• [00:20:18] * jkridner|work (n=a0321898@nat/ti/x-3168126a7345449d) has joined #beagle
  • [00:37:47] * RogerMonk (n=a0740758@nat/ti/x-3d6645879850e34a) Quit (Remote closed the connection)
  • [00:40:43] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) Quit ()
  • [00:44:06] * RogerMonk (n=a0740758@nat/ti/x-d20e54867d1aed7e) has joined #beagle
  • [02:58:11] * RogerMonk (n=a0740758@nat/ti/x-d20e54867d1aed7e) Quit (Remote closed the connection)
  • [02:59:51] * RogerMonk (n=a0740758@nat/ti/x-b29fbd5114ec44c6) has joined #beagle
  • [03:57:19] * lgentili (n=lgentili@host19.190-224-236.telecom.net.ar) has joined #beagle
  • [03:58:21] <lgentili> hello
  • [03:58:30] <NishanthMenon> lgentili: hello there
  • [03:59:06] <lgentili> I'm from Argentina, where are you from NishanthMenon
  • [03:59:52] <lgentili> It's almost 1:00 am here
  • [03:59:58] <NishanthMenon> lgentili: US. is there something about Beagle board you'd like to talk about?
  • [04:00:16] <lgentili> sure, Im just trying to be friendly :P
  • [04:00:22] <lgentili> ok going to the grain
  • [04:00:27] <lgentili> I want to buy a board :P
  • [04:00:33] <lgentili> haha
  • [04:01:40] <lgentili> Is there any change to put my hands over an omap35XX and linux, w/o to wait couple of months?
  • [04:01:47] <NishanthMenon> lgentili: i think the boards are on beta rt now. I am not sure about the wide spread availability yet. jkridner might be the right person to comment on timelines for general availability.
  • [04:02:07] <lgentili> ok, they just hang here, or they chat?
  • [04:02:19] <lgentili> maybe is late in USA i think
  • [04:02:40] <jkridner> it is late, but I can tell you where to get an OMAP35x board with Linux.
  • [04:02:46] <jkridner> http://www.ti.com/omap35x
  • [04:02:59] <lgentili> :(
  • [04:03:04] <jkridner> Beagles are expected in June.
  • [04:03:18] <NishanthMenon> lgentili: yep it is late here.. I am kinda messing around with U-boot V2 rt now.. http://www.beagleboard.org/irclogs/ has old chat logs.. i think there has been a bit of talks on beagle requests.
  • [04:03:22] <jkridner> no need to wait on the EVM if you can pay the $1499.
  • [04:03:51] <jkridner> BTW, we are solidifying what we will flash on the boards right now.
  • [04:04:11] <jkridner> I'm saying it should be x-loader + u-boot 1.3.2 (no kernel)
  • [04:04:13] <jkridner> thoughts?
  • [04:04:16] * NishanthMenon has woken up ;)
  • [04:04:17] <lgentili> jkridner, $1499 is too high for a hobbist like me
  • [04:04:43] <jkridner> oh, and no USBTTY in the next round of boards. :(
  • [04:04:55] <NishanthMenon> lgentili: u can look at zoom mdk -> but i think even that is not yet on general availability.
  • [04:05:11] <NishanthMenon> jkridner: ouch -> musb not available on next rev?
  • [04:05:14] <jkridner> I think that we can live with that for this next round and add USBTTY after that.
  • [04:05:29] <jkridner> MUSB support in Linux, but not u-boot.
  • [04:05:34] <lgentili> yep, if logicpd do the same like the IMX31... it will be ready in like 6 months
  • [04:06:27] <jkridner> I had wanted to support USBTTY in u-boot to eliminate the need for the serial cable.
  • [04:06:42] <NishanthMenon> lgentili: http://www.omapzoom.org/wiki/tiki-index.php?page=BuyingYourOwnZoomMdk
  • [04:06:54] <lgentili> jkridner, why there is no LCD interface in the board?
  • [04:06:58] <jkridner> musb support is already in Linux and will be supported in the next round of boards.
  • [04:07:08] <jkridner> b/c it is easy to add one externally.
  • [04:07:24] <jkridner> Beagle is the "bring your own peripherals" board.
  • [04:07:53] <NishanthMenon> jkridner: it part of my U-Boot V2 plans. but i want to get a proof of concept at the moment - step 1: SRAM boot UART, step 2: SRAM+MUSB step 3: flash support in SDRAM+ previous supports
  • [04:08:01] <lgentili> yep, but how do you connect a LCD with digital interface, not DVI or SVGA
  • [04:08:03] <jkridner> There is a DVI-D port and an S-Video port.
  • [04:08:15] <jkridner> DVI-D is digital.
  • [04:08:36] <lgentili> true, but cant connect a openframe LCD like I would like to
  • [04:08:51] <jkridner> I must really get familiar with U-Boot v2.
  • [04:09:07] <jkridner> what do you mean by "openframe"?
  • [04:09:41] <lgentili> I mean a LCD w/o TCON (timming controller o any controller board)
  • [04:09:48] <jkridner> Beagle is really meant for software developers more than hardware developers. It does offer expansion for A/D + D/A converters and a few other things...
  • [04:10:15] <jkridner> but the point is to provide a low-cost option for software development.
  • [04:10:24] <jkridner> Think of Beagle as a computer.
  • [04:10:30] <lgentili> I think still is a great idea, but would love to add just a connector with 20 LCD pins :)
  • [04:11:06] <lgentili> It would get a broader applications ... just with 20 pins more :)
  • [04:11:09] <NishanthMenon> jkridner: http://focus.ti.com/pdfs/wtbu/swpu114g.pdf page 3759 - you can kick out x-loader altogether :D
  • [04:11:10] <jkridner> If you really want to add it, use the DVI-D connection and convert it back to parallel.
  • [04:11:31] <lgentili> Yes, I need to study that option
  • [04:11:34] <NishanthMenon> jkridner: for a nand boot.
  • [04:12:20] * NishanthMenon is a x-loader hater
  • [04:12:30] <jkridner> fantastic.
  • [04:12:35] <jkridner> I'm opening the doc now.
  • [04:13:06] <jkridner> that is just the OMAP34xx TRM.
  • [04:13:31] <lgentili> jkridner, are you a hardware guy right?
  • [04:13:31] <jkridner> NM, flipping pages
  • [04:13:57] <NishanthMenon> jkridner = driving force behind beagle - wiz kid of h/w + s/w :D
  • [04:14:01] <lgentili> haha
  • [04:14:28] <jkridner> ah, bootloader support for SDRAM configs!
  • [04:14:41] <jkridner> and flash.
  • [04:14:48] <NishanthMenon> jkridner: yep..
  • [04:15:22] <jkridner> sweet. So, when will you have a conversion script for u-boot.bin for Beagle? :)
  • [04:15:46] <NishanthMenon> jkridner: lol... not me.. conflict....
  • [04:17:39] <lgentili> Is the omap3 already in the kernel mainline?
  • [04:18:00] <NishanthMenon> #define kernel mainline linus tree?
  • [04:18:17] <lgentili> yes
  • [04:18:46] <lgentili> Sorry, for my sucky english, I mean if I downlaod a vanilla kernel from kernel.org
  • [04:19:22] <lgentili> is there any omap3 support?
  • [04:19:49] <NishanthMenon> lgentili: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=tree;f=arch/arm/configs;h=d362f5efb21dcd78e907815820d45585d3d11cd6;hb=HEAD
  • [04:19:49] <jkridner> CH (configuration header) is on page 3438 of the OMAP3503 TRM (SPRUF98a).
  • [04:20:43] <NishanthMenon> lgentili: 1610 and p2 are there.. not even 2420 :( i guess it is blocked somewhere between tony->russel->andrew->linus
  • [04:20:48] <jkridner> for the latest OMAP3 boards, stuff is still just getting out on linux-omap git.
  • [04:20:58] <lgentili> I don't find any beagle_defconf :(
  • [04:21:16] <lgentili> thanks for the link anyway
  • [04:21:30] <jkridner> http://source.mvista.com/git
  • [04:21:49] * BThompson (n=BThompso@cpe-76-183-86-15.tx.res.rr.com) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [04:21:55] <lgentili> I think russel is not happy, because omap has a different newsgroup :P
  • [04:22:37] * NishanthMenon not looking at the latest revolt on arm kernel mailing list (nico,rmk,et.al.)
  • [04:23:26] <lgentili> Im following close the arm mailinglist
  • [04:25:34] * NishanthMenon out for chow...
  • [04:26:00] <lgentili> yesterday I ran a test to a S3C2440 board at 400Mhz, and I still can't belive how slow run QTopia 4
  • [04:26:09] <lgentili> I got like 4 fps with the demos :(
  • [04:26:14] <lgentili> it sucks
  • [04:26:47] <lgentili> jkridner did you tried QT in beagle?
  • [04:27:26] <jkridner> I haven't tried QT on Beagle yet.
  • [04:27:39] <jkridner> Hildon, yes.
  • [04:28:01] <jkridner> And matchbox via OE.
  • [04:28:29] <lgentili> I haven't tried OE yet
  • [04:29:05] <lgentili> but I will soon
  • [04:29:29] <lgentili> There are alot of people talking good about OE
  • [04:30:52] <lgentili> How many ppl worked in the beagle development?
  • [04:30:59] <lgentili> I mean hw
  • [04:32:32] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [04:37:56] <jkridner> hi khasim
  • [04:38:34] <jkridner> Khasim, have you looked at all into using config headers to avoid x-loader?
  • [04:39:39] <jkridner> lgentili, gerald is the "father of the beagle". He did all of the hardware design work. several people have contributed to various hardware aspects, but I don't know exactly how many.
  • [04:56:14] <khasim> Jason: regarding x-loader, no I have not looked into using config headers.
  • [04:56:36] <khasim> Will look into this some time...
  • [04:56:45] <jkridner> k. NishanthMenon mentioned them here.
  • [04:57:09] <NishanthMenon> khasim: it is kind of omap3 support -> kinda neat, u can do clock configs, SDRAM and i think even gpmc configs all as part of the header itself. there is a size constraint on MMC, but nand -> no issues
  • [04:57:22] <jkridner> doesn't seem to be critical. I was soliciting any feedback on how we flashed the boards for production.
  • [04:58:08] * NishanthMenon puts disclaimer: as far as i understand.. if boiled eggs pops out.. blame elsewhere ;)
  • [04:58:52] <jkridner> nope, you brought the idea up. you will be to blame. :)
  • [04:59:05] <NishanthMenon> why do i think i have heard it before :D
  • [04:59:17] <lgentili> I'm taking a look at flickr, really nice pics.
  • [04:59:45] <sakoman> khasim: hi! have you done any work with the SND_OMAP_SOC stuff in the linux-omap tree?
  • [05:00:03] <sakoman> do you know if there is any chance of it actually working?
  • [05:00:12] <jkridner> I'm hoping koen|dk will come back from vacation and take some new pics. :)
  • [05:00:32] <lgentili> who is who? alot grey hair there, nice
  • [05:01:21] <jkridner> sakoman: is SND_OMAP_SOC ALSA or ALSA_SOC?
  • [05:01:30] * jkridner is digging through the tree now.
  • [05:01:51] <sakoman> it is also soc
  • [05:02:01] <sakoman> alsa :-)
  • [05:02:43] <sakoman> Doing a build with it enabled just for grins
  • [05:04:41] <NishanthMenon> sakoman: i think some one pushed a t2 patch for audio alsa in linux-omap rt?
  • [05:05:24] <NishanthMenon> enabling alsa support might be a silly matter of adding a device entry in board file...
  • [05:06:58] * jkridner is exposing his ignorance again, but...
  • [05:07:23] <jkridner> why is there both sound/arm/omap and sound/soc/omap ?
  • [05:08:22] <NishanthMenon> sakoman: i was wrong! sound/arm/omap/ does not have t2 :(
  • [05:08:41] <jkridner> http://mailman.alsa-project.org/pipermail/alsa-devel/2007-May/001031.html
  • [05:09:05] <lgentili> because that kind of mess is not in the mainline kernel :P
  • [05:10:16] <NishanthMenon> jkridner: if my recollection is right... sound/arm/omap/ is for t2,aic,tsc2101 and the board specific ones.. the soc specific wrappers are in sound/soc/omap. probably khasim can correct me.. last time i looked at sound was when i was doing oss drivers..
  • [05:12:17] <jkridner> funny naming scheme, but that seems to be what is going on. sound/arm/omap is for board-level stuff and sound/soc/omap is for device-level stuff.
  • [05:12:55] <NishanthMenon> time to prove myself wrong... well.. omap-alsa-dma.c does not belong in sound/arm/omap/... ok i am confused now..
  • [05:12:57] <jkridner> of course, the n810 has its own entry in sound/soc/omap. :(
  • [05:13:25] <lgentili> haha
  • [05:14:44] <jkridner> NishanthMenon: your name is all over omap-alsa-dma.c. didn't you put it in sound/arm/omap ? ;)
  • [05:15:33] <NishanthMenon> nope.. blame it from sound/oss/omap-audio-dma-intfc.c
  • [05:15:59] <jkridner> sakoman: it is clear that the TWL4030 code is not in linux-omap git.
  • [05:16:22] <NishanthMenon> i think the first version of alsa was put in my iNDT brazil. i think they did for aic and tsc2101..
  • [05:16:26] <sakoman> yeah, just noticed that
  • [05:16:34] <NishanthMenon> lost track of things later on.. :(
  • [05:17:07] <jkridner> sound/arm/omap/omap2-audio-twl4030.c is clearly required.
  • [05:17:31] <NishanthMenon> jkridner: s/omap2-audio/omap-audio/g
  • [05:17:59] <jkridner> good call. just looking at what the file is called in my current source tree.
  • [05:18:35] <NishanthMenon> jkridner: viola.. google is my best friend: looks like omap-audio-twl4030.c is already in TI tree! http://linux.omap.com/pipermail/linux-omap-open-source/2007-January/008926.html
  • [05:19:59] <jkridner> I don't see /drivers/sound in linux-omap.
  • [05:22:43] <NishanthMenon> drat.. bad thought.. that was oss.. darn.. my own reply!! bah koen has it in the boot log.. http://www.flickr.com/photos/koenkooi/2346626459/ ALSA device list: #0: TWL4030 -> beagle code here http://code.google.com/p/beagleboard/downloads/list needs to be converted to patches to linux-omap i guess
  • [05:23:03] * NishanthMenon needs sleep :(
  • [05:24:37] <khasim> Sakoman, SOC was done by mistral
  • [05:24:45] <khasim> I can get you more info if required.
  • [05:24:55] <khasim> But we are mainly looking at ALSA
  • [05:27:46] <ds2> shouldn't the existing ALSA driver work with the beagle?
  • [05:27:59] <sakoman> khasim: Hmm . . . I'm confused! I'm referring to the ALSA for SoC audio support
  • [05:35:35] <jkridner> ALSA should work, but I do not think there is ALSA_SoC support in the TI release.
  • [05:35:44] <sakoman> NishanthMenon: Koen's boot log on flickr is the 2.6.22 kernel from Mistral
  • [05:35:50] <jkridner> I don't see either on the linux-omap git, though I would expect them to be there.
  • [05:36:05] <ds2> hmmm
  • [05:36:34] <jkridner> khasim, do you know where the SOC code from Mistral is posted?
  • [05:38:24] <sakoman> jkridner: looks like the stuff they did for 2.6.22 isn't SoC Alsa
  • [05:38:40] <sakoman> I'm looking at the source tree that ships with the EVM
  • [05:39:16] <jkridner> TI did the source tree for the EVM. Mistral did a port as well, but TI didn't pull all support from that port.
  • [05:39:20] <sakoman> the 4030 audio stuff is in sound/arm/omap
  • [05:39:55] <sakoman> I get confused about who did what -- the EVM has Mistral all over it so Ijust assume they did it :-)
  • [05:39:59] <jkridner> I believe khasim was saying that Mistral had done ALSA SoC, but that TI didn't use it. is that right, khasim?
  • [05:40:23] <jkridner> they did the board, more or less. :)
  • [05:40:36] <jkridner> the board is theirs and they did a base port for it.
  • [05:40:53] <ds2> the sound driver for the SDP should work, shouldn't it?
  • [05:40:55] <jkridner> TI had that to start with, but didn't deliver it as-is.
  • [05:41:55] <jkridner> ds2: I believe there are some GPIO tweaks, but khasim would know. it is awful that I bug him this much. :)
  • [05:42:38] <NishanthMenon> http://linux.omap.com/pub/kernel/3430zoom/linux-ldp-v1.0b.tar.gz has ./arm/omap/omap2-audio-twl4030.h and ./arm/omap/omap2-audio-twl4030.c
  • [05:43:08] <ds2> interesting, the linux-omap tree is missing the twl4030 driver
  • [05:43:44] <sakoman> I'll look at moving it over as part of my EVM patch series
  • [05:44:19] <sakoman> Though the patches aren't getting any comments, so maybe they'll just sit there forever :-(
  • [05:44:38] <NishanthMenon> sakoman: if you do.. please rename omap2 with omap
  • [05:44:55] <sakoman> OK
  • [05:45:31] <ds2> ah I see why...the TI drops have it
  • [05:45:43] <jkridner> need to get those folks on linux-omap to buy some EVMs.
  • [05:45:56] <ds2> heh
  • [05:46:18] <NishanthMenon> jkridner: SDP3430, EVM, zoom MDK, beagle.. tons of choices :)
  • [05:46:21] <ds2> buy? doesn't tony do enough to deserve a donation?
  • [05:46:24] <ds2> :)
  • [05:46:45] <jkridner> ds2: did you ever track down one? I know a few were headed that direction.
  • [05:47:05] <jkridner> I was hoping to save a few bucks and send him a Beagle, but it is clear we need to break this logjam.
  • [05:47:18] <jkridner> I'll drop him a note and see what I can do.
  • [05:47:45] <ds2> jkridner: K told me you guys sorted it out? I been buried in debugging some of the WTBU stuff ;)
  • [05:49:00] <jkridner> ds2: I wasn't sure it was sorted out, but I knew some boards where headed in the general direction. Couldn't nail down the exact recipients.
  • [05:51:21] <ds2> jkridner: It seems like it
  • [06:06:19] <lgentili> gnite all
  • [06:06:26] * lgentili (n=lgentili@host19.190-224-236.telecom.net.ar) Quit (Remote closed the connection)
  • [06:09:10] * RogerMon1 (n=a0740758@nat/ti/x-ab155e7a4c5cc606) has joined #beagle
  • [06:13:19] * RogerMonk (n=a0740758@nat/ti/x-b29fbd5114ec44c6) Quit (Remote closed the connection)
  • [06:44:59] * RogerMon1 (n=a0740758@nat/ti/x-ab155e7a4c5cc606) Quit (Remote closed the connection)
  • [06:52:22] * PibbRelay (n=supybot@nat/janrain/x-056f6c461e4a6d42) has joined #beagle
  • [06:53:04] * RogerMonk (n=a0740758@nat/ti/x-7ceb18d25c35827b) has joined #beagle
  • [06:55:09] <jkridner|work> welcome back PibbRelay
  • [10:37:27] <koen|dk> test kernel for usb problems: http://amethyst.openembedded.net/~koen/beagleboard/usbtest/
  • [10:51:22] * likewise (n=likewise@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [11:15:51] <Crofton> koen|dk, what is in that kernel?
  • [11:16:12] <Crofton> I am off to pick up a usb sound and network dongles in a bit
  • [11:22:40] <koen|dk> Crofton: latest git + felipes timeout patch
  • [11:22:58] <koen|dk> Crofton: or "whatever OE .dev builds" :)
  • [11:23:03] <koen|dk> everything's checked in
  • [11:23:07] <Crofton> ok
  • [11:23:22] <Crofton> I will run it in a bit
  • [11:23:27] <Crofton> or play GtA IV
  • [11:23:31] <koen|dk> usb is modular now, so we can pass in the overcurrent stuff
  • [11:23:43] * koen|dk will be on the beach Real Soon(TM)
  • [11:24:05] <Crofton> I really should give people my phone number so if I drop off irc they can call and tell me to get back to work ....
  • [11:24:30] <Crofton> in America we have have real beaches, with sand and warm water
  • [11:24:35] <Crofton> at least on the east coast
  • [11:24:48] <Crofton> sadly, the money is no longer real :(
  • [11:25:21] * koen|dk whistles innocently
  • [11:25:24] <Crofton> very funny :)
  • [11:25:38] <NishanthMenon> Crofton: dallas=no sea,no sand,lots of sun
  • [11:27:03] * Crofton is about 6-10 hours from real beaches
  • [11:27:39] <NishanthMenon> Houston is 3 hrs.. i feel better already ;)
  • [11:29:59] <Crofton> this is the problem with America, everything is so far apart :)
  • [11:31:11] <NishanthMenon> generic question: if i want an UI which will talk serial port.. would eclipse be better or mozilla
  • [11:31:27] <NishanthMenon> should need to work on linux and windows
  • [11:31:52] * NishanthMenon knows koen hates mozilla :D
  • [11:34:46] <koen|dk> webkti ftw
  • [11:34:50] <koen|dk> webkit*
  • [11:42:40] <likewise> koen|dk: have fun at the beach!
  • [12:25:13] <jkridner> koen|dk: what is this overcurrent stuff? (oh, and enjoy the beach!)
  • [12:28:11] * Crofton is building the latest kernel from koen's over night commits
  • [12:28:17] <Crofton> and heads to play gta iv
  • [12:29:26] <koen|dk> bah, stupid wether
  • [12:29:29] <koen|dk> +a
  • [12:29:31] <koen|dk> jkridner: http://marc.info/?l=linux-omap&m=120954707313789&w=2
  • [12:33:21] * RogerMonk (n=a0740758@nat/ti/x-7ceb18d25c35827b) Quit (Remote closed the connection)
  • [12:46:32] * bazbell (n=a0192809@nat/ti/x-39e2e3cb24fc42c1) Quit (Remote closed the connection)
  • [13:16:23] <Crofton> usb 1-1: new high speed USB device using ehci-omap and address 2
  • [13:16:23] <Crofton> usb 1-1: configuration #1 chosen from 1 choice
  • [13:16:23] <Crofton> usb 1-1: usbfs: USBDEVFS_CONTROL failed cmd usrper rqt 64 rq 160 len 4 ret -71
  • [13:16:35] <Crofton> usb host is flaky still
  • [13:17:04] <Crofton> Tony committed some clock stuff that involved usb
  • [13:18:24] <koen|dk> could you post your error message and dmesg to linux-omap?
  • [13:19:17] <Crofton> ok
  • [13:19:36] <Crofton> let me bump git rev
  • [13:19:39] <koen|dk> it seems we all get different error messages
  • [13:20:45] <koen|dk> bump srcrev to what?
  • [13:20:49] <koen|dk> it already is at HEAD ;)
  • [13:20:58] <Crofton> Tony pushed some stuff last night
  • [13:21:03] <Crofton> not anymore
  • [13:21:06] <Crofton> :)
  • [13:21:59] * koen|dk tries a "are you sure?"
  • [13:22:04] <Crofton> not sure if it will help
  • [13:22:19] <Crofton> http://source.mvista.com/git/gitweb.cgi?p=linux-omap-2.6.git;a=summary
  • [13:22:40] <Crofton> you had bumped to get the LL hack
  • [13:24:04] <koen|dk> http://lists.linuxtogo.org/pipermail/openembedded-commits/2008-April/014352.html
  • [13:24:14] <koen|dk> 59586cd959b31f91f29cf96c1d3f8ff52c3c0607 is still HEAD :)
  • [13:26:13] <Crofton> I have the 4bb9.... rev in my kernel bb file ...
  • [13:26:18] * bazbell (n=a0192809@nat/ti/x-76f6d345606ec856) has joined #beagle
  • [13:26:29] <koen|dk> mtn sync ; mtn up
  • [13:30:59] <koen|dk> Crofton: I also found out why people use local branches so much in git
  • [13:31:20] <koen|dk> it's the only way to do changes without resorting to quilt
  • [13:31:27] * koen|dk resorts to quilt from now on
  • [13:31:38] <Crofton> yeah
  • [13:31:46] <Crofton> dirk does his patches with quilt
  • [13:31:56] <koen|dk> it's easier to branch, reapply diffs than to try to do a revert
  • [13:32:45] <koen|dk> I need to figure out if git-format-patch can work on diffs instead of commits
  • [13:48:13] * BThompson (n=BThompso@nat/ti/x-30dd2c7e6d6588eb) has joined #beagle
  • [14:05:16] <jkridner|work> btw, it seems the pull-up on the over-current detect pin might not be configured.
  • [14:05:42] <jkridner|work> that might be the reason for the USB instability, but that is just wild speculation at the moment.
  • [14:06:24] <jkridner|work> if you are actively hacking, feel free to look at the GPIO settings.
  • [14:06:52] <jkridner|work> T2 pin N12, GPIO.1
  • [14:27:06] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
  • [15:35:41] <koen|dk> Crofton: did you test with 'modprobe ehci_hcd ignore_oc=1' and a powered hub?
  • [15:45:12] <koen|dk> jkridner: 2.6.22 treats the tfp410 as an lcd, right?
  • [15:45:26] <jkridner> yes.
  • [15:47:04] * koen|dk wonders how much of sakomans http://www.sakoman.net/omap3/user.collection/packages/linux/linux-omap3evm/lcd_omap3evm.c can be reused
  • [15:48:18] <sakoman> koen: that link is a bit old. The latest lcd-omap3evm.c was submitted as a patch to linux-omap this morning
  • [15:48:42] <koen|dk> ah, right
  • [15:48:59] <koen|dk> the question about reusing still stands :)
  • [15:49:14] <sakoman> I haven't updated the web stuff cause things are just changing too fast
  • [15:49:49] <sakoman> Will probably submit touchscreen support later today followed (hopefully) by audio
  • [15:49:52] <koen|dk> but the first step would be to get the omapfb driver to recognize the 3530 fb
  • [15:50:05] <koen|dk> sakoman: does the evm also use twl4030 audio?
  • [15:50:12] <sakoman> yes
  • [15:50:19] <koen|dk> nice
  • [15:50:32] <sakoman> with my lcd file, you will definitley get a linux fb
  • [15:50:50] <sakoman> Runs x11 images just fine (even with touchscreen)
  • [15:51:53] <sakoman> you'll need to tweak for resolution, timings, etc
  • [15:52:05] <koen|dk> and s/153/170/g ::)
  • [15:52:45] <koen|dk> part of the original code does this: http://rafb.net/p/Y70eQi71.html
  • [15:53:10] <koen|dk> and according to beagle hwrm gpio 170 is tfp410 power
  • [15:56:13] <koen|dk> sakoman: but I'm unsure how to hook into the lcd_panel struct
  • [15:56:28] <koen|dk> 2.6.22 is really different from current git
  • [15:56:51] <sakoman> yeah it sure is! even more so on audio :-(
  • [15:57:23] <koen|dk> it's not as horrific as some vendeor kernel I've seen, though
  • [15:57:50] <sakoman> Are there any other boards that do something close to what you need. An example is always nice :-)
  • [15:58:21] <koen|dk> afaik beagle is the only one with a tfp410 chip to drive dvi
  • [15:59:45] <koen|dk> maybe one of the davince evms has one
  • [16:00:18] <koen|dk> I'm afraid we'll need big X to make use of EDID, which means xorg.conf hell :(
  • [16:03:04] <sakoman> When I get a beagleboard I'd be happy to help
  • [16:03:11] <sakoman> tough without hw :-)
  • [16:03:17] * koen|dk discovers http://gitweb.freedesktop.org/?p=xorg/driver/xf86-video-intel.git;a=tree;h=3ce14d7676d3d8a3d294c267465279c9843078e3;hb=a0ced923bb793aa22e6bfbeeec0888d3b42ce176;f=src/tfp410
  • [16:03:24] <sakoman> Plenty to keep m ebusy on the EVM though
  • [16:03:30] <koen|dk> sakoman: wasn't dhl supposed to drop one of last week?
  • [16:03:39] <sakoman> delayed :-(
  • [16:05:01] <koen|dk> bummer
  • [16:07:08] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [16:07:40] <koen|dk> sakoman: let's hope tony applies your patches soon so we can copy stuff for beagle
  • [16:08:47] <sakoman> agreed, maintaining all those patches across revs till they are accepted could potentially be a real pain!
  • [16:09:27] <koen|dk> "inline conflict markers"
  • [16:09:46] <koen|dk> git is too rcs like for me
  • [16:09:56] * koen|dk hugs quilt
  • [16:10:59] <sakoman> I suppose I should learn to love quilt. I've never had to live in a patch based world before.
  • [16:12:23] <koen|dk> quilt isn't really for long term patch management
  • [16:12:36] <koen|dk> nothing can replaces pushing stuff upstream
  • [16:12:40] <koen|dk> -s
  • [16:19:15] * dirk2 (n=dirk@F33df.f.strato-dslnet.de) has joined #beagle
  • [16:23:05] <DJWillis> agreed, the patch cycle is proving a bit of fun ;-)
  • [16:25:28] * christian00 (i=WinNT@217-133-13-16.b2b.tiscali.it) Quit (Read error: 110 (Connection timed out))
  • [16:31:05] <prpplague> welcome to my world
  • [16:31:25] <prpplague> DJWillis: 90% of our drivers will never be accepted to the main tree
  • [16:31:35] <prpplague> DJWillis: so we have to maintain complete trees
  • [16:36:53] * NishanthMenon (n=gnat@nat/ti/x-5186dffd4ebc6a7a) Quit ("Download IceChat at www.icechat.net")
  • [16:37:07] <DJWillis> prpplague: yep, feels like my 2.4.old projects but this is a new take as there is no whole tree but just the patch sets off mainline.
  • [16:41:09] <sakoman> arghh. the twl4030 audio stuff is not going to move cleanly from 2.6.22
  • [16:41:44] <sakoman> lots of conflicts with the state of mcbsp stuff in git
  • [16:42:11] <sakoman> and several mcbsp patches ubmitted but not yet accepted
  • [16:42:21] <DJWillis> urggg
  • [16:42:49] <sakoman> I think I will move on to something else while that gets sorted out
  • [16:42:57] <DJWillis> Good plan ;-)
  • [16:43:12] <sakoman> grumble :-(
  • [16:44:24] <DJWillis> I really should get back into working on the EVM, makes a change from the day job and trying to work out wifi chips that are really GPL friendly.
  • [16:45:24] <sakoman> Would have been quite motivating to play some audio :-)
  • [16:46:11] <sakoman> Damn touch screen jitter is *really* annoying!
  • [16:46:57] <DJWillis> *cough* ;-), some tweaking with TSLib helps but not got to the bottom of that, something seems a little odd at the hardware level.
  • [16:47:19] <sakoman> yeah, I'm 99% convinced it is hw
  • [16:47:37] <DJWillis> Still on the TODO, is your driver still the same as it was when I grabbed the web stuff to work from?
  • [16:48:19] <sakoman> Very close -- I'll submit a patch to linux-omap later today
  • [16:48:44] <DJWillis> cool
  • [17:01:46] * jkridner|work (n=a0321898@nat/ti/x-3168126a7345449d) Quit (Remote closed the connection)
  • [17:34:58] * jkridner|work (n=a0321898@nat/ti/x-ab4d26de868115c4) has joined #beagle
  • [17:46:11] * ldesnogu (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) has joined #beagle
  • [17:46:18] <ldesnogu> hello
  • [17:51:27] <ldesnogu> in case some people are interested, Catalin Marinas (ARM kernel developer) confirmed me the better place to enable L2 is in the boot loader/monitor, so that the kernel remains generic enough
  • [18:05:29] * christian00 (i=WinNT@217-133-13-16.b2b.tiscali.it) has joined #beagle
  • [18:12:28] <jkridner|work> sakoman: let's talk TS jitter for a minute.
  • [18:12:42] <sakoman> OK
  • [18:12:49] <jkridner|work> I keep getting derailed, but I'm told to simply change the /etc/ts.conf settings.
  • [18:12:58] <sakoman> to what?
  • [18:13:11] <sakoman> I can try it easily
  • [18:13:21] <jkridner|work> pulling up the e-mail.
  • [18:13:30] <jkridner|work> computer is slow (doing a backup)
  • [18:14:00] <sakoman> currently have: module dejitter delta=100
  • [18:17:06] * NishanthM (n=Nishanth@cpe-24-175-68-203.tx.res.rr.com) has joined #beagle
  • [18:18:48] <sakoman> changing delta to 200 makes no noticable difference
  • [18:19:26] <jkridner|work> In /etc/ts.conf file: 1) Change ???dejitter delta??? from 100 to 800. 2) Change ???variance delta??? from 30 to 300.
  • [18:19:50] <jkridner|work> cancelled my backup... system still trying to come back to life. :(
  • [18:20:35] <sakoman> OK, one moment
  • [18:20:37] <jkridner|work> so, let me know if that makes it tolerable. if it does, I still won't be happy, but at least we have a watermark set.
  • [18:24:28] <sakoman> That makes it better -- still jumps in the y axis, but not nearly as bad
  • [18:24:57] <sakoman> I still believe that there is a hw issue. Two reasons:
  • [18:25:22] <sakoman> 1. The y jitter is *dramatically* worse than the x
  • [18:25:25] <DJWillis> :-o, that is mad values for the correction
  • [18:26:02] <sakoman> 2. I've worked with 2 other ts controllers and the default tslib values work beautifully (as they do for the x axis)
  • [18:26:24] <sakoman> DJWillis: agreed, those are absolutely insane values!
  • [18:28:09] <Crofton> ldesnogu, thanks for the info
  • [18:29:25] <NishanthM> sakoman: which ts on evm? what is the mclk+ what display is it? NEC/sharp?
  • [18:30:32] <sakoman> ts controller is ads7846
  • [18:30:50] <sakoman> don't know who the lcd vendor is
  • [18:30:51] <jkridner|work> sakoman: can you list off a competitive product that has better touchscreen performance? I want to set a benchmark with the team to beat.
  • [18:31:07] <sakoman> ucb1400
  • [18:31:15] <Crofton> koen|dk, FATAL: Module ehci_hcd not found.
  • [18:31:35] <sakoman> tsc2003 works but has brain dead hw design :-)
  • [18:32:07] <sakoman> interrupt logic is bizarre
  • [18:32:23] <NishanthM> i recollect one major difference when we moved from 2430 to 3430 -> the tsc2046 was the same, but they swapped the display from NEC to sharp. i had seen very bad noise in capture rate till we toned down from 1.5mhz to something like .96mhz or so..if my memory serves me right that is..
  • [18:32:51] <sakoman> phone call - back in a bit
  • [18:33:34] <NishanthM> i mean sdp...2430SDP->3430SDP..
  • [18:36:29] * dirk2 (n=dirk@F33df.f.strato-dslnet.de) has left #beagle
  • [18:45:51] <ds2> thought the 2430SDP had a different touchscreen chip
  • [18:47:13] <NishanthM> tsc2101 was with H4 1610/1710/2420 if i recollect.. thought we switched to tsc2046 when t2 came along.... am i confusing 3430 ES1 SDP Vs 3430 ES2 SDP?
  • [19:01:47] * Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) Quit (Read error: 113 (No route to host))
  • [19:04:00] * PibbRelay (n=supybot@nat/janrain/x-056f6c461e4a6d42) Quit (SendQ exceeded)
  • [19:05:02] <NishanthM> ds2: just pulled the kernel from linux.omap.com -> "arch/arm/configs/omap_2430sdp_2430_defconfig" CONFIG_OMAP_TSC2046_TS=y
  • [19:05:08] <sakoman> back now
  • [19:07:31] * Crofton (n=balister@66-207-66-26.black.dmt.ntelos.net) has joined #beagle
  • [19:09:54] <sakoman> TSC2046 is a next-generation version to the ADS7846
  • [19:10:57] <sakoman> allegedly pin & sw compatible
  • [19:12:40] <NishanthM> aaah.. interesting.. i dont know if the jitter has to do with the ADC or with the display resistive stuff
  • [19:13:19] * Crofton wonders if there is a trick to this gadget usb stuff
  • [19:13:24] <NishanthM> it was kinda interesting to see we had jitter when we moved displays..
  • [19:14:02] <NishanthM> sakoman: i am speaking for CSST perspective where we wrote SDP diagnostics.. not linux..
  • [19:15:02] <NishanthM> (assuming ads==tsc) u could try to reduce sampling rate by reducing mclk. spi if i am not mistaken
  • [19:16:02] <sakoman> It could be board layout, noise, component mis-load, or display
  • [19:16:22] <sakoman> Yes it is spi
  • [19:17:03] <sakoman> I'm tempted to break out a scope and probe it
  • [19:17:18] <sakoman> Too many other things to do right now though
  • [19:19:10] <sakoman> U7 is pretty accessible though so it looks like it would be easy to probe
  • [19:33:37] <Crofton> if I power the beagel through a hub, should I expect it to see USB speakers attached to that hub?
  • [19:43:50] <likewise> sakoman: are there drivers for the ads7846?
  • [19:44:33] <NishanthM> likewise: borrowing from sakoman's patch #include <linux/spi/ads7846.h>
  • [19:46:58] <sakoman> right -- and don't forget to enable the driver in defconfig :-)
  • [19:50:42] <NishanthM> sakoman: + .max_speed_hz = 1500000, probably might be able to improve that noise u see by reducing the frequency.. dont have access to sdp3430 values rt now.. but will post 2morrow once i reach office.. u can give it a shot and see..
  • [19:51:33] <sakoman> 1I'll take a look at the sdp3430 values and see if that makes a difference
  • [19:55:28] <likewise> sakoman, NishanthM: thanks, where do those patches go? (I'm new on the TI block).
  • [19:56:13] <sakoman> NishanthM: sdp3430 also sets .max_speed_hz = 1500000
  • [19:56:57] <sakoman> likewise: those patches apply against latest linux-omap from http://source.mvista.com/git/gitweb.cgi?p=linux-omap-2.6.git;a=summary
  • [20:27:16] <ds2> ah right the tsc2046 had 2 names
  • [21:01:36] <likewise> sakoman: Understood, and where is your patch itself available from?
  • [21:27:46] * ldesnogu (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) has left #beagle
  • [21:30:24] * jkridner|wor1 (n=a0321898@nat/ti/x-3e571d2d84d2084d) has joined #beagle
  • [21:40:44] * jkridner|work (n=a0321898@nat/ti/x-ab4d26de868115c4) Quit (Remote closed the connection)
  • [21:55:43] * erikg (n=erik@teach.laptop.org) has joined #beagle
  • [21:57:14] * erikg (n=erik@teach.laptop.org) has left #beagle
  • [21:57:18] * bazbell (n=a0192809@nat/ti/x-76f6d345606ec856) Quit (Remote closed the connection)
  • [22:04:53] * BThompson (n=BThompso@nat/ti/x-30dd2c7e6d6588eb) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [22:08:00] <sakoman> likewise: I posted the patches to the linux-omap mailing list. If you need a copy I could tar them up for you.
  • [22:21:39] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
  • [22:24:05] <likewise> sakoman: I am looking for a well supported touchscreen chip with power management functions ,but without audio. The ads7846 might be my plan B solution. Thanks, I'll find the driver on that list when I might need it.
  • [22:26:45] <sakoman> likewise: OK, good luck!
  • [22:34:56] * ds2 smacks the person responsible for a very irritating typo in the docs
  • [22:36:19] <ds2> TSC2004 is a valid chip and is not the same as a TSC2046!
  • [22:41:42] * BThompson (n=BThompso@cpe-76-183-86-15.tx.res.rr.com) has joined #beagle
  • [23:13:14] <jkridner|wor1> ds2: which doc?
  • [23:13:27] * jkridner|wor1 is now known as jkridner|work
  • [23:19:21] * RogerMonk (n=a0740758@nat/ti/x-89cd379d182bdc2a) has joined #beagle
  • [23:43:44] <ds2> jkridner: one of the Zoom docs