• [01:03:35] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [02:22:59] * rsalveti (n=salveti@189.70.143.68) has joined #beagle
  • [03:22:40] * RogerMonk (n=a0740758@nat/ti/x-2c108a8813c63567) Quit (Remote closed the connection)
  • [03:23:08] * RogerMonk (n=a0740758@nat/ti/x-3d905f96499f3251) has joined #beagle
  • [03:31:01] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [03:31:21] <ds2> doh just missed khasim
  • [03:33:23] <ds2> khasim/jkridner: if you check the logs - video and sound drivers may be available for the beagle
  • [04:00:06] * Olipro (n=Olipro@unaffiliated/olipro) Quit (Read error: 110 (Connection timed out))
  • [04:29:43] * ldesnogu (n=ldesnogu@fw-tnat.cambridge.arm.com) Quit (Read error: 101 (Network is unreachable))
  • [04:33:08] * kstar (n=kstar@59.92.133.61) has joined #beagle
  • [04:33:36] * kstar (n=kstar@59.92.133.61) has left #beagle
  • [04:34:03] * ldesnogu (n=ldesnogu@fw-tnat.cambridge.arm.com) has joined #beagle
  • [04:54:40] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [05:04:58] * khasi1 (n=a0393720@192.163.20.231) has joined #beagle
  • [05:04:58] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [05:12:21] * khasi1 (n=a0393720@192.163.20.231) has left #beagle
  • [05:17:40] * khasi1 (n=a0393720@192.163.20.231) has joined #beagle
  • [05:20:03] * bazbel1 (n=a0192809@nat/ti/x-45ba981f1af32310) has joined #beagle
  • [05:20:12] * bazbel1 (n=a0192809@nat/ti/x-45ba981f1af32310) Quit (Remote closed the connection)
  • [05:22:02] * dirk2 (n=dirk@F31da.f.strato-dslnet.de) has joined #beagle
  • [05:26:05] * bazbell (n=a0192809@nat/ti/x-7fdddcf04eeb7f01) Quit (Remote closed the connection)
  • [05:47:34] * khasi1 (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [05:53:39] * khasim (n=a0393720@192.163.20.231) has joined #beagle
  • [06:14:13] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) Quit ("Trillian (http://www.ceruleanstudios.com")
  • [06:57:04] <RogerMonk> Koen - ping!
  • [06:57:07] <RogerMonk> gm
  • [07:31:17] <koen> pong
  • [07:33:45] * darkerror05 (n=darkerro@cpe-72-230-185-21.rochester.res.rr.com) Quit (Read error: 104 (Connection reset by peer))
  • [07:34:27] * darkerror05 (n=darkerro@cpe-72-230-185-21.rochester.res.rr.com) has joined #beagle
  • [07:37:51] <ds2> khasim: ping
  • [07:37:52] * hagisbasheruk (n=hagisbas@78.148.131.1) has joined #beagle
  • [07:57:20] * hagisbasheruk (n=hagisbas@78.148.131.1) Quit ("Ex-Chat")
  • [08:02:44] <dirk2> sakoman: I had a look to your beagle NAND patches at OMAP list from 2008/06/11. Two remarks:
  • [08:03:58] * roxfan2 (n=dunno@253.0-200-80.adsl-dyn.isp.belgacom.be) Quit ()
  • [08:04:57] <dirk2> sakoman: I'd like it not to create a new file board-omap3beagle-flash.c, instead put this stuff directly into board-omap3beagle.c
  • [08:06:02] <dirk2> sakoman: And the third patch, the defconfig changes, gives me rejects. Currently checking this...
  • [08:08:13] <dirk2> sakoman: defonfig rejects: The defconfig patch contains the swtich from CONSISTENT_DMA_SIZE=2 to 4, too. But this seems to be already upstream.
  • [08:18:13] <RogerMonk> Koen - hi! - I finally got OE tools installed and org.openembedded.dev tree checked out!
  • [08:18:42] <RogerMonk> can you send me a local.conf I should use to build the angtrom image for beagle please?
  • [08:30:23] <koen> RogerMonk: http://rafb.net/p/vLG01o74.txt
  • [08:30:47] <koen> and "echo MACHINE=beagleboard > conf/auto.conf"
  • [08:43:21] <RogerMonk> thanks - and then 'bitbake ?' ?
  • [08:44:30] * CyruzDraxs (n=CyruzDra@S0106001cf0719b7b.ok.shawcable.net) has joined #beagle
  • [08:50:35] <koen> bitbake beagleboard-demo-image
  • [08:50:56] <koen> or plain 'bitbake foo' to see if you setup is working correctly
  • [08:51:22] <RogerMonk> http://rafb.net/p/3FcgZh84.html from bitbake foo
  • [08:51:48] <RogerMonk> same with beagleboard-demo-image
  • [08:56:19] <koen> echo 0 > /proc/sys/vm/mmap_min_addr
  • [09:06:41] <ldesnogu> koen: playing with some emulator? :)
  • [09:15:02] <RogerMonk> Koen - building!!! - woo hoo! - how long should I expect it to take (packages coming down at 600k, pc is 2GHz)?
  • [09:15:10] <RogerMonk> (btw - thanks!)
  • [09:20:16] <koen> RogerMonk: I can build it in 2 hours from scratch on a core2quad
  • [09:22:15] <RogerMonk> ok, thanks - I'll let you know how it goes!
  • [09:22:16] <koen> RogerMonk: the initial build time only has to be done once :)
  • [09:22:33] <koen> after than you can use packaged-staging to do rebuilds if needed
  • [09:44:50] * CyruzDraxs (n=CyruzDra@S0106001cf0719b7b.ok.shawcable.net) Quit ()
  • [10:39:39] * vsriram (i=7aa60de8@gateway/web/ajax/mibbit.com/x-cbadc77af1df59e8) has joined #beagle
  • [10:48:18] * DJW|Home (i=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) has joined #beagle
  • [11:06:18] * DJWillis (i=djwillis@82-46-19-72.cable.ubr02.bath.blueyonder.co.uk) Quit (Read error: 110 (Connection timed out))
  • [11:07:05] <koen> RogerMonk: is it still chugging along?
  • [11:08:54] <RogerMonk> yep, curl-native-7.18.0
  • [11:15:32] * vsriram (i=7aa60de8@gateway/web/ajax/mibbit.com/x-cbadc77af1df59e8) has left #beagle
  • [11:18:47] <koen> sakoman, Crofton|work: http://lists.arm.linux.org.uk/lurker/message/20080703.180533.1b7612a1.en.html
  • [11:20:39] <Crofton|work> does that work?
  • [11:26:49] <koen> haven't tried yet
  • [11:27:11] <koen> beagle does have armv7 oprofile support now :)
  • [11:27:43] * esden`away is now known as esden
  • [11:29:14] <koen> I really want to figure out the ffmpeg slowness problem before LRL
  • [11:35:50] * likewise (n=chatzill@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [11:35:54] <likewise> gm
  • [11:36:11] <koen> hey likewise
  • [11:36:16] <likewise> hoi koen
  • [11:36:52] <koen> likewise: did you already see http://www.nu.nl/news/1640426/122/rss/Poepterrorist_plaagt_Eindhoven.html ?
  • [11:37:08] <ldesnogu> koen: did you look why you get the glibc error message when running ffmpeg?
  • [11:37:21] <likewise> koen: yes, heard it on the radio yesterday. Luckily not my area, and luckily and don't commute to work by bike.
  • [11:38:08] <koen> ldesnogu: no, but I ingore such message when I build from trunk/head
  • [11:39:17] <ldesnogu> well I don't get such messages from ffmpeg trunk + mru NEON stuff (running both on my x86_64 and through qemu)
  • [11:43:04] <koen> http://amethyst.openembedded.net/~koen/beagleboard/uImage-2.6.25+2.6.26-rc8+r38+gitc32c81d59d2d8a66e63f82c9732db256d302068e-r38-beagleboard.bin
  • [11:43:08] <koen> that has oprofile support
  • [11:44:00] <jkridner> good morning.
  • [11:44:22] <ldesnogu> koen: you could perhaps try to run only libavcodec/dct-test and check your results with mru; that could help track the discrepancy
  • [11:44:25] <ldesnogu> hello
  • [11:49:27] <ldesnogu> koen: mru dct-test results http://archives.free.net.ph/message/20080627.194908.935fd909.en.html
  • [11:50:02] <koen> /OE/angstrom-tmp/work/armv7a-angstrom-linux-gnueabi/ffmpeg-0.4.9+r7+gitrf0dac08fdf2ef97196dad8393adffe8bf5340f82-r7/git/libavcodec/dct-test.c:551: undefined reference to `mm_support'
  • [11:50:20] <ldesnogu> oh you need to use trunk version of dct-test.c
  • [11:50:27] <ldesnogu> plus updates
  • [11:50:37] <ldesnogu> I can send you the updated files to apply to the trunk
  • [11:58:00] <koen> mru: any chance you can update you git tree to merge in those changes?
  • [12:01:18] * koen heads back to the library
  • [12:01:39] * dirk2 (n=dirk@F31da.f.strato-dslnet.de) has left #beagle
  • [12:15:34] <koen> drat, I just realised that I really shouldn't crash the beagle now
  • [12:15:39] <koen> since it's 2km away :)
  • [12:18:05] <jkridner> khasim: ping
  • [12:19:03] * jkridner wishes someone had a software solution to cut power to USB ports for remote Beagle power cycling.
  • [12:20:38] <ldesnogu> jkridner: a wife who stays at home might be a solution, though it can't be patched
  • [12:21:00] <likewise> koen: hardware watchdog support?
  • [12:21:01] <jkridner> latency and reliability issues.
  • [12:23:59] * DJW|Home is now known as DJWillis
  • [12:24:58] <koen> jkridner: there used to be a sysfs file where you could do that
  • [12:25:29] <jkridner> I'm thinking the best way to do it is with a second Beagle, that way the hardware is predictable.
  • [12:25:37] <jkridner> but, we need the host port for that. :(
  • [12:26:09] * jconnolly (n=jconnoll@ool-18ba051f.dyn.optonline.net) has joined #beagle
  • [12:26:30] <jkridner> well, maybe it is possible with the OTG port, but I don't know of any software control for the power from that port.
  • [12:26:49] <keesj> koen: I do it with "simple" hardware http://www.htvision.com/product.asp?intProdID=80
  • [12:26:50] <koen> well, the linux usb stack used to have that feature
  • [12:26:56] <koen> dunno if it's still there
  • [12:27:27] * jconn|building (n=jconnoll@ip-64-32-229-194.dsl.nyc.megapath.net) Quit (Remote closed the connection)
  • [12:27:59] <keesj> in the f_file_storage you can listen to scsi eject commands and reboot
  • [12:28:15] <keesj> at least what would be a nice place.
  • [12:28:38] <Crofton|work> rschuster is up
  • [12:28:39] <keesj> the "eject" command under linus does send those scsi eject command
  • [12:30:19] <koen> it was something like echo 0 > /sys/bus/usb/device/4838,4389/power
  • [12:36:42] <jkridner> Crofton|work: I was looking at that....
  • [12:37:14] <jkridner> should we point him to koen's builds on amethyst or the angstrom-distribution/unstable builds?
  • [12:40:44] <Crofton|work> He's another OE guy :)
  • [12:40:59] <jkridner> ah. I won't insult him then.
  • [12:41:07] <koen> tell him to fix his java stuff vor gcc 4.3 :)
  • [12:41:07] <jkridner> is he the phoneme guy?
  • [12:41:37] <jkridner> ah, excellent. I'll tell him you told me to do it. :)
  • [12:44:34] * jconnolly (n=jconnoll@ool-18ba051f.dyn.optonline.net) Quit (Read error: 110 (Connection timed out))
  • [12:45:12] * jconnolly (n=jconnoll@ool-18ba051f.dyn.optonline.net) has joined #beagle
  • [12:47:10] <koen> he works for tarent, a java shop
  • [12:48:07] <koen> tarent uses OE for most of their linux work
  • [12:48:19] <koen> even to target maemo, since they got fed up with scratchbox
  • [12:55:06] <Crofton> do the Liniuxtag boards have the cap needed for reliable OTG/host operation?
  • [12:55:55] <jkridner> Crofton: yes.
  • [12:56:07] <jkridner> Note the mention of revision "B4".
  • [12:56:22] <Crofton> ok
  • [12:56:49] <jkridner> Just when I was about to start mailing them out, Gerald told me to mail them back to him. Next day, he had reworked boards for me.
  • [12:56:59] * rsalveti (n=salveti@189.70.143.68) Quit (Read error: 110 (Connection timed out))
  • [12:57:10] <jkridner> so, B4 boards have the cap and do NOT have the EHCI port populated at all.
  • [13:05:19] * jconnolly (n=jconnoll@ool-18ba051f.dyn.optonline.net) Quit (Read error: 110 (Connection timed out))
  • [13:05:49] * jconnolly (n=jconnoll@ool-18ba051f.dyn.optonline.net) has joined #beagle
  • [13:07:49] * jkridner|work (n=a0321898@nat/ti/x-760d08013006a7fc) has joined #beagle
  • [13:08:02] * rsalveti (n=salveti@200.184.118.132) has joined #beagle
  • [13:18:02] <sakoman> koen: you may have noticed from irc logs that I haven't had much success with beagle-demo-image in nand
  • [13:18:37] <sakoman> seems the jffs2 image is damaged/corrupted
  • [13:19:05] <sakoman> same results whether flashed from u-boot or mtd-utils
  • [13:19:40] <sakoman> omap3evm-demo-image works without a problem though!
  • [13:20:31] <sakoman> Do you know if any of the packages in beagle-demo-image attempt to create hard links?
  • [13:21:00] <sakoman> I seem to recall that is a no-no for jffs2
  • [13:24:01] * jconnolly (n=jconnoll@ool-18ba051f.dyn.optonline.net) Quit (Read error: 110 (Connection timed out))
  • [13:24:36] * NishanthM (n=Nishanth@cpe-24-27-74-89.tx.res.rr.com) has joined #beagle
  • [13:29:33] <koen> sakoman: not that I know off
  • [13:30:48] <sakoman> koen: interesting! In order to track down what is breaking beagle-demo-image I started adding packages from beagle-demo-image to omap3evm-demo-image. Hit it on the first try with e-wm
  • [13:31:06] <sakoman> Will do some investigation
  • [13:31:09] <koen> heh
  • [13:31:19] <koen> I only included e-wm for its shininess
  • [13:31:36] <koen> it is utter shit from a distro POV
  • [13:40:01] <koen> the always present firefox icon is a hint at how broken it is wrt assumptions
  • [13:41:54] <koen> Crofton|work: making progress on boost?
  • [13:42:00] <Crofton|work> yeah
  • [13:42:07] <Crofton|work> i botched the testing
  • [13:42:19] <Crofton|work> and pushed to DEF-PREF on 33.1 though :)
  • [13:42:25] <koen> RogerMonk, sakoman: btw, 'opkg install swfdec-mozilla' will give you flash support in webkit
  • [13:42:30] <Crofton|work> I'll fix as soon as .0 builds :)
  • [13:42:36] <RogerMonk> koen : build still going ...webkit-gtk now...
  • [13:42:53] <koen> RogerMonk, sakoman: no youtube video support yet, I need to figure out which codec it needs and add that as a dep
  • [13:43:06] <Crofton|work> the Fedora 9 SRPMS has a gcc-4.3 patch
  • [13:43:33] <RogerMonk> swfdec uses gstreamer, right?
  • [13:43:35] <koen> Crofton|work: could you put that patch under a fedora/ subdir like the gcc 4.3 patches?
  • [13:43:46] <koen> RogerMonk: yes, but you can hook in gst-ffmpeg :)
  • [13:44:12] <Crofton|work> I'll take a look
  • [13:44:19] <koen> RogerMonk: my beagle currenly has gstreamer using the NEON optimized ffmpeg
  • [13:44:29] <Crofton|work> Fedora splits the patches, debian has one patch it looks like
  • [13:44:33] <koen> (which for some reason is twice as slow as it should be)
  • [13:44:37] * trickie (n=trickie@basesoft.xs4all.nl) Quit ("Leaving")
  • [13:44:46] <koen> Crofton|work: debian sucks in that regard
  • [13:45:11] <koen> Crofton|work: debian has no policy about patches, so sometimes you get lucky (gcc), sometimes you don't
  • [13:46:42] <koen> RogerMonk: the dsplink chat is still on for next monday?
  • [13:47:36] <RogerMonk> yep - hope u can still make it
  • [13:48:05] <koen> it looks like it
  • [13:48:08] <RogerMonk> looking forward to hearing all your feedback!
  • [13:48:27] <likewise> Is this dsplink chat an open chat?
  • [13:48:52] <koen> depends on how well echelon works ;)
  • [13:49:43] <koen> I'd really like to use 'upstream' link and CE on the neuros
  • [13:50:10] <RogerMonk> it's a discussion to review koen's/crofton's pains wrt integrating dsplink into OE build. Send me an email if you'd like to join r dash monk at ti
  • [13:51:11] <Crofton|work> We need to understand why they need such a painful build system :)
  • [13:53:34] <likewise> RogerMonk: well, we rejected a design-in for the omap because of that issue, and one other issue. Not that we make TI rich, my employer's company does low-volume broadcast equipment.
  • [13:57:51] * docelic (n=docelic@78.134.195.216) has joined #beagle
  • [13:59:03] <docelic> hey folks, boards for ltag 2008 winners shipped yet?
  • [13:59:17] <koen> docelic: some have
  • [13:59:20] <jkridner> yes, but I don't have all the tracking numbers back.
  • [13:59:35] <jkridner> all have now shipped.
  • [13:59:46] <docelic> no problems, just informative question
  • [13:59:49] <jkridner> some have arrived (without me getting the tracking numbers back)
  • [14:00:49] <docelic> btw are those rev.2 boards?
  • [14:00:57] <jkridner> Rev B4.
  • [14:01:08] * koen heads back home
  • [14:04:25] <Crofton|work> RogerMonk, did you work out your firewall issues?
  • [14:22:34] <Crofton|work> well, 1.34.1 didn't build easily
  • [14:23:05] * Crofton|work curses project that have to be special and create their own build system
  • [14:23:39] <RogerMonk> Crofton, no... I'm working from home today so no firewall. Wanted to see 'how it's supposed to work' before tacking the firewall issues!
  • [14:23:55] <Crofton|work> :)
  • [14:23:58] <Crofton|work> good idea
  • [14:24:15] <RogerMonk> ok.... just got an error though....
  • [14:24:18] <RogerMonk> damn
  • [14:24:44] <RogerMonk> ERROR: ..../packages/e17/exhibit_cvs.bb' failed... any ideas?
  • [14:24:49] <Crofton|work> pastebin?
  • [14:25:07] <RogerMonk> ok
  • [14:25:10] <Crofton|work> pastyebin more lines
  • [14:26:35] <RogerMonk> http://pastebin.com/m36438a83
  • [14:28:09] <RogerMonk> ...restarted... seems to be continuing...
  • [14:30:53] <Crofton> might be a transient fetch error
  • [14:31:26] <Crofton> || builds can make it hard to find the errors, since the error occurs, but you are still getting messages from tasks that are still running
  • [14:37:32] * shoragan just got his board from dhl
  • [14:37:34] <shoragan> :)
  • [14:46:04] <likewise> shoragan: congrats
  • [14:49:04] <koen> RogerMonk: the sf.net cvs server only allows one connection per host
  • [14:49:24] <koen> RogerMonk: we don't have a way to express that constraint yet :(
  • [14:49:58] <koen> RogerMonk: bitbake -c fetchm
  • [14:50:00] <koen> ehm
  • [14:50:26] <koen> RogerMonk: bitbake -c fetchall beagleboard-demo-image will fetch everything before starting a build
  • [14:53:03] <RogerMonk> thanks - I guess though if I retry, it'll eventually get it anyway?
  • [14:55:23] <Crofton|work> yeah, and once it gets a local copy, it does not need to redownload stuff again
  • [14:56:00] <RogerMonk> it's very cool this OE... I just need a faster DSL connection I think...
  • [14:56:03] <Crofton|work> The boost people need to learn to use major rev # changes when they break stuff
  • [14:57:32] * banderson (n=irc@69.71.183.7) Quit ("Leaving")
  • [15:00:27] <koen> likewise: I think I have a fix for gnash
  • [15:01:42] <koen> gah
  • [15:02:00] <koen> I can't update mrus branch because git 1.4.4.4 can't merge branches....
  • [15:02:23] <likewise> koen: would be nice
  • [15:02:54] * koen waits for packaged-staging to repopulate staging from scratcht
  • [15:11:48] * Crofton|work curses boost
  • [15:16:55] <shoragan> do i need some special hdmi/dvi cable?
  • [15:17:34] <koen> you need a passive one
  • [15:17:40] <koen> the signal are compatible
  • [15:17:53] <koen> they are like ???4 in stores
  • [15:18:00] <shoragan> so i just buy the cheapest one i find in a normal store?
  • [15:26:16] <koen> that should work
  • [15:27:32] <Crofton|work> shoragan, that is what I did
  • [15:27:40] <shoragan> ok :)
  • [15:28:12] * likewise (n=chatzill@82-171-51-231.ip.telfort.nl) Quit ("ChatZilla 0.9.83 [Firefox 3.0/2008061015]")
  • [15:29:56] * Olipro (n=Olipro@unaffiliated/olipro) has joined #beagle
  • [15:32:55] <ldesnogu> Catalin submitted the updated v7 cache printing functions
  • [15:33:20] * BeagleLogBotTest (n=PircBot@ec2-75-101-156-174.compute-1.amazonaws.com) Quit (Remote closed the connection)
  • [15:35:02] * gadiyar (n=a0393673@192.163.20.231) has joined #beagle
  • [15:36:25] <khasim> welcome gadiyar
  • [15:36:32] <Crofton> koen, I am guessing they changed all the options for bjam
  • [15:36:44] <gadiyar> hi khasim
  • [15:36:58] <khasim> gadiyar: let me introduce to few beagle giants... :)
  • [15:37:00] <gadiyar> I love the beagleboard :)
  • [15:37:22] <gadiyar> Hi all, I'm new to this board - have been using it for the last 20 minutes
  • [15:37:55] <khasim> gadiyar is helping us in fixing the Kernel hang in GIT code base
  • [15:38:27] <gadiyar> I am currently running git-bisect to see if I can find a bad commit
  • [15:38:34] <gadiyar> If anyone has already tried this, let me know
  • [15:51:48] <keesj> The [b]eagle has landed! off we go!
  • [15:52:27] * keesj is very happy
  • [16:06:14] <gadiyar> 545 revisions to test - Ten iterations to go
  • [16:06:38] <koen> gadiyar: wouldn't you hit a rev where tehre isn't beagle support?
  • [16:12:21] * esden is now known as esden`away
  • [16:15:30] * koen is developing a strong dislike for gnash
  • [16:22:55] * NishanthM (n=Nishanth@cpe-24-27-74-89.tx.res.rr.com) Quit (Read error: 110 (Connection timed out))
  • [16:23:42] * NishanthM (n=Nishanth@cpe-24-27-74-89.tx.res.rr.com) has joined #beagle
  • [16:29:33] * Olipro (n=Olipro@unaffiliated/olipro) Quit (Read error: 110 (Connection timed out))
  • [16:30:56] <gadiyar> koen: yes, I usually do.
  • [16:35:17] * alphaone (n=alphaone@a064.apm.etc.tu-bs.de) has joined #beagle
  • [16:38:44] * ldesnogu_ (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) has joined #beagle
  • [16:39:19] <gadiyar> A couple of weeks ago, I decided to hunt for a solution to this problem
  • [16:39:43] <gadiyar> Here's what I ran into: http://www.ussg.iu.edu/hypermail/linux/kernel/0704.1/0636.html
  • [16:39:47] <gadiyar> This works perfectly for me
  • [16:47:09] * NishanthM (n=Nishanth@cpe-24-27-74-89.tx.res.rr.com) Quit (Read error: 110 (Connection timed out))
  • [16:47:17] * NishanthMenon (n=Nishanth@cpe-24-27-74-89.tx.res.rr.com) has joined #beagle
  • [16:47:45] <khasim> koen: is there a way to reduce the delay with Configuring libgnome-2-0
  • [16:49:45] <Crofton|work> http://goodliffe.blogspot.com/2008/05/cross-compiling-boost.html
  • [16:53:27] * gadiyar (n=a0393673@192.163.20.231) has left #beagle
  • [16:58:13] * khasim (n=a0393720@192.163.20.231) Quit (Remote closed the connection)
  • [16:58:58] <ds2> morning
  • [17:04:29] <keesj> hi
  • [17:17:01] * jconnolly (n=jconnoll@ool-18ba051f.dyn.optonline.net) has joined #beagle
  • [17:20:12] * rsalveti (n=salveti@200.184.118.132) Quit (Read error: 113 (No route to host))
  • [17:21:47] * rsalveti (n=salveti@200.184.118.132) has joined #beagle
  • [17:30:08] * BThompson (n=BThompso@cpe-76-185-93-11.tx.res.rr.com) has joined #beagle
  • [17:37:32] * gadiyar (n=Smash@192.163.20.231) has joined #beagle
  • [17:39:16] <gadiyar> hey all... Looks like the offending commit is between # good: [3ffec4e18484c34838fa341de3848306c29ecd5d] 24XX: PM: Move pm.c to pm24xx.c and sleep.S to sleep24xx.S
  • [17:39:42] <gadiyar> and # bad: [458776cfe389ff03bd6c56c47e059df0778cdfca] OMAP3430SDP: Enable config options CONFIG_OMAP_RESET_CLOCKS and CONFIG_SUSPEND
  • [17:42:03] <gadiyar> all 7 commits in between are compilable for beagle.
  • [17:42:38] <gadiyar> i had to leave for now. will get back to this tomorrow unless someone beats me to it
  • [17:48:12] * jconnolly (n=jconnoll@ool-18ba051f.dyn.optonline.net) Quit (Read error: 110 (Connection timed out))
  • [17:52:34] <jkridner> started a list of verified peripherals at http://beagleboard.org/hardware
  • [17:57:31] <jkridner> I'll look for new entries at http://del.icio.us/ with tags "beagleboard" and "peripheral". Once you've verified, please also add the "verified" tag.
  • [18:01:34] * jconnolly (n=jconnoll@ool-18ba051f.dyn.optonline.net) has joined #beagle
  • [18:11:27] * ldesnogu_ (n=ldesnogu@ven06-2-82-247-86-183.fbx.proxad.net) Quit ()
  • [18:26:01] * NishanthMenon (n=Nishanth@cpe-24-27-74-89.tx.res.rr.com) Quit (Read error: 110 (Connection timed out))
  • [18:32:58] * gadiyar (n=Smash@192.163.20.231) has left #beagle
  • [18:38:39] <shoragan> is there some downloadable sc card image that works?
  • [18:39:12] <shoragan> i got nothing on serial and dvi, so i think my sd card has something wrong
  • [18:47:54] <shoragan> jkridner, should i get anything on serial even if i have no sd card?
  • [18:50:41] <Crofton|work> shoragan, you should see 40T
  • [18:51:04] <shoragan> hmm...
  • [18:51:10] <shoragan> no 40T
  • [18:51:42] <Crofton|work> try pushing down both buttons, then release one then the other
  • [18:54:54] <shoragan> nothing :/
  • [18:55:14] <shoragan> but any character i enter is echoed back to me
  • [18:59:25] <shoragan> most times only the PWR led is on, but sometimes the other three are also on
  • [19:06:09] <shoragan> anything else i could try?
  • [19:18:01] * jconnolly (n=jconnoll@ool-18ba051f.dyn.optonline.net) Quit (Read error: 110 (Connection timed out))
  • [19:22:29] <keesj> shoragan: you say that every key press is echoed back, even when the power is off?
  • [19:23:17] <shoragan> no, only when i supply power via usb (so that the PWR led is on)
  • [19:26:52] * CyruzDraxs (n=CyruzDra@S0106001cf0719b7b.ok.shawcable.net) has joined #beagle
  • [19:27:18] <keesj> I need more experiance to be able to help you. I will be soldering a cable later today I hope
  • [19:34:13] <shoragan> keesj, i'm currently using a null-modem-cable...
  • [19:38:38] <shoragan> it doesn't even show up on the host's usb anymore
  • [19:38:45] <shoragan> (which it did in the beginning)
  • [19:49:41] <shoragan> jkridner, do have any suggestion?
  • [19:58:21] <alphaone> shoragan: Still no luck?
  • [19:58:26] <shoragan> no :/
  • [19:58:35] <alphaone> damn :-(
  • [19:58:52] <shoragan> the other leds haven't lit up again
  • [19:59:04] <alphaone> Strange..
  • [19:59:25] <alphaone> I think they stopped after I plugged in the audio cable
  • [19:59:25] <shoragan> also it dosn't show up on the usb bus any more
  • [19:59:55] <alphaone> When I looked here and it didn't show up at all
  • [20:00:39] <shoragan> it showed on when i tried it here at first (with sd card and no serial/hdmi)
  • [20:00:56] <alphaone> Hmm.
  • [20:01:18] <shoragan> the reference manual has some testpoints where we could measure the voltages
  • [20:01:24] <alphaone> We can try to measure if all the power rails are what they're supposed to be.
  • [20:01:28] <alphaone> Right :-)
  • [20:01:49] <alphaone> My multimeter is running low on batteries.
  • [20:02:03] <alphaone> I don't know if I can trust it...
  • [20:02:08] <shoragan> you have some spares?
  • [20:05:43] <shoragan> but at least the boot rom should show up :/
  • [20:06:28] <shoragan> alphaone, do you have another of these serial cables?
  • [20:09:55] <alphaone> shoragan: No, it needs a 9
  • [20:10:13] <alphaone> 9V battery. I don't have one of those
  • [20:10:28] <alphaone> I have a few serial cables.
  • [20:11:04] <alphaone> I could also solder the Beagleboard<->D-Sub thingy
  • [20:11:28] <shoragan> lets try some other serial cables first
  • [20:11:37] <shoragan> maybe it's just the cable
  • [20:11:53] <alphaone> shoragan: Okay, but with a normal cable it didn't work at all
  • [20:11:59] <alphaone> i.e. no echoes
  • [20:12:11] <alphaone> And that Null-
  • [20:12:30] <shoragan> i tried with my straight cable (with and without null-modem-converter)
  • [20:12:37] <alphaone> And that null-modem cable is good. I'm pretty sure of that
  • [20:12:44] <alphaone> And it worked?
  • [20:13:40] * jkridner|work1 (n=a0321898@nat/ti/x-b3401f3d2254d7b3) has joined #beagle
  • [20:13:47] <shoragan> no, it didn't
  • [20:13:51] <shoragan> same effect
  • [20:14:00] <shoragan> but sometime not all characters are echoed
  • [20:14:50] <alphaone> That is really weird.
  • [20:16:25] <shoragan> jkridner|work1, the board i just got today won't do anything besides lighting the PWR led...
  • [20:16:34] <shoragan> no 40 T on serial
  • [20:17:05] <shoragan> it showed up on the host's usb bus only once in the beginning, now it doesn't even do that
  • [20:23:34] * jkridner|work (n=a0321898@nat/ti/x-760d08013006a7fc) Quit (Remote closed the connection)
  • [20:46:14] <jkridner> shoragan: hi
  • [20:46:23] <shoragan> hi jkridner
  • [20:46:32] <jkridner> did you reprogram the flash?
  • [20:47:03] <shoragan> jkridner, no
  • [20:47:18] <jkridner> do you have anything connected to the audio out to hear the test tone at boot?
  • [20:47:38] <shoragan> alphaone tried that
  • [20:47:39] <jkridner> if you still hear the test tone, then it could be that there is a problem with the serial port.
  • [20:47:47] <jkridner> any tone?
  • [20:48:01] <shoragan> no, do i need to have a sd card to hear the tone?
  • [20:48:14] <jkridner> I shared a link to the serial adapter on http://beagleboard.org/hardware
  • [20:49:01] <jkridner> no, the rev B4 boards are shipped with a version of u-boot that puts out an audio tone and a 1280x1024 splash screen out the DVI-D.
  • [20:49:34] <jkridner> SD card is good to load a new version of u-boot.
  • [20:50:04] <shoragan> i've connected only the headphones
  • [20:50:25] <shoragan> then i plug in the usb connector, but i hear nothing
  • [20:50:44] <shoragan> PWR led is on
  • [20:51:28] <shoragan> do i need to do something special to prepare the board?
  • [20:55:05] <jkridner> no. you should see USR0, USR1, and PMU_STAT light up, in addition to PWR.
  • [20:55:26] <jkridner> if not, then there is a problem booting the u-boot that comes on the board.
  • [20:56:27] <jkridner> you should only see something on the serial port if you hold the USER button while applying power (via USB or power connector).
  • [20:57:06] <jkridner> I'm sorry, that isn't right, you'll see status messages, you just won't see the ...40T...
  • [20:57:21] <jkridner> so, you should see stuff on the serial port either way.
  • [20:57:30] <shoragan> 115200 8N1?
  • [20:57:44] <jkridner> yes.
  • [20:58:13] <jkridner> not the same serial connector as the SDP or Zoom MDK, if you have one of those.
  • [20:58:25] <jkridner> http://www.pccables.com/07120.htm is the right connection.
  • [20:58:40] * jkridner will be back after a bit.
  • [20:58:53] <shoragan> i used a straight 10pin to dsub connector + a null-modem cable
  • [20:59:20] <shoragan> it is the same as the one linked
  • [20:59:31] <shoragan> i see no status message at all
  • [20:59:52] <shoragan> and even when powering on with USER held, i see no 40T
  • [21:00:11] <shoragan> the other leds lit up the first time i tried it
  • [21:00:25] <shoragan> but no longer
  • [21:07:05] <shoragan> jkridner, i've got to go, but read the backlog tomorrow
  • [21:09:46] <shoragan> strangely if the board has power, all characters are echoed straight back to me
  • [21:20:25] <CyruzDraxs> So, has Beagleboard developed enough yet that someone could make a program that you load up in Linux or Windows to partition an SD card and setup an OS on it so all you have to do is plug it into the Beagleboard and go?
  • [21:22:22] <shoragan> alphaone, ping?
  • [21:22:41] <CyruzDraxs> I've seen a video of a whole DE running on it, but it didn't really have any indications of stability.
  • [21:23:24] <kulve> CyruzDraxs: afaik, it's not there yet
  • [21:23:46] <kulve> it's under development (both the hw and the sw) and you can't even buy the board yet
  • [21:26:10] <Crofton|work> alphaone, do you have a beagle?
  • [21:40:16] <alphaone> Crofton: No, I had shoragans for a couple hours today
  • [21:40:18] <alphaone> shoragan: pong
  • [21:40:32] <Crofton|work> ah
  • [21:40:36] <Crofton|work> no luck?
  • [21:42:04] <alphaone> no, it seems somewhat bricked... :-(
  • [21:42:17] <alphaone> Serial communication was erratic
  • [21:42:51] <Crofton|work> I had a problem with mine once, but I recovered by pushing both buttons, then releasing one c4 the other
  • [21:42:55] <alphaone> I got a few characters, but most was just garbage...
  • [21:43:07] <Crofton|work> hmm
  • [21:43:12] <alphaone> Hmm
  • [21:43:15] <Crofton|work> sounds like a flaky cable
  • [21:43:31] <alphaone> Yeah, but I checked that and it seemed okay
  • [21:45:14] <alphaone> Still could be. I might build another one just to be sure
  • [21:46:11] <Crofton|work> I'm fighting boost and gcc 4.3 atm
  • [21:46:27] <Crofton|work> when that builds, I can try the usrp with the beagle :)
  • [21:51:10] * CyruzDraxs (n=CyruzDra@S0106001cf0719b7b.ok.shawcable.net) Quit ()
  • [21:52:39] * docelic (n=docelic@78.134.195.216) Quit ("http://www.spinlocksolutions.com/")
  • [22:03:18] <alphaone> Crofton|work: Cool :-)
  • [22:03:47] <alphaone> I walked past a PS3 in a store today thinking whether I should get one for the usrp
  • [22:14:06] <jkridner> The 40T is actually just the only ASCII values in a string of numbers emitted when the serial bootloader starts. It could be the terminal package used translates non-ASCII values into something visible, resulting in more junk than just "40T".
  • [22:14:54] <jkridner> Getting the cable right is a big deal though. I'm working on making sure that it can be ordered with the board when we go for normal sale.
  • [22:15:55] * likewise (n=chatzill@82-171-51-231.ip.telfort.nl) has joined #beagle
  • [22:16:06] <likewise> koen: no problem, see your email box
  • [22:17:31] <koen> great!
  • [22:17:41] <koen> likewise: and I got gnash to configure :)
  • [22:18:58] <likewise> koen: excellent :-)
  • [22:19:28] <likewise> I'm still fighting with Thunderbird/GoogleMail not screwing up my kernel patches...
  • [22:19:31] <koen> only took me 3 hours of trial and error
  • [22:19:39] <koen> likewise: git send-email
  • [22:20:50] <likewise> koen: yes, but I mostly work against older kernels. Is git send-mail stand-alone? (I found 'nbsmtp' and just started to test it)
  • [22:23:05] <koen> yes
  • [22:23:13] <koen> git send-email <patchfile>
  • [22:23:35] <koen> it parses teh SOBs and ACK and CC:s those people
  • [22:23:53] <koen> or git-send email <dir>
  • [22:26:52] <likewise> fatal: Not a git repository at /usr/share/perl5/Git.pm line 197.
  • [22:30:26] <koen> heh
  • [22:30:33] <koen> gotta love git :)
  • [22:31:58] * Olipro (n=Olipro@unaffiliated/olipro) has joined #beagle
  • [22:32:31] <ds2> hey how important is getting a working sound driver + video driver for the beagle?
  • [22:33:09] <mru> it's pretty high on my list
  • [22:33:35] <ds2> is mid next week too late?
  • [22:33:36] <mru> but first I need to figure out how to get a picture at all
  • [22:33:58] <mru> ds2: do you have something up your sleeve?
  • [22:34:08] <ds2> mru: yes
  • [22:34:13] <koen> ds2: isn't there already a working sound driver?
  • [22:34:24] <ds2> what kernel version does it need to be in?
  • [22:34:36] <mru> ds2: please tell us more
  • [22:34:38] <ds2> koen: Not sure sure what state is sakoman's SoC driver in
  • [22:35:06] <koen> it works for most apps, except e17 manages to kill it
  • [22:35:18] <ds2> mru: the code exists, I need to get some 2 more nods and I can either post it or get it ported to the latest
  • [22:35:34] <koen> and it would be foolish to not base a driver on sakomans ASoC drivefr
  • [22:35:42] <ds2> koen: this driver is non SoC
  • [22:35:46] <koen> eeew
  • [22:35:51] <koen> that is so 1990's
  • [22:35:53] <mru> ds2: when you say video, what do you mean exactly?
  • [22:35:57] <ds2> koen: this is fully working AFAIK
  • [22:36:16] <koen> ds2: if it's like the one in the wtbu kernel, it is crap
  • [22:36:25] <ds2> mru: full TI driver + display lib
  • [22:36:45] <ds2> koen: so in your opinion, you wouldn't care for it then?
  • [22:36:51] <mru> I mean, what functionality does it provide?
  • [22:37:15] <koen> ds2: I wouldn't care for it and usb soundcards work great as well
  • [22:37:35] <ds2> mru: it _SHOULD_ have the same functionality as the WTBU driver but has been massaged for use on OMAP git (currently, the code is for a 2.6.24 snapshot of omap git)
  • [22:37:57] <ds2> so by extension, it should support svideo out + dvi out
  • [22:38:05] <mru> wtbu?
  • [22:38:23] <ds2> the TI part that originally did the 3430
  • [22:38:27] <koen> ds2: but to be fair, I suspect that the WTBU driver bugginess is caused by 'generic' omap3 issues, not by the driver itself
  • [22:38:32] <mru> what programming interface are we talking about here?
  • [22:38:54] <ds2> koen: know what bugs specifically? I have fixed a few
  • [22:39:15] <koen> wrong pitch is one
  • [22:39:23] <ds2> mru: frame buffer
  • [22:39:40] <koen> random distortion is another
  • [22:39:43] <ds2> koen: I did not see the problem on the SDP board
  • [22:39:57] <koen> the beagle is 'special' in that way
  • [22:40:00] <ds2> random distortion, I beleive is due to a TWL4030 issue
  • [22:40:13] <koen> ask sakoman how the beagle compares to the EVM
  • [22:40:14] <ds2> and I have seen it on older TWL4030's
  • [22:40:29] <ds2> the driver works on the EVM
  • [22:40:31] <koen> I suspect it's something like the twl loosing sync
  • [22:41:02] <ds2> koen: the distortion you are refering to sounds like clipping but it doesn't go away regardless of volume, right?
  • [22:41:59] <koen> that, and noise mixed in to it is another
  • [22:42:31] <ds2> koen: and you have seen it on the EVM?
  • [22:43:03] <ds2> I have only see it on a 2430SDP... the 3430SDP/EVM don't seem to show it
  • [22:43:19] <koen> I only have a beagle
  • [22:43:53] <ds2> anyways -- given that crude description, there is no interest in posting it?
  • [22:44:25] <koen> I think sakoman would like to see it to scan for tweaks :)
  • [22:44:41] <ds2> fair enough
  • [22:45:16] <koen> but I don't see any benefit in using it for the beagle kernel, since non-soc alsa driver are on their way out
  • [22:45:40] <ds2> if all goes well, I should be able to post it by mid week
  • [22:45:57] <ds2> I agree with you but there is also code maturity issues
  • [22:50:56] <mru> ds2: do you know any trick for getting a usable dvi output?
  • [22:51:09] <mru> whatever that u-boot splashscreen does, it works
  • [22:51:23] <mru> but I've failed to replicate the settings it uses in omapfb
  • [22:53:48] <koen> likewise: NOTE: package gnash-0.8.3-r2: task do_rm_work_all: completed
  • [22:53:49] <ds2> it is the display lib
  • [22:53:55] <ds2> you need that code
  • [22:53:59] <koen> likewise: for armv5, but still progress :)
  • [22:54:12] <ds2> it is a major pain to port from the WTBU kernel
  • [22:54:16] <mru> code I can write, I need the magic
  • [22:54:46] <ds2> grab the wtbu kernel, look at the stuff in arch/arm/mach-omap2/display.c, IIRC
  • [22:54:57] <koen> and dispc.c
  • [22:55:11] <ds2> yep, it is scattered all over the place
  • [22:55:19] <koen> and have a tissue ready in case you are allergic to spaghetti code
  • [22:55:37] <mru> so nobody knows what settings differ between the u-boot splash and current omapfb?
  • [22:55:43] <ds2> try a cell phone with 999 (or 911) pre entered
  • [22:56:08] <mru> damn it, all I'm asking for is a few register values
  • [22:56:24] <ds2> which U-boot? EVM or ?
  • [22:56:39] <mru> the one that displays the beagleboard image
  • [22:56:50] <mru> don't know how to describe it better
  • [22:57:28] <mru> and what's the procedure to talk to the dispc?
  • [22:57:31] <ds2> I haven't looked at the beagle u-boot code yet but I know on the EVM, U-boot's usage of resources is a lot like the TI kernel which is very different from the linux-omap kernel; it uses different clocks for one
  • [22:59:03] <RogerMonk> mru - what sort of display are you using - standard pc dvi input monitor?
  • [22:59:17] <RogerMonk> (or hdmi tv?)
  • [22:59:49] <mru> I've tried two hdmi TVs and a dvi monitor
  • [23:00:06] <mru> and the splash screen is fine
  • [23:00:18] <mru> so why can't I get the same timings in omapfb?
  • [23:00:23] <RogerMonk> yep, it's a clock/timings thing I think
  • [23:00:43] <mru> of course it is, the tv kindly tells me the mode is unsupported
  • [23:00:55] <mru> not so kindly, it gives no details about the bad mode
  • [23:01:02] <mru> so I have no idea what's being sent
  • [23:01:08] <koen> omapfb calculates clocks in a funky way
  • [23:01:14] <mru> tell me
  • [23:01:19] <koen> it uses an 48MHz pixclock for one
  • [23:01:30] <koen> instead of the 65MHz it's being told to use
  • [23:01:54] <ds2> that might be a DSS2 vs DSS1 thing
  • [23:02:11] <RogerMonk> koen - do you have the 12.18 wtbu kernel?
  • [23:02:12] <koen> so we get a 1024x768@57hz or something like that
  • [23:02:25] <RogerMonk> I think that should work - maybe worth a diff
  • [23:02:30] <koen> RogerMonk: only we one that are on beaglebord google code
  • [23:02:33] <mru> I want 1280x720
  • [23:02:40] <mru> and I know it's possible
  • [23:03:10] <koen> mru: edit lcd_omap3beagle.c
  • [23:03:17] <mru> I already modified omapfb to use the same dispc settings as that magic u-boot
  • [23:03:30] <mru> didn't help
  • [23:03:36] <koen> drivers/video/omap/lcd_omap3beagle.c
  • [23:03:40] <mru> some clock somewhere must be different
  • [23:03:48] <koen> you should be able to fill in 1280x720 there
  • [23:03:51] <mru> koen: I found that file
  • [23:04:06] <mru> it's just not giving me any timings any monitor I own will accept
  • [23:04:28] <koen> iirc sakoman looked into that this week
  • [23:04:50] <mru> was he the one who bought a new monitor in the end?
  • [23:04:54] <RogerMonk> (and Khasim is as well) - we're seeing the same issue at TI
  • [23:05:11] <mru> I guess I'm wasting my time here
  • [23:05:16] <koen> Crofton|work bought a new screen
  • [23:05:26] * darkerror05 (n=darkerro@cpe-72-230-185-21.rochester.res.rr.com) Quit (Read error: 60 (Operation timed out))
  • [23:05:34] <ds2> mru: are you sure you are using the same clock setup? there are apparently 2 clocks that can be used, DSS1 and DSS2
  • [23:05:51] <mru> I'm not sure of anything
  • [23:05:54] <mru> that's why I'm asking
  • [23:06:09] <mru> I have a 4000-page manual that *might* contain the answer somewhere
  • [23:06:10] <mru> or not
  • [23:06:16] <ds2> let me see if I can find the register that has the info
  • [23:06:33] <mru> would that be a dss register or a cm register?
  • [23:07:42] <RogerMonk> koen http://pastebin.com/m1dec5a57
  • [23:07:59] <koen> ds2: my current wishes for beagle are: working overlay for fast colour conversions, working SGX with git kernel, tv out
  • [23:08:03] <ds2> CM_FCLKEN_DSS
  • [23:08:03] <RogerMonk> any thoughts on this one.... (btw we're still going 14 hrs on :))
  • [23:08:11] <ds2> the lower 2 bits show what clock is enabled
  • [23:08:32] <koen> RogerMonk: like 37 :)
  • [23:08:49] <sakoman> mru: I have the same issue that you do -- most of my monitors don't work with current beagle timings
  • [23:08:55] <ds2> on the EVM, U-boot turns on a clock that the rest of the code doesn't know about and causes other issues
  • [23:08:57] <koen> s/like/line/
  • [23:09:00] <mru> ds2: that register reads as 1 here
  • [23:09:23] <ds2> let me decide that
  • [23:09:38] <mru> changing it makes no visible difference
  • [23:10:31] <sakoman> mru: I spent a little time on it earlier this week trying to get sync widths and front/back porch values more standard
  • [23:10:36] <ds2> that is DSS1 and I think the linux-omap tree uses DSS2 so that might account for the different clock rates you are seeing
  • [23:10:45] <ds2> the next bit is DSS2 enable
  • [23:11:10] <ds2> bbl
  • [23:11:33] <mru> the manual talks about setting all 3 bits to 1 during init
  • [23:12:06] <RogerMonk> http://pastebin.com/m4ba526cb
  • [23:12:33] <sakoman> mru: I think that there is a bug in the current git code that sets up the PLL to generate the clock
  • [23:12:34] <ds2> that enables both clocks
  • [23:12:48] <sakoman> Haven't had time to debug it
  • [23:12:55] <ds2> the side effect of that is PM/screen blanking can be broken
  • [23:13:09] <mru> sakoman: PLL4?
  • [23:13:25] <sakoman> don't recall offhand
  • [23:13:25] <mru> ds2: well, right now I have permanent screen blanking
  • [23:13:30] <koen> RogerMonk: ah, one of those stupid "needs internet to generate docs" errors....
  • [23:14:19] <ds2> I think if you check, there should be a message about a FATAL error
  • [23:15:09] <sakoman> mru: I kind of doubt that it is a blanking bug, since there *are* monitors that can sync to the signal
  • [23:15:12] <ds2> I am not being too detailed because I have only done bug fixes
  • [23:15:50] <mru> I'm sure those monitors exist
  • [23:15:51] <sakoman> My plasma display just complains "Out of Range" when I connect beagle
  • [23:16:02] <mru> but I'd rather use the ones I already own
  • [23:16:20] <mru> my samsung lcd tvs say the same
  • [23:16:29] <sakoman> And I know for certain that sync width, front and back porch values are just wrong, as is the pixel clock that is being used
  • [23:16:32] <mru> an hp monitor goes to sleep
  • [23:17:14] <mru> with a 48MHz pixel clock, there is no way of getting correct sync timings per hdmi spec
  • [23:17:27] <sakoman> Today is a holiday here, so I'm not working on it. Will look at it again later this weekend.
  • [23:17:48] <mru> 1280x720 should have a pixel clock of 74.25MHz
  • [23:18:20] <sakoman> mru: as I said, the timings are just wrong :-)
  • [23:18:53] <mru> unless you hadn't noticed, I'm the one to take no for an answer
  • [23:20:23] <sakoman> Even if you set up the correct values in the beagleboard lcd source file it still won't work because of the pixel clock generation bug -- that's the part I was looking at last
  • [23:20:41] <sakoman> Hopefully it will be a trivial fix
  • [23:21:04] <mru> doesn't look like it to me
  • [23:21:24] <sakoman> I'm an optimist ;-)
  • [23:21:26] <mru> what else does that clock drive?
  • [23:21:38] <sakoman> AFAIK, nothing
  • [23:22:31] <sakoman> Well, time for traditional 4th of July picnic!
  • [23:22:37] <mru> so which clock are we talking about anyway?
  • [23:22:38] <ds2> traditional?!
  • [23:22:40] <mru> there are so many
  • [23:22:49] <sakoman> Cookout
  • [23:22:51] <ds2> is that the one where you roast meats over a forest fire? ;)
  • [23:22:55] <sakoman> burn stuff over charcoal
  • [23:23:12] <sakoman> yeah, that's the idea
  • [23:23:30] <ds2> the big "fireworks" :)
  • [23:23:48] <sakoman> our 2500 acre campfire is finally out
  • [23:24:15] <sakoman> mru: pixel clock
  • [23:24:27] <mru> pixel clock is derived from another clock
  • [23:24:33] <mru> which is derived from yet another clock
  • [23:24:46] <mru> the last stage is the easy bit
  • [23:25:01] <mru> and it's done correctly afaict
  • [23:25:32] <sakoman> then why doesn't it work? ;-)
  • [23:25:45] <sakoman> and why is it the wrong frequency?
  • [23:25:46] <mru> because one of the master clocks is wrong
  • [23:26:29] <sakoman> Well, got to run. Hope you fix it first!
  • [23:26:29] <mru> the pixel clock is the dss functional clock divided by 2-255
  • [23:26:40] <mru> the functional clock is currently 96MHz
  • [23:26:42] <mru> go figure
  • [23:26:49] <sakoman> Otherwise I'll get back ion it later this weekend
  • [23:27:03] <sakoman> ttyl
  • [23:27:12] <mru> have fun
  • [23:32:35] <RogerMonk> koen - how can I turn off libgnomeprint html docs generation in OE? - can't see any reason why it's failing to load the file from web - and don't really need it anyway... I'd like to kick this thing off building the rest over night if possible?
  • [23:33:17] * mru has a 100% gnome-free system
  • [23:33:27] <koen> RogerMonk: edit its Makefile to exclude the docs subdir
  • [23:36:51] <RogerMonk> ... new it had to be an easy answer - continuing now... - thanks
  • [23:38:02] <RogerMonk> is that a known issue BTW?
  • [23:38:24] <RogerMonk> if so, can we make OE recipe patch the makefile?
  • [23:41:47] * likewise (n=chatzill@82-171-51-231.ip.telfort.nl) has left #beagle
  • [23:42:34] <koen> I don't see the issue on my builds
  • [23:42:42] <koen> but yes, OE can patch that out
  • [23:42:49] <koen> just write a patch for it :)
  • [23:42:58] <koen> (to Makefile.am, Makefile is a generated file)
  • [23:43:50] <RogerMonk> are you building the same rev of libgnomeprint-2.8.2 as me?
  • [23:45:06] <koen> bitbake beagleboard-demo-image works for me, if that's what you are asking
  • [23:46:06] <koen> 1:45 am, time for bed
  • [23:46:48] <RogerMonk> ok, thanks - me too... l8r
  • [23:56:46] * Olipro (n=Olipro@unaffiliated/olipro) Quit (Read error: 104 (Connection reset by peer))
  • [23:58:31] * Olipro (n=Olipro@unaffiliated/olipro) has joined #beagle