quit (Read error: 110 (Connection timed out))
  • [22:49:34] <Crofton|work> yes
  • [23:01:21] * gerrynjr (n=gerrynjr@gentoo/user/gerrynjr) has joined #beagle
  • [23:05:24] * robclark (n=robclark@nat/ti/x-af12463ebb0cb0a5) Quit ()
  • [23:05:34] <gcohler> Where do we find the patch for irq-33? I missed the earlier part of the chat.
  • [23:05:44] * zodttd is now known as zodttd|away
  • [23:05:49] <Crofton|work> on the l-o list
  • [23:06:24] <Crofton|work> do you need a link?
  • [23:06:28] * john280z (n=johnm@user-0c2h5av.cable.mindspring.com) has joined #beagle
  • [23:06:33] <Crofton|work> are you hitting the porblem?
  • [23:07:22] <gcohler> Yes. Everytime I run either xmms or mplayer, it runs for a while (anywhere from seconds to an hour), and then eventually the system goes into an infinite loop of irq-33's and I have to power down. A link would be great.
  • [23:07:31] <Crofton|work> hang on
  • [23:09:59] * jrmuizel (n=jrmuizel@mozca02.ca.mozilla.com) Quit ()
  • [23:14:30] <Crofton|work> http://article.gmane.org/gmane.linux.ports.arm.omap/12657
  • [23:14:46] <Crofton|work> I've had bad time getting patches out of gmane ....
  • [23:16:15] <gcohler> That's awesome. I'll give it a try. I haven't done a patch with Bitbake before. Do I just add it to the linux- omap_git.bb file as another patch to include for beagleboard? If so, I think I can do that. Thanks very much.
  • [23:17:10] <gcohler> BTW, I just finally was able to get on IRC. And I noticed that you are Philip. So thank you for all the help you have been. This is quite a journey this open-source. :-)
  • [23:17:18] <Crofton|work> heh
  • [23:17:48] <Crofton|work> I've been doing open source as a user or developer since the early days of Linux
  • [23:19:18] <gcohler> I was a Unix guy in the 70s and 80s. But I've been away for 15 or so years. So all this git and patching is new.
  • [23:20:07] <Crofton|work> cool
  • [23:20:32] <Crofton|work> We all have our 733t irc nicks
  • [23:20:45] <TangerinePentode> indeed
  • [23:20:50] <Crofton|work> my mother lived in "Crofton Park"
  • [23:21:30] <Crofton|work> gcohler, it should be clear how to add the patch to the bb file
  • [23:21:44] <Crofton|work> but, I have had whitespace issues getting stuff from gmane
  • [23:22:29] <gcohler> I'll give it a try tomorrow. Thanks for all the help. Time to sign off for now. But I'm glad I dropped in before shutting down.
  • [23:22:42] * gcohler (n=gcohler@pool-72-93-108-183.bstnma.fios.verizon.net) has left #beagle
  • [23:23:15] * zodttd|away is now known as zodttd
  • [23:25:08] <TangerinePentode> wow, that's old school
  • [23:28:25] * roxfan2 (n=dunno@46.53-136-217.adsl-dyn.isp.belgacom.be) has joined #beagle
  • [23:33:47] * _apkII (n=kows@81-208-83-210.fastres.net) has joined #beagle
  • [23:37:52] * roxfan (n=dunno@2.214-200-80.adsl-dyn.isp.belgacom.be) Quit (Read error: 110 (Connection timed out))
  • [23:41:07] * jrmuizel (n=jrmuizel@CPE001f5be79d0f-CM0017ee62f8b0.cpe.net.cable.rogers.com) has joined #beagle
  • [23:42:02] * kows__ (n=kows@85-18-136-98.fastres.net) has joined #beagle
  • [23:42:41] * _apk (n=kows@85-18-136-98.fastres.net) Quit (Read error: 104 (Connection reset by peer))
  • [23:47:12] * Olipro_ (n=Olipro@uncyclopedia/Olipro) has joined #beagle
  • [23:59:47] * _apkII (n=kows@81-208-83-210.fastres.net) Quit (Read error: 110 (Connection timed out))