• [00:08:32] * Thanatos_desktop (n=rwvens@p578FEF95.dip.t-dialin.net) has joined #beagle
  • [00:16:32] <ds2> FWIW, I believe I just saw a serial port hang WITH the work around compiled in
  • [00:36:52] * cian (n=cian@cian.ws) Quit ()
  • [00:40:15] * dcordes (n=dcordes@unaffiliated/dcordes) has joined #beagle
  • [00:40:39] <dcordes> I just noticed my computer has no serial ports. what's the preferred way to get one? usb to serial?
  • [00:41:00] <[1]ianr> check the motherboard, there are sometimes serial headers hiding on there
  • [00:41:17] <[1]ianr> you use the same everex cable you do for the beagle
  • [00:42:29] <dcordes> I think I will get one adapter anyway
  • [00:42:42] <[1]ianr> yeah probably less effort than unscrewing a pc case !
  • [00:42:44] <dcordes> if I need to flash from a notebook I will need it
  • [00:44:06] <davr> usb to serial can be painfully slow, depending on how the serial is being used
  • [00:45:59] * cian (n=cianh@cian.ws) has joined #beagle
  • [00:49:40] <dcordes> [1]ianr: you're right. my mainboard has a 9 pin com port
  • [00:50:15] <dcordes> now all I need is a cable with two of those connectors
  • [00:52:48] <dcordes> how are those connectors called? I can't find any everx products
  • [00:53:15] <mru> I don't know what they're called
  • [00:53:54] <mru> when I asked for the idc to dsub cable in a shop, they just stared at me
  • [00:55:39] <dcordes> idc to dsub? what's that for?
  • [00:58:46] <[1]ianr> everex - they're on jason's validated h/w page, have you seen that?
  • [00:59:07] <[1]ianr> interesting idea an IDC - IDC null modem cable but I dont think they exist
  • [01:00:49] <mru> should be easy to make one
  • [01:03:10] <[1]ianr> @dcordes: basically you need two IDC-dsub cables, which gives the beagle and the PC standard 9-pin Dsub serial ports, then you need a null modem cable to connect them
  • [01:03:24] <ds2> arrggggg
  • [01:03:37] <ds2> anyone tried doing heavy I/O to the SDIO slot?
  • [01:04:47] <dcordes> [1]ianr: it's not possible to connect them directly?
  • [01:06:06] * Thanatos_desktop (n=rwvens@p578FEF95.dip.t-dialin.net) has left #beagle
  • [01:06:56] <mru> you could fashion an idc-idc null modem cable
  • [01:07:04] * RogerMonk (n=a0740758@nat/ti/x-b8144a10389b15b4) Quit (Remote closed the connection)
  • [01:08:25] <mru> you need to cross a few pairs
  • [01:09:09] <dcordes> mru: http://www1.conrad.de/xl/9000_9999/9700/9760/9762/976253_BB_00_FB.EPS.jpg two of those and a dsub cable will do?
  • [01:09:37] <mru> yes
  • [01:10:23] * Olipro (n=Olipro@unaffiliated/olipro) Quit (Read error: 104 (Connection reset by peer))
  • [01:10:32] * Olipro_ (i=Olipro@unaffiliated/olipro) has joined #beagle
  • [01:35:56] * rsalveti (n=salveti@200.184.118.132) Quit (Read error: 113 (No route to host))
  • [01:43:03] * rsalveti (n=salveti@189.70.119.44) has joined #beagle
  • [01:47:52] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) Quit (Read error: 104 (Connection reset by peer))
  • [01:55:29] * cian (n=cianh@cian.ws) Quit ()
  • [01:59:48] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [02:04:57] * dcordes (n=dcordes@unaffiliated/dcordes) Quit ("Lost terminal")
  • [02:11:36] * DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) Quit (Read error: 104 (Connection reset by peer))
  • [02:16:17] * Olipro_ (i=Olipro@unaffiliated/olipro) Quit (No route to host)
  • [02:40:12] * bazbell (n=a0192809@nat/ti/x-75ea7d06f44fb188) has joined #beagle
  • [02:52:54] * bazbell (n=a0192809@nat/ti/x-75ea7d06f44fb188) Quit (Remote closed the connection)
  • [02:54:37] * gulick (n=gulick@rrcs-208-125-50-2.nys.biz.rr.com) Quit (Read error: 110 (Connection timed out))
  • [03:28:34] * jeh (n=jeh@vpn-pub2.hard.ware.fi) has joined #beagle
  • [03:41:41] * bazbell (n=a0192809@nat/ti/x-0040c4317e0581cd) has joined #beagle
  • [03:59:44] * bazbell (n=a0192809@nat/ti/x-0040c4317e0581cd) Quit ("Leaving.")
  • [04:17:57] <sakoman> khasim: ping
  • [04:18:10] * jeh (n=jeh@vpn-pub2.hard.ware.fi) Quit (Read error: 60 (Operation timed out))
  • [04:30:23] * chakie_work (n=jeh@vpn-pub2.hard.ware.fi) Quit (Read error: 113 (No route to host))
  • [04:39:51] * jeh (n=jeh@vpn-pub2.hard.ware.fi) has joined #beagle
  • [04:53:41] <khasim> sakoman: r u there?
  • [04:53:51] <sakoman> khasim: yes
  • [04:53:57] <khasim> good
  • [04:54:26] <khasim> what is the host side s/w that you are running to get USB pheripheral boot?
  • [04:55:02] <sakoman> I've tried 2 different utilities. both work with Nishanth's uboot-v2 image
  • [04:55:44] <sakoman> first is:
  • [04:55:46] <khasim> I am just trying to understand your setup before going into issues ...
  • [04:55:46] <sakoman> TI OMAP3 USB boot ROM tool, version 0.1
  • [04:56:08] <sakoman> (c) 2008 Martin Mueller <martinmm@pfump.org>
  • [04:56:43] <khasim> so u-boot v2 is running when booted over USB but not x-load.bin
  • [04:56:50] <sakoman> correct
  • [04:57:02] <sakoman> work flawlessly
  • [04:57:31] <sakoman> x-load.bin loads without error, but then there is no sign of life
  • [04:57:50] * jeh_work (n=jeh@vpn-pub2.hard.ware.fi) has joined #beagle
  • [04:58:30] <sakoman> I sent you hex dumps of the binaries -- looks like Nishanth's is not signed
  • [04:58:44] <khasim> yeah I saw both the dumps
  • [04:58:56] <sakoman> Any ideas?
  • [04:59:13] <khasim> is this x-load.bin working when booted over MMC?
  • [04:59:29] <khasim> same thing signed and renamed as MLO
  • [04:59:38] <sakoman> yes. it works from nand and mmc
  • [04:59:56] <sakoman> (renamed MLO from mmc)
  • [05:00:34] <sakoman> I also tried your mlo/x-load with the same result. but I now realize they were signed
  • [05:01:11] <sakoman> have you had success doing this?
  • [05:02:19] <khasim> I havent tried USB peripheral boot.
  • [05:02:40] <khasim> I think it is going to exception some where
  • [05:03:20] <sakoman> Am I correct that it should be an unsigned image?
  • [05:03:33] <sakoman> The TRM is very contradictory on this point
  • [05:03:39] <khasim> I need to confirm this with ROM boot document
  • [05:03:57] <sakoman> I guess Nishanth's image is evidence that unsigned is correct
  • [05:04:17] <khasim> are you trying this on overo or Beagle?
  • [05:04:24] <sakoman> tried on both
  • [05:04:41] <sakoman> Beagle was a few weeks back
  • [05:05:09] * jeh_work (n=jeh@vpn-pub2.hard.ware.fi) Quit (Remote closed the connection)
  • [05:05:28] <sakoman> I think I tried EVM too
  • [05:06:21] <sakoman> I would expect that they all should behave the same since there sn't very much external hw involved (other that 4030 which is the same for all three)
  • [05:07:58] <khasim> since u-boot v2 is working with this setup, I only suspect some issues with x-loader source it selft
  • [05:08:11] <sakoman> could very well be
  • [05:08:12] <khasim> I am sure it is going for some exception
  • [05:08:48] <sakoman> do you know of any differences in the execution environment between nand/mmc and peripheral boot?
  • [05:08:50] <khasim> and this could happen if memory timings are corrupted or x-loader trying to access a peripheral
  • [05:09:26] <khasim> in x-loader you can comment out NAND and MMC booting funcitions. Just remove them from the board_init function
  • [05:09:43] <sakoman> i.e. should there be a difference in the x-load hw setup between nand/mmc and peripheral boot
  • [05:10:05] * jeh (n=jeh@vpn-pub2.hard.ware.fi) Quit (Read error: 110 (Connection timed out))
  • [05:10:07] <sakoman> or should the same pinmux, gpmc, etc setup be the same
  • [05:10:17] <khasim> sakoman can I get USB bootrom tool
  • [05:11:00] <sakoman> yes, I'll email it to you
  • [05:11:05] <khasim> sometimes x-loader assumes few things
  • [05:11:32] <khasim> and ROM Code doesnt flow through same instructions, it depends on the boot options
  • [05:11:49] <khasim> so it could be some thing that ROM Code missed and x-loader relying on it
  • [05:12:50] <khasim> If you can send me the ROM boot Hostside utility and u-boot v2 image that is booting, I will quickly run that here and will try to grab a lauterbach and see for the hang if possible.
  • [05:13:31] <khasim> It will take couple of hrs before I start working on this, so take rest till then :)
  • [05:14:22] <sakoman> OK, utility is on the way. Will send ubootv2 binary now
  • [05:16:19] <sakoman> you should have both of them now
  • [05:16:50] <sakoman> khasim: it is 10:15pm, so I will likely be sleeping in a couple of hours :-)
  • [05:22:56] <ds2> the beagle needs a connector for a lithium battery#!@#!@$@#@
  • [05:23:10] <ds2> so tired of resetting the date constantly
  • [05:23:51] <khasim> sakoman: got both the images. I will try it out on Beagle and get back to you
  • [05:24:11] <sakoman> khasim: thanks! I appreciate the help
  • [05:24:27] <sakoman> ds2: I use ntp to avoid that
  • [05:24:39] <ds2> sakoman: ntp to what?
  • [05:24:51] <sakoman> does require a usb/ethernet dongle
  • [05:25:26] <sakoman> then set up ntp to try pool.ntp.org for a time server
  • [05:25:26] <ds2> works better when one is not powering it from the USB port :)
  • [05:25:35] <sakoman> there is that :-)
  • [05:25:55] <khasim> sakoman: one dumb question, you run omap3dl x-load.bin == correct
  • [05:26:07] <sakoman> but in that case you at least could have a usb network connection so it could still work
  • [05:26:25] <sakoman> khasim: yes that is correct
  • [05:27:06] <sakoman> khasim: use sudo since that might be required for usb access
  • [05:27:12] <khasim> what is the sequence
  • [05:27:24] <sakoman> sequence should be:
  • [05:27:24] <khasim> first connect cable then run utility?
  • [05:27:34] <sakoman> power off, usb cable connected
  • [05:27:52] <sakoman> sudo omap3dl x-load.bin
  • [05:28:04] <sakoman> you will see periods being printed then
  • [05:28:20] <sakoman> power up the board and you will get messages from the utility
  • [05:28:43] <sakoman> should download successfully in a fraction of a second
  • [05:29:01] <sakoman> you should then see u-boot v2 prompt on serial
  • [05:29:16] <sakoman> or nothing at all if you downloaded x-load :-(
  • [05:29:30] <khasim> thanks sakoman, will follow this
  • [05:29:48] <sakoman> so I guess that sudo command should actually be for uboot-serial, not xload :-)
  • [05:30:11] <khasim> ok, will catch up with you in an hour..
  • [05:30:17] <khasim> leaving to office now.
  • [05:30:19] <sakoman> OK, thanks!
  • [05:36:42] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [05:47:01] * ap02 (n=abhijit@c-24-6-99-89.hsd1.ca.comcast.net) has joined #beagle
  • [06:01:18] * jeh_work (n=jeh@vpn-pub2.hard.ware.fi) has joined #beagle
  • [06:06:34] * a_min (n=ubuntu@pd95b4498.dip0.t-ipconnect.de) has joined #beagle
  • [06:07:46] * shrirama (i=3b6030f5@gateway/web/ajax/mibbit.com/x-5eae96aebcf25d78) has joined #beagle
  • [06:09:46] * jeh_work is now known as chakie_work
  • [06:29:36] * ap02 (n=abhijit@c-24-6-99-89.hsd1.ca.comcast.net) Quit ("Leaving")
  • [06:37:26] * shrirama (i=3b6030f5@gateway/web/ajax/mibbit.com/x-5eae96aebcf25d78) Quit ("http://www.mibbit.com ajax IRC Client")
  • [06:43:21] * trickie|work (n=trickie@basesoft.xs4all.nl) has joined #beagle
  • [06:50:01] <chakie_work> as a more or less total hw-n00b i could use some pointers for a small design issue
  • [06:50:37] <chakie_work> in a car environment i need to provide some simple means for user input for simple navigation
  • [06:51:48] <chakie_work> so something like arrow keys + an "ok" key would be more than sufficient. there even seems to be ready made small key pads that one could use or just hack up something own
  • [06:52:10] <chakie_work> however, getting the signals to the beagle is where i have no idea
  • [06:53:02] <chakie_work> another idea is to use a small generic remote control and some ir receiver, but that's a bit clumsy
  • [06:53:53] * Viral_Sachde (n=Viral_Sa@122.167.141.148) has joined #beagle
  • [07:05:28] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [07:06:49] * DJWillis (n=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) has joined #beagle
  • [07:07:26] * valhalla (n=valhalla@81-174-21-16.dynamic.ngi.it) has joined #beagle
  • [07:30:55] * [1]ianr (n=ianr@adsl-76-244-162-79.dsl.sndg02.sbcglobal.net) Quit (" HydraIRC -> http://www.hydrairc.com <- The alternative IRC client")
  • [07:32:33] <khasim> beagle board didnt hang for 20 hrs after removing C70
  • [07:35:51] * Olipro (i=Olipro@unaffiliated/olipro) has joined #beagle
  • [07:38:25] * Viral_Sachde (n=Viral_Sa@122.167.141.148) Quit ("Leaving.")
  • [08:21:45] <koen> good morning all
  • [08:21:53] <mru> morning
  • [08:31:44] <khasim> error: ??usb_dev_handle?? undeclared
  • [08:31:54] <khasim> I have installed lib-usb
  • [08:32:06] <khasim> http://sourceforge.net/project/showfiles.php?group_id=1674&package_id=277626&release_id=614530
  • [08:32:30] <khasim> I am trying to build the sources sakoman sent for USB host
  • [08:32:37] * guillaum1 (n=gl@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  • [08:32:39] <khasim> for peripheral boot
  • [08:33:44] * guillaum1 (n=gzba4143@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) has joined #beagle
  • [08:33:49] <guillaum1> hi all
  • [08:53:26] * calculu5 (n=calculus@gentoo/user/calculus) has joined #beagle
  • [08:54:59] * calculus (n=calculus@gentoo/user/calculus) Quit (Read error: 104 (Connection reset by peer))
  • [09:06:15] <koen> Persistence of Vision(tm) Ray Tracer Version 3.7.0.beta.25b (ccache arm-angstrom-linux-gnueabi-g++ -march=armv7-a -mtune=cortex-a8 -mfpu=neon -mfloat-abi=softfp 4.3.1 @ i686-pc-linux-gnu)
  • [09:06:40] <koen> it would probably run a lot faster on cortex-a9 with it's pipelined vfp3
  • [09:11:26] <khasim> *** was able to reproduce the problem sakoman reported, now have to fix the x-loader to boot on usb for beagle
  • [09:12:17] <koen> khasim: do you use sakomans x-load tree? http://www.sakoman.net/cgi-bin/gitweb.cgi?p=x-load-omap3.git;a=summary
  • [09:12:55] <khasim> no, I am just using my local copy for now, once this works may be I can try with GIT as well
  • [09:36:04] <koen> that directfb dsp thingy looks interesting
  • [09:37:26] <kulve> mru: did you start that X.Org display driver for omap3? I remember somebody planning it a while ago..
  • [09:50:02] <koen> kulve: you mean for the overlay?
  • [09:51:10] * ldesnogu (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) has joined #beagle
  • [09:51:22] <ldesnogu> hello
  • [09:52:32] <koen> hey ldesnogu
  • [09:53:58] <kulve> koen: I meant something that "accelerates" xrender stuff. Overlay would be the next step though (afaik, but I'm not an expert on that field)
  • [09:54:38] <koen> if you want to accelerate xrender some NEON stuff in pixman would probably help
  • [09:56:29] <ldesnogu> koen, it looks like DirectFB davinci stuff is using a dedicated interface between ARM and c64x CPU, isn't it ?
  • [09:56:56] <ldesnogu> at least it comes with its own kernel module
  • [09:57:16] <koen> and it patches 2.6.20-mvl
  • [09:57:19] <koen> not a good sign
  • [09:57:39] <kulve> koen: so, when are you adding that NEON stuff to pixman? ;)
  • [09:58:02] * koen isn't a coder
  • [09:58:31] <kulve> who made the neon stuff to ffmpeg?
  • [09:58:37] <ldesnogu> mru did
  • [09:58:39] <koen> mru:
  • [09:58:50] <koen> http://git.mansr.com/?p=ffmpeg.mru;a=shortlog;h=refs/heads/arm
  • [09:59:32] <kulve> right. So he's the right guy for the pixman ops too :)
  • [09:59:42] <koen> getting the neon colourspace conversion from http://git.mansr.com/?p=omapfbplay;a=blob;f=omapfbplay.c;h=3f82ed5d67546b851468d695e8fff1401feb781d;hb=HEAD into pixman or liboil would nice as well
  • [10:00:01] <ldesnogu> what is pixman?
  • [10:00:24] <koen> "Pixman contains lowlevel pixel manipulation routines and is used by
  • [10:00:25] <koen> both xorg and cairo."
  • [10:00:25] <suihkulokki> pixel-manipulation library for X and cairo
  • [10:01:39] <koen> ldesnogu: http://gitweb.freedesktop.org/?p=pixman.git;a=summary
  • [10:02:24] <koen> mru says "It is correct that I mentioned it
  • [10:02:24] <koen> in the past, and hinted that I might look into it since nobody else was
  • [10:02:24] <koen> doing it. First, however, we need to sort out the framebuffer and
  • [10:02:24] <koen> overlay drivers. Perhaps we should start by making a list of requirements
  • [10:02:25] <koen> for apps (including X) that will be using the kernel drivers."
  • [10:02:30] <ldesnogu> thanks
  • [10:02:41] * koen is playing firewall proxy
  • [10:03:15] <ldesnogu> can't mru login to IRC through beagle website?
  • [10:05:09] <kulve> 11:21 < mru> morning
  • [10:11:08] <ldesnogu> koen, does webkit use cairo?
  • [10:11:56] <koen> ldesnogu: the gtk port does
  • [10:12:13] <ldesnogu> that's what I just found, I should google before asking :-(
  • [10:16:33] <koen> making pixman faster on arm would benefit a lot of projects
  • [10:18:22] <kulve> yes it would.
  • [10:19:05] <koen> I read somewhere that patches for armv5te exist
  • [10:19:26] <koen> probably in some companies private repo that they aren't going to publish
  • [10:21:29] <Crofton> http://www.phdcomics.com/comics.php?n=1055
  • [10:22:02] <kulve> I would guess there are several operations that could be optimized, so even if there are some patches, it may not be a lot..
  • [10:22:57] <ldesnogu> wouldn't cairo also be a target for dedicated optimizations or does it rely on other libs (such as pixman) for costly operations?
  • [10:23:17] <kulve> We made some armv6 patch for pixman: https://bugs.freedesktop.org/show_bug.cgi?id=13445
  • [10:25:53] <koen> ldesnogu: cairo relies on pixman for most fallbacks
  • [10:26:07] <ldesnogu> kulve, if I may make a comment about that patch: for recent ARM processors long sequences of conditional instructions (especially memory ones) are not very good
  • [10:26:38] <chakie_work> kulve: you're at movial?
  • [10:26:41] <ldesnogu> koen, ok so to improve cairo speed, improving pixman is the way to go
  • [10:26:44] <kulve> chakie_work: yup
  • [10:27:11] <chakie_work> kulve: should have guessed :)
  • [10:27:21] <kulve> ldesnogu: feel free to comment on the bug (I'm no asm expert) :)
  • [10:27:29] <kulve> chakie_work: how so?
  • [10:27:51] <koen> ldesnogu: and when pixman gets faster, so will the xserver since that's using pixman for software fallbacks as well :)
  • [10:27:56] <ldesnogu> kulve, I know what I said is applicable to Cortex A9, I have no other platform to test :p
  • [10:28:10] <chakie_work> kulve: if a company in finland is involved with linux and/or embedded stuff it's usually movial or nokia
  • [10:28:16] <kulve> right.. I guess A9 is still quite rare? :)
  • [10:28:28] <ldesnogu> kulve, :)
  • [10:29:41] <koen> kulve: that patch seems to enable those assembler options for every arm architecture
  • [10:30:07] <koen> kulve: wouldn't that break armv4, armv4t and armv5te?
  • [10:30:30] <ldesnogu> it looks like it's at least armv5te compatible
  • [10:30:58] <ldesnogu> and v4 too
  • [10:31:08] <kulve> koen: I'm not sure what ARM that is. Based on the comments it might be armv5 then
  • [10:31:13] <kulve> well, v4..
  • [10:32:58] <bjdooks> that code looks v2 compatible
  • [10:33:06] <ldesnogu> kulve, I think this patch should be a full assembly file instead of mixed C and asm
  • [10:33:12] <ldesnogu> bjdooks, no it uses strh :)
  • [10:33:13] <bjdooks> oh no, there is one v4 instrictuin in it
  • [10:33:34] <kulve> ldesnogu: write a better one ;)
  • [10:33:37] <bjdooks> but then again, who has anything pre v4 around anyway
  • [10:33:50] <ldesnogu> bjdooks, ask rusell :-)
  • [10:33:57] <ldesnogu> russell*
  • [10:34:05] <bjdooks> I know he hasn't
  • [10:34:17] <bjdooks> i've got ARMv3, but it hasn't got linux on it atm
  • [10:34:21] <ldesnogu> kulve, it looks like there is no assembly file in pixman at all
  • [10:34:29] <bjdooks> and a StrongARM-110
  • [10:35:02] <ldesnogu> wasn't strongarm v4?
  • [10:35:37] <bjdooks> hmm, it is
  • [10:35:37] <bjdooks> CPU: StrongARM-110 [4401a104] revision 4 (ARMv4), cr=0000517f
  • [10:36:19] <ldesnogu> humpf that '44' kind of hurts
  • [10:36:27] * ldesnogu still remember Digital
  • [10:39:32] <bjdooks> right, new psu in and my amd64 has been running all morning
  • [10:44:07] * keesj (n=keesj@ip49-193-210-87.adsl2.static.versatel.nl) Quit (Read error: 110 (Connection timed out))
  • [10:44:10] * keesj (n=keesj@ip49-193-210-87.adsl2.static.versatel.nl) has joined #beagle
  • [10:49:42] <koen> there, my good deed for today: https://bugs.freedesktop.org/show_bug.cgi?id=13445
  • [10:55:03] <ldesnogu> there doesn't seem to be a dedicated mailing list for pixman
  • [10:55:41] <koen> ldesnogu: I see some pixman mails popping up on the cairo list from time to time
  • [10:55:58] <koen> but then again, cairo used to have an internal copy of it
  • [11:00:20] <khasim> In "board/omap3530beagle/config.mk" if I change TEXT_BASE to 0x80e80000 from 0x40200800 then the USB peripheral booting works fine for x-loader
  • [11:00:23] <ldesnogu> I am trying to see how pixman relates to xserver, any hint?
  • [11:00:58] <koen> ldesnogu: X uses it as software fallback for xrender operations iirc
  • [11:01:10] <ldesnogu> koen, OK thanks
  • [11:01:31] <koen> so if you run X on a dumb framebuffer you want a fast pixman :)
  • [11:01:44] <khasim> but then its just x-loader we need to add USB support to x-loader to download u-boot from x-loader
  • [11:02:12] <ldesnogu> phew I think I need to refresh my X understanding, I thought xrender was optional, I guess x11r5 is long gone :-)
  • [11:04:38] <kulve> it might be optional but I guess quite many libs (like cairo) uses it heavily
  • [11:04:50] <kulve> again, just guessing here..
  • [11:05:19] <kulve> maybe I should just shut up, and not guess on things I don't really know..
  • [11:06:34] <ldesnogu> kulve, you seem to be right, xrender is an extension of X
  • [11:16:00] * cian (n=cianh@cian.ws) has joined #beagle
  • [11:22:55] * docelic (n=docelic@78.134.198.146) has joined #beagle
  • [11:30:41] * Thanatos_desktop (n=rwvens@p578FEFA2.dip.t-dialin.net) has joined #beagle
  • [11:33:54] * cian (n=cianh@cian.ws) Quit ()
  • [11:46:17] * fer_luck (n=fer_luck@200.96.204.33) has joined #beagle
  • [11:49:02] * Thanatos_desktop (n=rwvens@p578FEFA2.dip.t-dialin.net) has left #beagle
  • [11:50:06] * gulick (n=gulick@rrcs-208-125-50-2.nys.biz.rr.com) has joined #beagle
  • [11:51:55] <jkridner> good morning
  • [11:52:15] <koen> hey jkridner
  • [11:53:11] * mib_qsby8w (i=d95b5d22@gateway/web/ajax/mibbit.com/x-5140041e17f7bb77) has joined #beagle
  • [11:53:33] * mib_qsby8w (i=d95b5d22@gateway/web/ajax/mibbit.com/x-5140041e17f7bb77) has left #beagle
  • [11:53:53] * de (i=d95b5d22@gateway/web/ajax/mibbit.com/x-72592b122bd9a84a) has joined #beagle
  • [11:56:15] <de> Has anyboday tried running Jalimo on beagle?
  • [11:57:27] <koen> de: the jalimo people are testing on beagleboards now :)
  • [11:57:49] <koen> Robert had phoneme foundation running yesterday
  • [11:58:37] <koen> angstrom has a ready to install package for it: http://www.angstrom-distribution.org/repo/?action=details&pnm=phoneme-advanced-foundation
  • [11:59:09] <de> Cool! Can I just install it in your demo image?
  • [12:00:27] <koen> yes
  • [12:00:46] <koen> 'opkg update ; opkg install phoneme-advanced-foundation'
  • [12:00:55] <koen> if it doesn't work, let Robert know :)
  • [12:05:42] <de> probably the vm is not included. is there also a package for the vm?
  • [12:06:29] * a_min (n=ubuntu@pd95b4498.dip0.t-ipconnect.de) Quit (Read error: 110 (Connection timed out))
  • [12:07:21] <fer_luck> howdy all.
  • [12:09:31] * dcramer (n=davec@dcdsl.ebox.com) Quit (Read error: 104 (Connection reset by peer))
  • [12:10:08] <koen> de: jamvm?
  • [12:10:11] * koen is a java n00b
  • [12:10:20] * dcramer (n=davec@dcdsl.ebox.com) has joined #beagle
  • [12:18:56] <koen> jkridner: did you already see http://ldn.linuxfoundation.org/book/how-participate-linux-community ?
  • [12:19:08] <de> koen: ah, jamvm looks fine. thanks.
  • [12:19:31] * de (i=d95b5d22@gateway/web/ajax/mibbit.com/x-72592b122bd9a84a) Quit ("http://www.mibbit.com ajax IRC Client")
  • [12:22:20] * fer_luck_ (n=fer_luck@200.96.204.33) has joined #beagle
  • [12:22:20] * fer_luck (n=fer_luck@200.96.204.33) Quit (Read error: 104 (Connection reset by peer))
  • [12:22:34] <jkridner> koen: yeah.
  • [12:22:42] <jkridner> very nice doc.
  • [12:22:53] <jkridner> it is making the rounds at TI.
  • [12:29:05] * BThompson (n=BThompso@nat/ti/x-b5e5e6b2cf9c7158) has joined #beagle
  • [12:34:47] * Thanatos_desktop (n=rwvens@p578FEFA2.dip.t-dialin.net) has joined #beagle
  • [12:54:40] * cbrake_away is now known as cbrake
  • [12:58:30] * min (n=ubuntu@pd95b4498.dip0.t-ipconnect.de) has joined #beagle
  • [13:07:17] * robclark (n=robclark@nat/ti/x-e37115e97c5cf9da) has joined #beagle
  • [13:09:56] <sakoman> khasim: good detective work! Thanks :-)
  • [13:10:13] * fer_luck_ (n=fer_luck@200.96.204.33) has left #beagle
  • [13:10:17] * fer_luck_ (n=fer_luck@200.96.204.33) has joined #beagle
  • [13:11:44] * dcordes (n=dcordes@unaffiliated/dcordes) has joined #beagle
  • [13:12:43] * JuanG (n=Juan@nat/ti/x-dd90a9235dbd247a) has joined #beagle
  • [13:12:47] * JuanG (n=Juan@nat/ti/x-dd90a9235dbd247a) has left #beagle
  • [13:28:22] <guillaum1> de: phoneme-advanced-foundation package provides a CDC VM
  • [13:28:52] * Thanatos_desktop (n=rwvens@p578FEFA2.dip.t-dialin.net) Quit (Read error: 145 (Connection timed out))
  • [13:32:29] <khasim> sakoman: how are you planning to use USB peripheral booting
  • [13:34:04] * prpplague (n=dave@mail.americanmicrosystems.com) has joined #beagle
  • [13:34:17] <khasim> sakoman: you need USB support in x-loader and u-boot, I think we have USB support in u-boot for EVM, but that is only to work with Linux Host machine
  • [13:42:04] * JuanG (n=Juan@nat/ti/x-78f03243cb943501) has joined #beagle
  • [13:42:36] * JuanG (n=Juan@nat/ti/x-78f03243cb943501) has left #beagle
  • [13:57:25] * min (n=ubuntu@pd95b4498.dip0.t-ipconnect.de) Quit (Remote closed the connection)
  • [14:13:54] <sakoman> khasim: eventually I would like be able to use it for automated test and initial onenand image writes
  • [14:14:07] <sakoman> but as you say that requires usb support
  • [14:14:12] * Thanatos_desktop (n=rwvens@p578FEFA2.dip.t-dialin.net) has joined #beagle
  • [14:14:24] <sakoman> this is just step 1 along the way
  • [14:15:02] <sakoman> which version of evm u-boot has the usb support?
  • [14:16:46] <khasim> I think 1.0 has the support in u-boot
  • [14:17:16] <khasim> Another good one from LDN http://ldn.linuxfoundation.org/lsb/check-your-app
  • [14:17:42] <bjdooks> right, i'm off for about a week
  • [14:18:57] <prpplague> bjdooks: yea, yea, we know you are going to rehab, don't try to hide it
  • [14:24:49] * rsalveti (n=salveti@189.70.119.44) Quit (Read error: 110 (Connection timed out))
  • [14:26:23] * Thanatos_desktop (n=rwvens@p578FEFA2.dip.t-dialin.net) has left #beagle
  • [14:28:33] * Navi_ (n=Navi@unaffiliated/navi) has joined #beagle
  • [14:30:35] <sakoman> khasim: just to confirm -- USB peripheral boot images are *unsigned* Correct?
  • [14:38:15] * Navi (n=Navi@unaffiliated/navi) Quit (Read error: 110 (Connection timed out))
  • [14:38:39] <sakoman> khasim: sadly, changing TEXT_BASE does *not* work for me
  • [14:41:22] * Beagle1 (n=Beagle1@CPE-58-165-122-37.qld.bigpond.net.au) has joined #beagle
  • [14:43:49] <khasim> sakoman: they are unsigned
  • [14:44:14] <khasim> sakoman: can you quickly try this on Beagle?
  • [14:45:47] <khasim> sakoman: I have mailed you my x-load.bin. Just try this out. In my version I have commented out MMC, but it should atleast give the starting messages
  • [14:45:58] <khasim> sakoman: have you done a clean build?
  • [14:46:06] <sakoman> khasim: I see that Nishanth is using a slightly different lowlevel_init
  • [14:46:44] <sakoman> I will email you what he is using
  • [14:47:04] <khasim> I have pulled a laste u-boot v2 tree
  • [14:48:07] <khasim> just check if beagle works for you, we can find the difference between overo and beagle
  • [14:48:14] * Beagle1 (n=Beagle1@CPE-58-165-122-37.qld.bigpond.net.au) Quit ()
  • [14:48:54] <sakoman> khasim: the image you sent does print the message, so perhaps something else hangs in my version of x-load since I still have nand/mmc enabled in that build
  • [14:49:31] <sakoman> I will send you my x-load.bin as well as the lowlevel init from u-boot-v2
  • [14:49:33] <khasim> I do have NAND enabled, only MMC is commented in default config
  • [14:50:29] <khasim> #define CFG_CMD_MMC 1
  • [14:52:25] <khasim> comment this macro
  • [14:55:40] <sakoman> khasim: I will try turning off MMC
  • [14:59:57] <khasim> I tried enabling MMC in my x-loader it still boots but waits for MMC - it works
  • [15:00:03] <khasim> Texas Instruments X-Loader 1.41
  • [15:00:03] <khasim> Starting on with MMC
  • [15:00:33] <khasim> are you using x-loader from code.google?
  • [15:02:09] <khasim> your binary image didnt boot on my board
  • [15:03:07] <sakoman> khasim: no I am using x-load from my git
  • [15:03:34] * rsalveti (n=salveti@200.184.118.132) has joined #beagle
  • [15:03:54] <sakoman> so clearly something in that code base is causing the hang
  • [15:04:09] <khasim> sakoman: ok, I think is shouldn't be lot different - do you have some time to take a diff and see or try code.google one.
  • [15:05:13] <sakoman> I will start doing diffs
  • [15:06:33] <sakoman> khasim: to answer your earlier question, all of my x-load builds are clean
  • [15:07:09] <khasim> ok, just a thought. I generally miss this step some times :)
  • [15:08:05] <sakoman> I've learned from bad experience, so I just do it by default. especially for small things like x-load where clean build is just a few seconds
  • [15:22:24] * trickie|work (n=trickie@basesoft.xs4all.nl) Quit ("Leaving")
  • [15:36:13] * jkridner|work (n=a0321898@nat/ti/x-ae3b30d234107489) has joined #beagle
  • [15:37:54] * RogerMonk (n=a0740758@nat/ti/x-e01cddfd9b409df8) has joined #beagle
  • [15:42:28] * GentooHag (n=haggis@cpc1-clyd1-0-0-cust187.renf.cable.ntl.com) has joined #beagle
  • [15:49:19] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [15:51:48] * uberfry (i=nooberfr@vodsl-10903.vo.lu) has joined #beagle
  • [15:51:53] <uberfry> hi everybody
  • [15:52:51] <uberfry> I don't mean to sound like an idiot, but is there any plan to make a beagle board with standard vga connection?
  • [15:53:04] <BThompson> not likely
  • [15:53:25] <uberfry> how about standard lcd connection?
  • [15:53:45] <uberfry> you know...like the one open-pandora uses
  • [15:53:51] <uberfry> or even psp
  • [15:55:33] * dust_t (n=wrongone@59.92.161.96) has joined #beagle
  • [15:58:38] <jkridner|work> open-pandora has an S-Video out, by my understanding.
  • [15:58:45] <jkridner|work> Beagle Board does as well.
  • [15:58:49] <jkridner|work> not VGA.
  • [15:59:36] <jkridner|work> VGA is possible to add to a design and possible to buy external converters, but there are no plans to do such on the Beagle Board.
  • [16:09:42] * RogerMon1 (n=a0740758@nat/ti/x-30bf1bfb93c4a7a4) has joined #beagle
  • [16:13:08] <uberfry> jkridner: I meant vga OR lcd output
  • [16:13:17] <uberfry> lcd as in psp lcd for instance
  • [16:15:55] <BThompson> the board does not have the connections for such small LCDs, only a display that supports DVI/HDMI digital video or S-Video will be possible, adding in the pinout for a LCD like on the PSP would be cost prohibitive for the board
  • [16:16:21] <uberfry> hmmm psp lcd costs 15USD in "retail"
  • [16:16:30] <uberfry> buy em in quantities and you can reach 10USD
  • [16:20:10] * RogerMonk (n=a0740758@nat/ti/x-e01cddfd9b409df8) Quit (Remote closed the connection)
  • [16:22:13] * wrongone_ (n=wrongone@59.96.41.127) has joined #beagle
  • [16:22:53] * guillaum1 (n=gzba4143@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) Quit ("Leaving.")
  • [16:23:14] * guillaum1 (n=gl@AMontsouris-153-1-67-145.w90-2.abo.wanadoo.fr) has joined #beagle
  • [16:24:28] * dust_t (n=wrongone@59.92.161.96) Quit (Read error: 110 (Connection timed out))
  • [16:41:23] * browser04 (n=treid@74.43.222.41) has joined #beagle
  • [16:43:34] <browser04> I have a question about the board - can both the dvi and s-video interfaces be used at the same time?
  • [16:43:57] <browser04> With different information displayed on each?
  • [16:45:30] <jkridner|work> browser04: yes, it is possible.
  • [16:46:21] <jkridner|work> in the diagnostic kernel, there are 3 windows: 1 graphics and 2 video.
  • [16:46:23] <jkridner|work> http://code.google.com/p/beagleboard/wiki/LinuxHints
  • [16:46:55] <jkridner|work> you can set each of the windows to display to either the 'tv' or 'lcd' (DVI-D).
  • [16:47:51] <browser04> Cool. Thanks. That makes this board very interesting for a project I am considering.
  • [16:48:15] <jkridner|work> uberfry: I don't believe there is a "standard" LCD connection. If there was a "standard", then it might be something to put on there for a broad market.
  • [16:48:37] <jkridner|work> browser04: the S-Video support has not made it into the linux-omap git kernel, as far as I know.
  • [16:49:26] <jkridner|work> a patch has been submitted, but it is not currently in the right condition to be merged into the git kernel.
  • [16:49:27] <browser04> Yeah - I've been reading the irc logs for a few days to get an idea of where things stand.
  • [16:50:15] <browser04> This means I will have to get my feet wet in embedded linux.
  • [16:50:47] <browser04> But I love trying new stuff, and I don't have a timeframe to complete this, so NP.
  • [16:55:06] <browser04> Thanks again for the info.
  • [16:58:30] * t_s_o (n=tso@173.84-49-129.nextgentel.com) has joined #beagle
  • [17:00:53] <Crofton> http://www.balister.org/~balister/shoulder-xray-dye/
  • [17:01:31] <jkridner|work> you share a lot of painful looking photos. you seem to hurt yourself often. :)
  • [17:01:42] <Crofton> stiil the smae wreck
  • [17:01:49] <Crofton> modern technology is grand
  • [17:06:16] <Crofton> I have a request to make an animated gif out of the mri shots :)
  • [17:11:04] * Navi_ is now known as Navi
  • [17:12:34] * Beagle1 (n=Beagle1@dyn.83-228-203-144.dsl.vtx.ch) has joined #beagle
  • [17:16:39] * Beagle1 is now known as sprotch
  • [17:17:02] <uberfry> so what have you guys done so far with your boards?
  • [17:17:47] <banderson> I take mine for a walk every morning...
  • [17:20:36] <uberfry> fat man made a funneh
  • [17:31:28] * wrongone_ (n=wrongone@59.96.41.127) Quit ("Leaving")
  • [17:32:50] * dirk2 (n=dirk@F323b.f.strato-dslnet.de) has joined #beagle
  • [17:48:32] <dirk2> keesj: Any JTAG news/progress?
  • [17:49:45] <prpplague> dirk2: we have some stuff in the works for jtag
  • [17:50:07] <prpplague> dirk2: hopefully have openocd working in the near future
  • [17:50:35] <dirk2> prpplague: :)
  • [17:56:07] <uberfry> does windows mobile run on it?
  • [17:57:08] <dirk2> I don't now a lot about Mamona, but just found http://blog.aloisiojr.com/?p=25 and read "supported: ... Omap3430sdp" so most probably this should work for Beagle, too (?)
  • [17:59:45] <dirk2> s/now/know/
  • [18:02:31] * sprotch (n=Beagle1@dyn.83-228-203-144.dsl.vtx.ch) Quit (Read error: 110 (Connection timed out))
  • [18:03:50] * Beagle1 (n=Beagle1@hunter.netmodule.com) has joined #beagle
  • [18:04:01] * Beagle1 is now known as sprotch
  • [18:08:33] <uberfry> anyone here used the gpios yet?
  • [18:08:44] <uberfry> within linux that is
  • [18:09:06] <ds2> what about the gpios?
  • [18:09:41] <uberfry> well how exactly do you use them?
  • [18:09:51] <ds2> you use them to drive a pin to a certain state
  • [18:10:00] <uberfry> but how?
  • [18:10:12] <ds2> what do you mean how? you hook up an input to a pin
  • [18:10:18] <uberfry> duh
  • [18:10:28] <uberfry> do you have some C code that outputs?
  • [18:10:38] <ds2> Oh you mean how do you control them in software
  • [18:10:44] <uberfry> yes :)
  • [18:10:46] <ds2> there should be examples in the existing code
  • [18:10:50] <ds2> let me check
  • [18:10:53] <uberfry> ok thanks
  • [18:11:50] * sprotch (n=Beagle1@hunter.netmodule.com) Quit ()
  • [18:12:04] * Beagle6 (n=Beagle6@hunter.netmodule.com) has joined #beagle
  • [18:12:08] <dirk2> uberfry: http://source.mvista.com/git/?p=linux-omap-2.6.git;a=blob;f=arch/arm/mach-omap2/board-omap3beagle.c;h=85119ce5aa1467c6782a41fa3491981edd89fa06;hb=HEAD line 142
  • [18:12:13] * Beagle6 is now known as sprotch
  • [18:12:20] <uberfry> thanks dude
  • [18:12:21] <ds2> look at the omap*gpio* functions in say, arch/arm/mach-omap2/board-omap3evm.c
  • [18:12:33] <ds2> the evm uses the same processor but has more things wired to it
  • [18:13:53] <ds2> that's just the hookup to the LEDs
  • [18:13:55] <uberfry> icic
  • [18:14:06] <ds2> you can control them with the omap*gpio* functions in your code
  • [18:14:39] <uberfry> looks kinda complicated
  • [18:14:47] <ds2> not really -
  • [18:14:57] <ds2> request it, set the direction, and set the state you want it to
  • [18:15:19] <uberfry> think I'll look at the low level code and mmap it myself
  • [18:15:34] <ds2> please don't
  • [18:15:44] <ds2> use the omap*gpio* macros/functions
  • [18:16:24] <uberfry> why shouldn't I?
  • [18:16:43] <ds2> those functions help prevent conflicts
  • [18:16:56] <uberfry> ahhh I see
  • [18:17:03] <ds2> so if you code is used by someone else later on, it can cause subtle bugs
  • [18:17:18] <uberfry> ultimately I want to use this for myself only
  • [18:17:28] <ds2> it maintains a list of who used what GPIO so if 2 drivers tries to do it, it will fail one of them
  • [18:17:30] <uberfry> put it in the car and monitor engine stuff
  • [18:17:41] <ds2> heh, OBDII stuff?
  • [18:17:41] <uberfry> ah ic
  • [18:17:46] <uberfry> what's OBDII?
  • [18:17:58] <uberfry> ah nvm
  • [18:18:00] <uberfry> yes kind of
  • [18:18:01] <ds2> that's a standard required by all US cars
  • [18:18:08] <uberfry> well I'm .eu ;)
  • [18:18:25] <ds2> a lot of EU cars also implement OBD II to make things easy on the mfg
  • [18:18:35] <uberfry> ah cool
  • [18:18:49] <uberfry> well does OBDII measure acceleration too?
  • [18:18:52] <uberfry> and cornering?
  • [18:19:17] <ds2> that, no. just engine parameters like speed, RPM, temperature at some place, throttle position, etc
  • [18:19:23] <uberfry> icic
  • [18:19:30] <prpplague> uberfry: only on vehicles installed with those options
  • [18:19:42] <uberfry> I'm getting a golf 3
  • [18:19:46] <ds2> you'd need an accelerometer to measure the stuff you are talking about
  • [18:19:50] <uberfry> yep
  • [18:20:01] <uberfry> 2 accelerometers
  • [18:20:08] <ds2> 2?
  • [18:20:13] <uberfry> why yes...
  • [18:20:14] <prpplague> uberfry: most large "18 wheeler" trucks in the US have a giant number of sensors
  • [18:20:30] <ds2> prpplague: those have a different standard
  • [18:20:36] <uberfry> one that measures acceleration and another that measures cornering
  • [18:20:39] <ds2> why 2? or you mean 2 axis only?
  • [18:20:45] * Beagle5 (n=Beagle5@dyn.83-228-203-144.dsl.vtx.ch) has joined #beagle
  • [18:20:56] <uberfry> well one for straight and one for sides
  • [18:21:03] <prpplague> ds2: you sure of that? it was my understanding it was based on OBDII
  • [18:21:08] <ds2> you should be able to do it with 1 3 axis sensor and derive them from math
  • [18:21:14] <ds2> let me find the name
  • [18:21:40] <ds2> browser crashed :(
  • [18:22:38] <uberfry> ;X
  • [18:22:39] <ds2> prpplague: http://en.wikipedia.org/wiki/J1587
  • [18:22:55] <ds2> unless there is a light duty 18 wheeler ;)
  • [18:23:14] <prpplague> ds2: ahh interesting, learn something new every day!
  • [18:23:37] <uberfry> imagine an app that flashes once you broke the high-speed record
  • [18:23:53] <ds2> prpplague: I didn't know til I interviewed with a company that was interfacing to that stuff a few years ago
  • [18:23:53] <uberfry> and sounds of a crowd cheering you on
  • [18:24:39] <prpplague> or uses the BT interface to call the police on your cell
  • [18:25:06] <ds2> ah the OnStar variant? ;)
  • [18:25:23] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [18:26:05] <keesj> dirk2: mamomna has been run on beagleboard
  • [18:26:18] * docelic (n=docelic@78.134.198.146) Quit (Read error: 110 (Connection timed out))
  • [18:26:52] <keesj> dirk2: I did get interesting docs from jkridner aboutg jtag indeed
  • [18:27:39] <prpplague> keesj: yea the interesting part is the comment about the scan chain being almost identical to the divinci
  • [18:27:58] * docelic (n=docelic@78.134.199.31) has joined #beagle
  • [18:28:26] * calculu5 is now known as calculus
  • [18:29:11] <keesj> indeed . still I find the documenation about emu0/1 a little vague
  • [18:30:12] <prpplague> keesj: yea, i did some reading up on those, for basic jtag operations, pulling them low is fine
  • [18:30:16] <uberfry> [20:24:39] <prpplague> or uses the BT interface to call the police on your cell <-- lmao what for? speed? :)
  • [18:30:24] <keesj> up right?
  • [18:30:26] <uberfry> it's all done on no-limit autobahn in germany
  • [18:30:38] <prpplague> keesj: from what i read, it is low
  • [18:30:48] <prpplague> uberfry: ahh
  • [18:30:57] <prpplague> uberfry: speed limits in the states of course
  • [18:31:08] <uberfry> here too
  • [18:31:11] <uberfry> but not everywhere :)
  • [18:31:42] <keesj> pwd
  • [18:32:50] <prpplague> keesj: http://wiki.davincidsp.com/index.php?title=Debugging_on_DaVinci_using_Olimex_dongle
  • [18:32:56] <prpplague> keesj: not the comment in the config
  • [18:32:59] <prpplague> note
  • [18:33:03] <prpplague> keesj: note the comment in the config
  • [18:33:16] <prpplague> #TI DM355 EVM: set EMU0 L and EMU1 L
  • [18:36:13] * sprotch (n=Beagle6@hunter.netmodule.com) Quit (Read error: 110 (Connection timed out))
  • [18:38:20] <Beagle5> l
  • [18:40:42] * trickie (n=trickie@hgaulton.xs4all.nl) Quit ("Leaving")
  • [18:41:47] * Beagle5 (n=Beagle5@dyn.83-228-203-144.dsl.vtx.ch) Quit ()
  • [18:43:18] * docelic (n=docelic@78.134.199.31) Quit ("http://www.spinlocksolutions.com/")
  • [18:46:36] <keesj> prpplague: I guess My adapter does it all wrong :P
  • [18:47:09] <prpplague> keesj: hehe, is your adapter working then with openocd?
  • [18:48:16] <keesj> "working" it a little ... but I do get some jtag scan chain output (I use the flyswatter)
  • [18:48:18] * RogerMon1 (n=a0740758@nat/ti/x-30bf1bfb93c4a7a4) Quit (Remote closed the connection)
  • [18:48:20] <keesj> http://www.paste-it.net/raw/public/b42c8f5/
  • [18:48:29] * dcordes (n=dcordes@unaffiliated/dcordes) Quit (Remote closed the connection)
  • [19:02:21] * cian (n=cianh@cian.ws) has joined #beagle
  • [19:06:26] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [19:06:28] * prpplague looks
  • [19:07:03] <prpplague> keesj: yea use the davinci config with the emu0/1 pulled low
  • [19:07:08] <mru> hi guys
  • [19:07:12] <prpplague> keesj: see what kind of output you get from openocd
  • [19:07:16] <prpplague> mru: greetings
  • [19:11:04] * dirk2 (n=dirk@F323b.f.strato-dslnet.de) has left #beagle
  • [19:21:17] * GentooHag (n=haggis@cpc1-clyd1-0-0-cust187.renf.cable.ntl.com) Quit ("Lost terminal")
  • [19:29:29] * AngstBeagle (n=angstrom@nat/ti/x-bbf26cdc61edc5a9) has joined #beagle
  • [19:53:01] * fer_luck_ is now known as fer_luck
  • [19:57:52] * Beagle6 (n=Beagle6@dyn.83-228-203-144.dsl.vtx.ch) has joined #beagle
  • [20:11:21] <jkridner|work> minor Angstrom issue: need to find the terminfo and add entry for cygwin: http://sources.redhat.com/ml/cygwin/2000-12/msg00203.html
  • [20:11:34] <jkridner|work> haven't found the terminfo file in Angstrom yet, but that shouldn't be too hard.
  • [20:12:06] * cian (n=cianh@cian.ws) Quit ()
  • [20:27:53] * RogerMonk (n=a0740758@nat/ti/x-c94c12bb08347999) has joined #beagle
  • [20:40:26] * AngstBeagle (n=angstrom@nat/ti/x-bbf26cdc61edc5a9) Quit (Remote closed the connection)
  • [20:43:07] * browser04 (n=treid@74.43.222.41) has left #beagle
  • [20:56:38] * Navi is now known as DrgnAK
  • [20:57:27] <keesj> jkridner|work: it can be found under /etc/terminfo
  • [20:58:24] * cbrake is now known as cbrake_away
  • [20:58:53] * ldesnogu (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) Quit ()
  • [20:59:53] * fer_luck (n=fer_luck@200.96.204.33) Quit ()
  • [21:00:37] <keesj> I thinkg the sources come from ncurses
  • [21:02:47] * cian (n=cian@cian.ws) has joined #beagle
  • [21:05:26] * DrgnAK is now known as Navi
  • [21:06:20] * Beagle6 (n=Beagle6@dyn.83-228-203-144.dsl.vtx.ch) Quit ()
  • [21:09:59] <jkridner|work> strange. still getting the error. anything I should use to reload it?
  • [21:13:57] <jkridner|work> I did an upgrade of all packages and am now getting this: Xorg: symbol lookup error: /usr/lib/xorg/modules//libfb.so: undefined symbol: pixman_fill
  • [21:13:58] <jkridner|work> :(
  • [21:14:20] <mru> missed dependency?
  • [21:17:46] * nooomem (i=d02f5302@gateway/web/ajax/mibbit.com/x-85550fbe2640b03f) has joined #beagle
  • [21:17:49] <nooomem> hey all
  • [21:17:53] <nooomem> how's everyone doing?
  • [21:20:32] <nooomem> anyone know the biggest SD card supported by this board?
  • [21:20:47] <mru> don't know
  • [21:20:52] <mru> I'm using an 8GB card
  • [21:21:05] <BThompson> supports SDHC, so probably most any SD card you can find
  • [21:21:55] * BThompson (n=BThompso@nat/ti/x-b5e5e6b2cf9c7158) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [21:21:55] <nooomem> BThompson: you see, SDHC comes in different classes (for speed) you know what max class beagle's reader approaches?
  • [21:22:05] <nooomem> mru: you see, SDHC comes in different classes (for speed) you know what max class beagle's reader approaches?
  • [21:23:11] <nooomem> would you know if this would work: http://www.newegg.com/Product/Product.aspx?Item=N82E16820141353
  • [21:24:15] <davr> i'm pretty sure even if beagle's not fast enough, the SD will still work, it just wont read at the maximum speed the card is capable of
  • [21:24:24] <nooomem> davr: k, cool
  • [21:27:07] <nooomem> what about wifi capability? would I have to give up the persistent memory to be able to shoot wireless or can I also just do a usb wifi drive with a usb hub? if so, where can I find a list of supported devices for this board?
  • [21:28:53] <jkridner|work> list of known-good peripherals is on http://beagleboard.org/hardware
  • [21:29:45] <jkridner|work> at the bottom is a list of "candidate" devices. I only add known-good peripherals to the list if I personally have seen them work well or from someone I know I can trust to have tested them well.
  • [21:30:37] <Crofton|work> How goos is openGL support in the beagle?
  • [21:30:43] <Crofton|work> http://lists.gnu.org/archive/html/discuss-gnuradio/2008-08/msg00124.html
  • [21:31:04] <jkridner|work> I haven't used DOGLESS.
  • [21:31:21] <jkridner|work> GLES support is fine, as soon as we have open source in the kerne.
  • [21:31:23] <jkridner|work> kernel.
  • [21:31:47] <jkridner|work> that seems like an interesting thing to profile.
  • [21:32:23] <Crofton|work> once I get data comnig from the USRP, the openGL displays would be really cool on the beagle
  • [21:32:31] <Crofton|work> it will turn heads :)
  • [21:34:26] <nooomem> and are the video unit's drivers being released anytime soon?
  • [21:34:38] <mru> don't hold your breath
  • [21:36:11] <nooomem> damn it... no word from whoever made it?
  • [21:36:26] <Crofton|work> my understanding is we have to sell jkridner's bosses :)
  • [21:37:02] <nooomem> Crofton|work: ok... I'll get cracking on an eBay page
  • [21:37:43] <Crofton|work> heh
  • [21:45:36] <jkridner|work> anyone know the mplayer arguments for playing an MP3 file in Angstrom?
  • [21:45:47] <jkridner|work> I'm not getting any audio and I'm wondering if I did something wrong.
  • [21:47:13] <mru> does audio output work?
  • [21:47:24] * mru wonders if he missed something
  • [21:47:51] <jkridner|work> got any simple test? I do get the *buzz* from u-boot.
  • [21:48:21] <mru> sakoman: you around?
  • [21:49:34] <sakoman> mru: yup
  • [21:49:44] <mru> what's the status of alsa?
  • [21:49:53] <mru> you were working on that, right?
  • [21:50:03] * mru sometimes gets names confused
  • [21:50:15] <sakoman> yes, multi-tasking on that and a few other things
  • [21:50:44] <sakoman> it is somewhat broken
  • [21:51:36] <sakoman> seems to work OK in OSS compatibility mode, but apps that use alsa apis die from a numm substream pointer fairly quickly
  • [21:52:10] <mru> jkridner|work: listening?
  • [21:52:17] * robclark (n=robclark@nat/ti/x-e37115e97c5cf9da) Quit ()
  • [21:52:28] <sakoman> The alsa asoc gurus that have looked at my drivers don't see anything wrong -- they suspect issues somewhere in alsa or the alsa library
  • [21:53:42] <sakoman> jkridner|work: you could try mplayer -ao oss song.mp3
  • [21:54:21] <sakoman> s/numm/null/ a few lines up
  • [21:54:52] <jkridner|work> Unable to handle kernel NULL pointer dereference at virtual address 00000000
  • [21:55:08] <sakoman> yup that is the symptom
  • [21:55:17] <jkridner|work> root@beagleboard:/home/angstrom# mplayer -ao oss lara-bach_128.mp3
  • [21:55:21] <jkridner|work> total lock-up
  • [21:55:37] <jkridner|work> SysRq works.
  • [21:56:55] <jkridner|work> things were working so much better before I did this upgrade.
  • [21:57:46] <sakoman> I've done a bit of debugging and it seems the null pointer is originating in the alsa library -- pcm.hw.c to be specific
  • [21:59:26] <sakoman> To be honest I'm not entirely convinced that the alsa lib in OE is completely compatible with the current linux-omap tree
  • [21:59:49] <sakoman> It is indeed the most recent version, but . . .
  • [22:00:23] <Crofton|work> still the kernel should protect itself from null ptr's?
  • [22:00:47] <sakoman> I would welcome help from anyone who cares to go mucking about in alsa/alsa lib
  • [22:01:41] <sakoman> Crofton|work: what should the kernel do when userland passes in a null pointer to be used by driver code?
  • [22:02:16] <Crofton|work> barf politely :)
  • [22:02:27] <Crofton|work> return an error code to luser space
  • [22:02:48] <sakoman> That sort of happens if you enable alsa debugging
  • [22:03:00] * mru looked at alsa-lib once
  • [22:03:49] <sakoman> Crofton|work: then it often gets caught by code snippets like:
  • [22:03:51] <sakoman> snd_assert(substream != NULL, return -ENXIO);
  • [22:04:11] <Crofton|work> amen
  • [22:04:20] * valhalla (n=valhalla@81-174-21-16.dynamic.ngi.it) Quit ("Leaving")
  • [22:04:40] <sakoman> If you enable alsa debugging you will get an error message from that assert and it will try to back out gracefully
  • [22:05:07] <sakoman> But not all of the alsa code checks return values, so things tend to degenerate anyway :-(
  • [22:05:48] <sakoman> I got burned out wading through the alsa code trying to debug this and decided to set it aside for a few days
  • [22:07:05] <Crofton|work> ok, I need to study my scuba gear
  • [22:07:11] <Crofton|work> going diving this weekend
  • [22:07:23] <sakoman> sounds like *much* more fun than alsa
  • [22:07:24] <Crofton|work> getting gear on and off will be amusing
  • [22:07:29] <Crofton|work> should be ok in the water :)
  • [22:07:32] <Crofton|work> right
  • [22:07:44] <Crofton|work> :)
  • [22:08:04] <Crofton|work> these xilinx tools are taking forever to install
  • [22:08:17] <Crofton|work> and I suspect they won't work in the end :)
  • [22:08:25] <Crofton|work> but I must try to be windows free
  • [22:09:34] <sakoman> To those who follow u-boot: the latest upstream changes break the beagle nand driver
  • [22:10:01] * prpplague (n=dave@mail.americanmicrosystems.com) Quit ("Leaving")
  • [22:10:05] <sakoman> Hopefully one of the TI nand experts will revise the driver
  • [22:10:13] <jkridner|work> I tried 'aplay' from Khasim's diagnostics and it crashed the kernel quickly.
  • [22:10:26] <sakoman> Onenand for evm and overo still work
  • [22:10:29] * cian_ (n=cianh@cian.ws) has joined #beagle
  • [22:10:30] <jkridner|work> same error.
  • [22:11:06] <sakoman> jkridner: yes, I would expect that -- aplay uses the alsa api's
  • [22:11:19] <jkridner|work> guess I would just be happy to get X running again, even without audio. :(
  • [22:11:25] <Crofton|work> we need to send a beagle to an alsa guy :)
  • [22:11:31] <sakoman> things have gotten much worse with recent linux-omap kernels
  • [22:12:02] <sakoman> Crofton|work: a great idea! I would love to get rid of this driver ;-)
  • [22:12:18] <jkridner|work> I've also noted that FAT writes are *extremely* slow to SD. ext2/3 writes to SD seem fine.
  • [22:12:34] * dcordes (n=dcordes@unaffiliated/dcordes) has joined #beagle
  • [22:12:46] <jkridner|work> give me a name and I'll find a board!
  • [22:12:49] <sakoman> jkridner: audio used to work perfectly on evm and overo until about 2 weeks ago
  • [22:12:54] <dcordes> is there a way to flash from sd without having a serial connection? like morse code with the user button?
  • [22:13:21] <sakoman> jkridner: now it doesn't work anywhere :-(
  • [22:13:35] <Crofton|work> sadly, a name I have not
  • [22:14:18] <sakoman> jkridner: seems to be a period of maximum instability -- upstream linux and u-boot have broken lots of stuff
  • [22:14:26] <Crofton|work> ok, time to reset the time date etc on this cursed dive computer
  • [22:14:58] <Crofton|work> sakoman, do yuo have a list of stuff?
  • [22:15:57] * Crofton|work needs to be quiet, he is going to be overworked updating his PLD skills from 22v10 to xilinx/altera fpga
  • [22:16:11] <sakoman> Crofton|work: well audio and nand for starters ;-)
  • [22:16:32] <Crofton|work> the 22v10 is the equivalent of 2 or 3 CLB's in a fpga ....
  • [22:17:43] <sakoman> Crofton: I'm sure most folks have clue what you are talking about
  • [22:18:06] <sakoman> They probably were still in grade school during the 22v10 era
  • [22:18:19] <sakoman> have no clue is what I meant
  • [22:21:16] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) has joined #beagle
  • [22:23:07] <jkridner|work> I setup the college lab to be able to program 22v10s. :)
  • [22:23:17] <jkridner|work> had a lot of fun with those.
  • [22:23:54] <jkridner|work> before that, everyone was using discrete logic gates.
  • [22:24:37] * mru admits to having no clue
  • [22:24:39] <jkridner|work> of course, the rest of the world was already on FPGAs with RAM-based LUTs.
  • [22:25:45] <banderson> I *programmed* for my first fpga only 4-5 months ago...
  • [22:27:05] <jkridner|work> I'll be happy going home today if I can at least get Xorg running again.
  • [22:29:03] * nooomem (i=d02f5302@gateway/web/ajax/mibbit.com/x-85550fbe2640b03f) Quit ("http://www.mibbit.com ajax IRC Client")
  • [22:30:38] <Crofton|work> anyone remember CUPL?
  • [22:31:18] <banderson> How has everyone been compiling x-loader/U-boot for the beagle? Directly with gcc (codesourcery) or ?
  • [22:31:37] <mru> what else?
  • [22:31:54] * Crofton hopes that the Lyrtech Small Form Factor support in u-boot 1.3.4 is ok :)
  • [22:31:58] * Thanatos_desktop (n=rwvens@p578FEFA2.dip.t-dialin.net) has joined #beagle
  • [22:32:13] <banderson> mru: wasn't sure if that was only way...
  • [22:32:39] <mru> I suppose you could compile/assemble it by hand...
  • [22:32:50] <mru> though I prefer using a compiler
  • [22:33:01] <banderson> LOL
  • [22:33:06] <Crofton|work> rofl, http://www.ese.upenn.edu/rca/software/CUPL/cuplindex.html
  • [22:35:53] <banderson> I guess my real question was that do most ppl have another toolchain outside of oe that is used to compile xload/uboot...
  • [22:36:19] <mru> oh, I don't have oe at all
  • [22:36:21] * asterick (n=unicdk@24-117-80-129.cpe.cableone.net) has joined #beagle
  • [22:41:04] <sakoman> banderson: I just use my OE recipe for both x-load and u-boot
  • [22:41:33] <sakoman> I'm pretty sure most people *don't* build x-load and u-boot!
  • [22:41:43] <jkridner|work> mru: did you get those clock settings I sent. Let me know if they are useful or not.
  • [22:41:48] <banderson> sakoman: That is what I was hoping could be done!
  • [22:41:55] <sakoman> Unless you are working on the code, there would be no point!
  • [22:43:24] <banderson> sakoman: just want to be able to give client the ability to compile as much from source as possible.
  • [22:45:02] <banderson> sakoman: Don't think I will be doing it on regular basis just need to be able to do it...and the easier the better....
  • [22:45:30] <sakoman> banderson: I understand -- that was my motivation too
  • [22:47:06] <jkridner|work> how do I find what package /usr/lib/xorg/modules//libfb.so is a part of (so I can figure out why it is missing the symbol 'pixman_fill')
  • [22:48:04] <mru> pixman_fill sounds like it should come from pixman
  • [22:48:16] <mru> do you have a libpixman.so somewhere?
  • [22:50:32] <jkridner|work> yes.
  • [22:50:43] <jkridner|work> er, no. I have libpixman-1.so.0
  • [22:50:54] <jkridner|work> should I 'ln -s'?
  • [22:50:56] <mru> close enough
  • [22:51:10] <mru> does that lib define the pixman_fill?
  • [22:51:21] <mru> does libfb.so link to it?
  • [22:51:24] <jkridner|work> I use 'nm' for that?
  • [22:51:35] <mru> yes, or readelf
  • [22:51:40] <jkridner|work> hmmm... no symbols.
  • [22:51:44] <mru> nm -D
  • [22:51:56] <mru> or readelf -s
  • [22:52:15] <jkridner|work> flashbacks. (years since I've done this)
  • [22:52:49] <jkridner|work> root@beagleboard:/usr/lib# nm -D libpixman-1.so.0 | grep fill
  • [22:52:49] <jkridner|work> 00010140 T pixman_image_create_solid_fill
  • [22:52:49] <jkridner|work> 00010198 T pixman_image_fill_rectangles
  • [22:53:06] <jkridner|work> no pixman_fill
  • [22:53:46] <mru> mine has it
  • [22:54:04] <mru> which version of pixman do you have?
  • [22:57:24] <jkridner|work> how do I query?
  • [22:57:37] <mru> don't know
  • [22:58:28] <jkridner|work> 0.11.8
  • [22:58:35] <jkridner|work> just based on the file names.
  • [23:04:30] <mru> http://gitweb.freedesktop.org/?p=pixman.git;a=commit;h=3dca89a677528845ccaee7ff6d355207c631ddd2
  • [23:06:48] <jkridner|work> hmmm... so I must have something older?
  • [23:06:55] <jkridner|work> prior to this commit?
  • [23:06:59] <mru> that commit is old
  • [23:07:06] <mru> are you sure it's not there
  • [23:07:27] <mru> what does readelf -s say?
  • [23:07:38] <mru> I don't trust nm with shared libs
  • [23:10:21] <jkridner|work> root@beagleboard:/usr/lib# readelf -s libpixman-1.so | grep fill
  • [23:10:21] <jkridner|work> 102: 00010198 708 FUNC GLOBAL DEFAULT 10 pixman_image_fill_rectang
  • [23:10:57] * Olipro_ (n=Olipro@unaffiliated/olipro) has joined #beagle
  • [23:12:00] <mru> weird
  • [23:14:06] <mru> does strings libpixman-1.so reveal the version number?
  • [23:14:16] <mru> in mine, it's the last string printed
  • [23:15:30] * bazbell (n=a0192809@nat/ti/x-adce310f1c61fbfa) has joined #beagle
  • [23:16:09] <calculus> I have pixman_fill in /usr/lib/libpixman-1.so.0.10.0
  • [23:22:11] * RogerMonk (n=a0740758@nat/ti/x-c94c12bb08347999) has left #beagle
  • [23:27:20] * cian (n=cian@cian.ws) Quit ()
  • [23:27:56] <dcordes> I'm making an cable for the rs232 port and I wonder how to connect the beagle's rs232 pins (documented in the reference manual) to their according pin on the d-sub (their is only a picture showing how some ribbon cable is connected to a d-sub)
  • [23:28:25] * Olipro (i=Olipro@unaffiliated/olipro) Quit (Read error: 110 (Connection timed out))
  • [23:28:26] <dcordes> there
  • [23:29:05] <mru> http://www.pccables.com/07120.htm
  • [23:31:19] <dcordes> mru: thanks, do the numbers on the everex connector there (second image) match the numbers on the board's pins?
  • [23:31:27] <mru> yes
  • [23:31:57] <mru> a ribbon cable is the easiest way
  • [23:32:51] <dcordes> mru: I alredy have cables on the pin side connector, only need to solder them to the d-sub.
  • [23:33:20] <dcordes> unlike on that website the d-sub I have here is female. do I need to twist a cable?
  • [23:35:23] <mru> what cable will you use to connect to the pc?
  • [23:36:12] <dcordes> the one I'm making right now
  • [23:36:38] <mru> wait, what cables do you have?
  • [23:36:43] <mru> what's connected to the beagle?
  • [23:36:56] <dcordes> I found 5 2pin connectors which fit the beagle's rs232 pins
  • [23:37:09] <mru> oh, you're making a cable to go directly from the beagle IDC10 to the pc dsub?
  • [23:37:22] <dcordes> is that not a good idea?
  • [23:37:28] <mru> it will work
  • [23:37:42] <mru> a bit non-standard only
  • [23:37:55] <dcordes> I'm not going to flash every day anyway
  • [23:38:34] <jkridner|work> For those of you brave with a soldering iron and looking for a fix to the GPT1 issue...
  • [23:38:44] <jkridner|work> you can try to remove C70.
  • [23:38:49] <mru> where is it?
  • [23:39:05] <mru> not that it's troubling me... just curious to see how brave I'd need to be
  • [23:39:47] <jkridner|work> I knew that was coming next. I was told it was not that hard to do.
  • [23:39:54] <jkridner|work> must run
  • [23:42:33] <dcordes> interesting
  • [23:43:32] <dcordes> mru: can I just take a floppy cable and connector?
  • [23:43:56] <mru> floppy cable?
  • [23:44:16] <dcordes> yes fdd
  • [23:44:31] * rsalveti (n=salveti@200.184.118.132) Quit (Read error: 113 (No route to host))
  • [23:44:32] <mru> don't they have many more pins?
  • [23:44:35] <mru> at least 20
  • [23:44:40] <mru> can't remember
  • [23:44:41] * rsalveti (n=salveti@189.70.119.44) has joined #beagle
  • [23:44:50] * mru doesn't have a floppy drive since many years
  • [23:44:54] <dcordes> yes they do but couldn't I still use it?
  • [23:45:14] <dcordes> me n'either but the mainboard box has one
  • [23:45:43] <mru> I suppose the extra wires won't do any harm
  • [23:45:53] <mru> if the you can physically plug it in
  • [23:46:04] <dcordes> yep that will work
  • [23:48:49] * asterick (n=unicdk@24-117-80-129.cpe.cableone.net) Quit ()
  • [23:54:22] * Ravi (n=chatzill@ip68-100-201-32.dc.dc.cox.net) has joined #Beagle
  • [23:54:30] * asterick (n=unicdk@24-117-80-129.cpe.cableone.net) has joined #beagle
  • [23:54:48] <Ravi> hey dudes... and dudettes.
  • [23:55:09] <Ravi> you think I'd be able to run a light jabber server on one of these boards?
  • [23:55:16] <Ravi> may be support like 40 people?
  • [23:55:22] <Ravi> small family / business?
  • [23:55:26] <mru> how much memory does it need?
  • [23:55:49] <Ravi> mru: good question, looking
  • [23:56:27] * NishanthM (n=Nishanth@cpe-24-27-74-89.tx.res.rr.com) has joined #beagle
  • [23:57:05] <Ravi> mru: how much do I have available?
  • [23:57:06] <Ravi> mru: 128?
  • [23:57:33] <mru> 128MB
  • [23:57:43] <Ravi> mru: shit... any way to extend that?
  • [23:57:57] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [23:58:54] <mru> no
  • [23:59:23] <ds2> you could swap to a USB drive ;)
  • [23:59:50] <Ravi> ds2 madness!