irclog2html for #oe on 20050321

00:07.28Ken|JLimeerr, i forgot. Which package do i have to rebuild in order for my Packages file to get updated?
00:07.31Ken|JLimetask-bootstrap?
00:09.49reenoo_'night all
00:17.01Ken|JLimei would like the rootfs to get tar'ed
00:18.10zeckehmm good nite
00:26.20RP'night zecke
00:26.24CIA-803RP 07 * r1.3123.1.1 10openembedded/packages/linux/linux-openzaurus_2.6.11.bb: openzaurus-2.6: Fix the mtd mapped PROM driver so it works and add definitions for tosa/spitz to it (patch commited to MTD CVS).
00:40.07*** join/#oe Laibsch (~leggewie@G1d95.g.pppool.de)
00:47.29*** join/#oe linuxwhore (~johnh@70-56-163-89.mpls.qwest.net)
00:51.14*** join/#oe zilt (~zilt@ottawa-dial-206-191-4-22.s-ip.magma.ca)
01:07.23*** join/#oe [g2] (~g2@g2.nslu2-linux)
01:20.37*** join/#oe exastra (~go@l5.org)
01:31.09*** join/#oe tigrux (~tigrux@charybdis-ext.suse.de)
01:31.25*** part/#oe tigrux (~tigrux@charybdis-ext.suse.de)
01:32.06_drak0_Ken|JLime: bitbake -c clean task-bootstrap; bitbake bootstrap-image
01:36.37*** join/#oe noclouds (~mhfan@60.166.164.26)
01:43.14Ken|JLime_drak0_, thx
01:44.10*** join/#oe tomimo (~kurre@a84-231-39-238.elisa-laajakaista.fi)
01:44.26*** join/#oe kurre2 (~tomimo@a84-231-39-238.elisa-laajakaista.fi)
01:53.04*** join/#oe Marsupilami23 (~Marsupila@152.139-60-66-nokia-static.surewest.net)
02:01.14kergothKen|JLime: hm? i epxlained this yesterday.
02:01.33kergothKen|JLime: the Packages file is updated whenever an _image_ is built.  it gets created only for the purposes of buliding an image.
02:03.23Ken|JLimeI missed it in the log last night, sorry. Btw where do you want me to send new packages?
02:27.15*** join/#oe wrobbie (~rob@cm168.sigma180.maxonline.com.sg)
02:39.27emteKen|JLime, why would you send the package anywhere?
02:39.41emteyou only use them for your personal feed or image
03:06.55*** join/#oe Marsupilami23 (~Marsupila@152.139-60-66-nokia-static.surewest.net)
03:06.58*** join/#oe Marsupilami24 (~Marsupila@152.139-60-66-nokia-static.surewest.net)
03:43.45gb2mark: 4:13:05
03:43.50gb2doh!
04:24.35chouimat|Zzznight
04:35.03*** join/#oe treke|home (~ggilbert@69-175-32-221.ventca.adelphia.net)
04:50.16*** join/#oe hufnus (~slonsiki@m283436d0.tmodns.net)
05:54.36*** join/#oe xep_ (misha@porter-kresge-apts-59-178.resnet.ucsc.edu)
06:09.31*** join/#oe bluelightning (~bluelight@203-173-245-90.akl.ihugultra.co.nz)
06:25.44kergothKen|JLime: you mean new .bb files? send them to the mailing list.
06:35.27emteheya kergoth
06:35.41emtei dont suppose you play with automatons ...
06:41.25bluelightninghi guys
06:44.05*** join/#oe _law_ (~law@mail.stiftadmont.at)
06:56.17*** join/#oe Pend-Gone (Pendalar@h54.115.140.67.ip.alltel.net)
07:35.36*** join/#oe greg72 (~szobatudo@ns1.sysdata.hu)
07:37.59greg72hi koen&kergoth
07:49.44*** join/#oe dkey| (~dkey@L0008P27.dipool.highway.telekom.at)
07:58.46*** join/#oe florian_kc (~fuchs@dns.kernelconcepts.de)
07:58.48florian_kchi all
08:04.00bluelightninghi florian_kc
08:24.15*** join/#oe schurig (~schurig@p54B28BEE.dip0.t-ipconnect.de)
08:30.53*** join/#oe xiaoxiao (~root@220.248.128.10)
08:31.25xiaoxiao...
08:31.56RPmorning all
08:32.09CoreDump|homemorning
08:32.14schurigmorning
08:32.40xiaoxiao16:33 now
08:33.04wrobbieindeed
08:33.04bluelightninghi RP
08:33.21bluelightninghi xiaoxiao, CoreDump|home, schurig, wrobbie
08:33.40xiaoxiaohi
08:33.45kergothi hate a document review today
08:33.46kergother
08:33.46kergothhave
08:33.53kergoththough i hate it too
08:33.56CoreDump|homexiaoxiao: in your timezone maybe ;) It's 09:34 here
08:33.58wrobbie:)
08:33.58kergothand i havent yet written the doc we're reviewing.
08:34.19xiaoxiaohehe .kergoth
08:34.41treke|homekergoth: so you're not sleeping much tonight? :p
08:35.05CoreDump|homeouch
08:35.11kergoth"uh .. my laptop's hard drive .. uh .. was .. uh .. formatted by a virus.. yeah."
08:35.26treke|homebah
08:35.34treke|homethis is what the excuse server is for
08:35.37kergothhehe
08:35.48treke|home"Vendor no longer supports the product "
08:35.51kergothon another note, Eurotrip is funny as hell
08:36.32treke|homewe're waiting for [the phone company] to fix that line
08:36.43schurigkergoth: you're going to Europe?   Where to?
08:36.44treke|homegod I can't believe how many times I've used that one
08:36.50kergothschurig: heh. its a movie :)
08:36.58kergothtreke|home: hehe
08:37.11treke|homeeuropean vacation?
08:37.20*** join/#oe obi (~obi@zischt.oettinger.im.karlshof.de)
08:37.28schurig~lart kergoth
08:37.29kergothnah, eurotrip.  its the road-trip like one, that had the bloodhound gang reference in the preview
08:37.44treke|homehaven't seen that
08:37.46kergoth"the hat! the hat!  the hat is on fire!"  "we dont need no water, let the mother fucker burn!"
08:37.50kergoth"no, really, its not a joke. its on fire."
08:38.35treke|home'The final Star Wars movie will be an emotional "Titanic in space"'
08:38.39treke|homewell that settles it
08:38.42kergothhah
08:38.47kergothscary
08:38.52schurigkergoth: get the doc for your chinese alarm clock. Copy the text. Maybe no one realizes it.
08:38.57kergothhehe
08:39.13treke|homeif I had any hope it wouldnt absolutly suck before, it's long gone
08:40.57kergothhey, i could call in sick.  for my last week at work
08:40.59kergothhehe
08:40.59hrw|workgello
08:41.25treke|homekergoth: no vacation time saved up?
08:41.46kergothdont think so, pretty sure i'm out
08:41.51kergothvaca & sick is the same time bank
08:41.58kergothjust one big personal time bank deal
08:42.18wrobbielast week? you quitting?
08:42.48kergothyep.  the job wasnt a good fit for me.  been having trouble getting shit done
08:42.53kergothbored out of my mind
08:43.12kergothtreke|home: you read the article about the coder who turned bicycle messenger?
08:43.20wrobbienew job requirement: no doc reviewing..
08:43.30kergothhttp://www.kuro5hin.org/story/2005/3/19/133129/548
08:46.43bluelightninghmm, strange... it seems on my desktop that CF cards are now mounting as "msdos" instead of "vfat"
08:51.19treke|homethat started happening to me in 2.6.9
08:51.58bluelightninghmm...
08:52.18bluelightninggoogling has not turned up a fix yet... I'm using 2.6.11 at the moment
08:52.39treke|homevi /etc/fstab
08:52.52treke|homeedit line for compact flash card. set fstype to vfat
08:53.48kergothyou can probalby just manipulate /etc/filesystems
08:53.50kergothchange the order
08:53.55kergothmove vfat above msdos
08:54.29treke|homeprobably would fix it as well
08:54.59*** join/#oe XorA (~dp@81-178-213-232.dsl.pipex.com)
08:55.10XorAmorning
08:55.11CoreDump|homekbot: hello
08:55.12bluelightninghmm... /etc/filesystems points to /proc/filesystems on gentoo
08:55.12kbotCoreDump|home: Error: "hello" is not a valid command.
08:55.15bluelightninghi XorA
08:55.31CoreDump|homekergoth: is your kbot a supybot?
08:55.35bluelightningCoreDump|home: hmm, how rude
08:55.39kergothCoreDump|home: yeah, been playing with its codebase
08:55.39bluelightning:)
08:55.42CoreDump|home:)
08:55.51CoreDump|home@lart me
08:55.56CoreDump|homeheh
08:56.07CoreDump|homewhere did you find a lart database if i may ask?
08:56.11kergothi didnt
08:56.15kergoththats the only one in it
08:56.16kergoth:)
08:56.19kergothbeen looking for one
08:56.20CoreDump|homemine shiped w/o one
08:56.23kergothsame.
08:56.31treke|homebluelightning: iirc the system looks at /etc/filesystems first and then /proc/filesystems if it isnt present
08:56.46kergothnormally you'd have /etc/filesystems so filesystem kernel m odules get loaded when using -t auto
08:56.49CoreDump|homewell, mine didn't know _any_ larts at all. Maybe the dian maintainer fucked up?
08:56.50treke|homesounds like gentoo just pretends like they are the same
08:56.54kergothsince naturally fs's that are modules wouldnt be listed in /proc/filesystems
08:57.02bluelightningtreke|home: ok, I guess I should break the symlink then
08:57.06kergothCoreDump|home: mine didnt either.  it doesnt ship with a lart db at all.
08:57.11CoreDump|homedam
08:57.22treke|homeubuntu for example doesnt even include an /etc/filesystems
08:57.25CoreDump|homewell, i entered most of blootbots (ie ibot's) larts
08:57.35kergothCoreDump|home: heh, thats what i was planning on doing
08:57.40CoreDump|home:)
08:57.50kergothtreke|home: that makes sense.  a symlink really makes no sense
08:58.05treke|homekergoth: well it is gentoo
08:58.21kergothhehe
08:59.03bluelightningthe question would be why is msdos above vfat in /proc/filesystems though...
08:59.25treke|homem comes before v :)
08:59.33bluelightningtreke|home: you may well be right :)
08:59.40kergothi'd bet its filesystem registration order, rather than alphabetical
08:59.46kergothand then someone changed the link order in the kernel...
08:59.48kergothbam
08:59.56kergothjust theorizing, but that seems a likely possibility
09:00.36bluelightningoh well, at least I know where to look now... thanks
09:02.09*** join/#oe alakdan (~alakdan@210.213.196.113)
09:03.50treke|homethis is sad. Had this laptop for nearly a year now and today is the first time sleep has ever worked
09:04.08*** join/#oe dkey (~dkey@L0008P27.dipool.highway.telekom.at)
09:04.20koengood morning all
09:07.07*** join/#oe AxTheB (~ax@cit2.natur.cuni.cz)
09:23.36CIA-803CoreDump 07 * r1.3131 10openembedded/packages/base-passwd/base-passwd_3.5.9.bb: Fix PR...
09:23.40CIA-803CoreDump 07 * r1.3130 10openembedded/ (8 files in 3 dirs): Upgrade base-passwd from 3.5.7 to 3.5.9 as 3.5.7 is no longer available from upstream
09:23.42*** join/#oe ade|desk (~adavey@194.200.143.249)
09:23.52RPXorA: I found and fixed the speed problem on the c7x0
09:24.54ade|deskhowdy folks
09:25.11RPmorning ade|desk
09:25.47bluelightninghi ade|desk
09:29.31alakdananybody here got some kernel image for zaurus with keybdev module?
09:29.42alakdankernel image for 5600
09:30.21XorARP: I noticed, flashed that ROM last night
09:30.31XorARP: cheers for that
09:42.20RPmickey|zzZZzz: In the latest c7x0 opie image, the fonts and taskbar all seem a lot smaller than they used to be. I'm not sure if that's be design or not...
09:45.06*** join/#oe zecke|aw1y (fakeuser@pD9E7CFBA.dip0.t-ipconnect.de)
09:48.25CIA-803koen 07 * r1.3132 10openembedded/packages/meta/task-bootstrap.bb:
09:48.25CIA-8task-bootstrap.bb:
09:48.25CIA-8<PROTECTED>
09:48.32kergothheh
09:48.47CoreDump|homeindeed
09:49.11CoreDump|homedidn't know that, sorry
09:49.49koenit has been fixed before the ml has been flooded :)
09:50.29CoreDump|homewho knows how long 3.5.7 was missing from the server? heh maybe a couple of days
09:50.54koendunno
09:51.04koenit's been in my DL_DIR for quite a while
09:51.23CoreDump|homesame here, i tend to keep my "sources" dir between trashing tmp
09:52.01kergothkeep between?  that makes it sound like you do it manually. i hope not :)
09:52.46RPMy sources dir still has some of the very first c7x0 2.6 kernel patches in it :)
09:53.48wrobbiehmm. How can I populate root_fs the way I like? Do I have to copy & change base_files? (which means also copy task-bootstrap and bootstrap-image)
09:54.15kergoththats a pretty broad question
09:54.36kergoththere's a lot more to a rootfs than the base directories in the base-files ipk
09:54.54CoreDump|homekergoth: IIRC "sources" is inside tmp per default. I've moved it to ..
09:55.42kergothCoreDump|home: yes, the point was, its just a matter of setting DL_DIR.  and in doing that, you arent 'keeping the sources between trashing tmp', you're just keeping them outside of tmp, always
09:55.44wrobbieI could copy a rootfs and modify it, but then later updates are tricky..
09:55.47kergothheh
09:55.49kergothwrobbie: ?
09:55.55kergothwhy dont you spit out what you're trying to do
09:55.58CoreDump|homeheh, right :)
09:56.02kergothwithout more information, we cant really answer that question
09:57.08CoreDump|homewrobbie: the rootfs.bbclass supports all kinds of bash hax0ring before completing the image
09:57.21wrobbieok, I want to create a rootfs that has some of the stuff created by bootstrap-image removed, and other things added.
09:57.30kergothso create a custom -image .bb
09:57.44kergothadd packages, remove packages
09:58.00kergothif something you want to add to the image isnt in oe, you've got a decision to make
09:58.14schurigwrobbie: I made my own (not in OE) nfsroot.bb packages which installs the IPKs I want (and not more) and also recursive copies the nfsroot/* files into it for misc config settigs that I don't want in some .ipk files
09:59.16kergothbleh, i havent done that doc yet
09:59.17kergothhmm
09:59.43wrobbieah, that sounds like what I wanna do.. also copy some config files in, to replace standard stuff (/etc/hosts, etc)
09:59.45schurigkergoth: do a /quit until you wrote the doc, IRC can be distracting
09:59.53kergothheh, very true
10:00.09CoreDump|homethat must be the understatement of the year :)
10:01.02schurigwrobbie: for /etc/hosts I used the OE way by adding a oe/packages/netbase/netbase/ramses/hosts file to OE's bitkeeper
10:01.46schurigwrobbie: other places for my own config files are in base-files/base-files/<MACHINE>
10:02.21schurigwrobbie: you could place yours there, too
10:02.37schurigwrobbie: OE has great per-machine or per-distro override capabilities!
10:02.43wrobbieok (having a look)
10:03.42wrobbieramses is what you're building?
10:03.50greg72moin, schurig
10:04.05schuriggreg72: how's your PowerPC build ?
10:05.00schuriggreg72: Jo reggelt  (I hope this is "good morning", I always confuse jo reggelt, jo eszokat)
10:05.26*** join/#oe zedkatuf (~dfutak@82-32-58-45.cable.ubr08.azte.blueyonder.co.uk)
10:05.42koenjo napot
10:05.57schurigviszont latasra
10:05.59koen(that's "good day", right?)
10:06.14greg72no success yet. I asked for help on various lists (oe, uclibc) but no response yet
10:06.28greg72schurig: you are right
10:06.31schuriggreg72: hehe, you're pioneering it
10:06.53schurigNem beszelek madyarul
10:07.22greg72schurig: don't say <tschuess> unless you are really leaving :-)
10:07.40schurigkoen: that was (or is supposed to be): I don't speak hungarian
10:07.40koenschurig: "I don't speak hungarian"?
10:07.48greg72koen is right
10:08.00koenmagyarul :)
10:08.04greg72oh yes
10:08.24greg72it's time to return
10:08.53greg72I spent 15 months in Leipzig, so we could talk in German if you wish, schurig
10:09.32schuriggreg72: hehe, I found many Hungarians that knew German quite good, might be a remnant of the K.u.K. time ...
10:09.59greg72well, it was a long time ago...
10:10.13schuriggreg72: especially in Budapest and in the western part of the country
10:10.22greg72that's right
10:12.33zecke|aw1yre
10:12.58koenmoin zecke
10:13.42Ken|JLimeK.u.K ?
10:13.44Ken|JLimehaha
10:14.11greg72Kaiserlich und Königlich, if I recall correctly
10:14.16Ken|JLimein swedish that would mean something in the lines of "dick"
10:14.41schurigKen|JLime: this is something like "King and Imporer"
10:14.54CoreDump|homes/I/E/ ;)
10:15.02greg72Emperor you mean?
10:15.08schurigKen|JLime: Austria used to have some countries conquered, and the Austrian King called himself also Emporer
10:15.21Ken|JLimeAha
10:15.28schurigKen|JLime: and so the monarchy there were not only a kingdom, but also an empire
10:16.23schurigKen|JLime: at that time Austria, Hungary, Slowakia, Czech, some Balcan countries where "one" country, "one" in a sense the Austrians ruled everything
10:17.00*** join/#oe chii (chii@chii.utility.freenode)
10:17.12bluelightningthe "Austro-Hungarian Empire"
10:17.30RPDoes anyone know why /etc/hotplug/mmc.agent uses /dev/mmc/part$n for the mmc card location? Which device does that work on?
10:17.38bluelightningRP: ipaq
10:17.40koenRP: ipaqs
10:17.52schurigbluelightning: yeah. And a german abbreviation for this is "K.u.K.", e.g. the K.u.K.-Hofreitschule (kind of a hippodrom with special horses in Vienna)
10:18.02bluelightningschurig: ah ok
10:18.17*** join/#oe cbrake (~cbrake@a22c174.neo.rr.com)
10:18.44RP2.4 ipaq or 2.6 ipaq?
10:18.51XorARP: there was a new one posted to oe list that worked on both
10:18.55bluelightningRP: 2.4... not sure about 2.6
10:19.15RPThe one posted to the list won't work on both as they introduced sharpisms
10:19.35koenI think you can work around that in 2.6 using some udev magic
10:19.42RP2.4 Zaurii don't use hotplug mmc do they?
10:20.02RPIf not, this is easy to fix...
10:20.07XorARP: ah sorry, I missed how he had coded it at the top
10:21.31RPkoen: udev makes sure the devices exist but doesn't take care of mounting the devices
10:22.19koenRP: can it be used to create /dev/mmc/* instead of /dev/mmcd*?
10:23.27RPkoen: I suspect it could but I'd rather we went with the 2.6 style device names and sorted the problem out once and for all
10:24.36CIA-803koen 07 * r1.3133 10openembedded/packages/xcursor-transparent-theme/xcursor-transparent-theme_0.1.1.bb:
10:24.36CIA-8xcursor-transparent-theme_0.1.1.bb:
10:24.36CIA-8<PROTECTED>
10:24.42schurigbluelightning: don't worry, my knowledge about NZ history is zero :-)
10:24.55koenI was under the assumption that block devices used /dev/<dir>/(part*,disc} in 2.6
10:25.17bluelightningschurig: yes but our history is much shorter of course :)
10:25.34bluelightningand perhaps not as interesting
10:25.58schurigbluelightning: maybe NZ goes out and conqueres Australia ...   then you'd have something to learn & remember
10:26.05bluelightningheh
10:26.11bluelightningthey would eat us alive
10:26.15bluelightningand all of our sheep too
10:26.17bluelightning;)
10:27.04bluelightningheh
10:31.21hrw|workkoen: basically 2.6 use old /dev/ naming style. udev can create anything anyway
10:34.40RPkoen: udev uses /dev/mmcblk?p?, devfs uses /dev/mmc. I have no idea which is "correct" :-/
10:35.02hrw|workRP: rather first as devfs is obsolete
10:35.27koenI do like the ideas of subdirs, it looks a bit more organized
10:36.28hrw|workkoen: how much entries you have in /dev/ tree?
10:36.45XorAkoen: The idea is udev keeps naming in userspace, the kernel gets on with its real job of driving devices
10:36.57koenhrw|work: too much
10:37.09koenXorA: I know, I just like the devfs naming scheme
10:37.31hrw|workkoen: then configure udev to give you devfs naming with compat symlinks
10:37.56koenhrw|work: I was doing that on my h2210
10:40.07hrw|workwho is "Michael Kropfberger"? he recently find many src_uri changes which he floods ML - maybe he should get somekind of write access?
10:40.50RPI don't think udev has any rules for mmc which is why it defaults to the mmcblk?p? naming
10:42.22RPI would add that /dev/mmc/ is wrong as what if you had two mmc card slots. It should really be /dev/mmc0...
10:44.21hrw|workRP: /dev/mmc/disc0 /dev/mmc/disc1
10:44.33hrw|workif you want devfs style
10:45.13hrw|workor even /dev/mmc/host0/bus0/slot0/disc and /dev/mmc/host0/bus0/slot1/disc ;)
10:45.32RPGiven devfs is obsolete, I'm tempted to go with udev's style. That means we don't need to patch udev...
10:48.10*** join/#oe mallum (~mallum@host81-153-50-205.range81-153.btcentralplus.com)
10:51.54Ken|JLimeI just want the kernel-modules to be installed in the rootfs
10:52.00Ken|JLimehow do i do that?
10:55.07hrw|workKen|JLime: add it into machine.conf for example
10:55.32Ken|JLimeas an RDEPEND?
10:55.41Ken|JLimeBOOTSTRAP-EXTRA-RDEPEND
10:55.42Ken|JLime?
10:56.06*** join/#oe zinga (~arnaudb@186.80-203-227.nextgentel.com)
10:59.14CIA-803RP 07 * r1.3127.1.1 10openembedded/packages/linux-hotplug/ (linux-hotplug_20040920.bb files/mmc.agent): linux-hotplug: Update mmc.agent to work with 2.6 kernels + udev
10:59.46kergothRP: eh, thats not right
11:00.00kergothat the time the hotplug script executes, the device in /dev isnt guaranteed to exist.
11:00.29kergoththe only way to properly do it with udev is to install a script into dev.d, and mount on the _device creation_ event, rather than the _hotplug_ event
11:00.40RPkergoth: I've just updated the script, not created it...
11:00.50RPkergoth: It does work as I just tested it...
11:00.52kergothyes.. and?
11:01.00kergothyou didnt properly update it to work with udev
11:01.02kergothdespite your claims
11:01.11kergothbecause, once again, the device isnt guaranteed to exist at that point
11:01.17kergothif it works, its luck.  timing.
11:01.21kergothit could break at any time
11:01.34RPWell, I altered it to use udev naming and its less broken than when I started...
11:02.32kergothgood, but its the wrong solution.  those scripts shouldnt be used for automount in a udev environment, only in a non-udev environment.  needs to be fixed
11:02.42kergothjust making sure people know.
11:03.36RPOk. Binning the mmc.agent script when using udev which is beyond my oe skills though...
11:04.00kergothi dont think that should be a build time thing.
11:04.07kergoththere are all sorts of situations whre udev could be installed, but not used
11:04.12kergothor what have you
11:04.15kergothor built, but not installed.
11:04.48RPSo how does the mmc.agent script know whether to do anything or not?
11:04.54kergothi'd 1) split out the mmc hotplug scripts into a seperate package, 2) make the .agent simply exit immediately if udev is in use, and 3) install a dev.d script.
11:05.13kergothjust look in /dev for the udev db.  the same way soem other startup scripts check for udev
11:07.46hrw|workKen|JLime: both _DEPEND and RDEPEND - look into collie.conf for example
11:08.09hrw|workhi mickeyl
11:11.01bluelightninghi mickeyl
11:12.55*** join/#oe _drak0_ (~rob@ip68-98-148-202.dc.dc.cox.net)
11:14.46hrw|workhmm.. kdm from kde 3.4 still looks awful (when compared to gdm)
11:15.05XorAhrw|work: kdm is just xdm, gdm was re-written
11:15.30hrw|workXorA: and it's login prompt cannot be made smaller;(
11:16.03mickeylmorning
11:18.16Ken|JLimeWhen adding my linux-kernel-2.6.11 to the RDEPEND, DEPEND it complains about not filling the DEPS for usbutils & pciutils, both of which is not very usefull for the jornada6xx
11:20.29CIA-803mickeyl 07 * r1.3126.1.1 10openembedded/ (3 files in 2 dirs): upgrade cumulus to 1.2.1 and lower the default preference of the cvs version
11:26.48zeckeI hate GPIOs :)
11:28.28CIA-803mickeyl 07 * r1.3136 10openembedded/packages/zaurus-updater/ (zaurus-updater.bb spitz/updater.sh.c3000): update updater.sh for c3000
11:28.47Ken|JLimesetting DEPEND = "usbutils pcituils" but not adding them to RDEPEND would make them fill the deps but not get included in the final bootstrap-image?
11:32.06RPkergoth`zzz: What you were saying shouldn't be a problem as udev is always called before the hotplug agents
11:33.05XorARP: it actually says in udev manual it is not designed to mount devices, use another tool
11:34.00RPXorA: It ok to use hotplug as the ordering problem kergoth mentioned isn't true
11:34.19schurigzecke: especially GPIO 5 ?
11:35.17hrw|work~lart kde 3.4 documentation
11:35.20zeckeschurig: Yeah, GPIO5 is evil(tm). I know that the switch is working
11:35.41kergoth`zzzRP: udev is a daemon.  it does hotplug event reordering and the like.  can you guarantee that the udev hotplug callout waits until the daemon is done creating the node before it returns?
11:35.59kergoth`zzzi dont recall, but i'd really advise checking your assumptions
11:36.35RPkergoth`zzz: udev is listed before default.hotplug in /etc/hotplug.d/default
11:36.38kergoth`zzzyes, and?
11:36.45kergoth`zzzthe agent just communicates with the daemon
11:36.58kergoth`zzzyou're saying that the agent is fully synchronous?
11:37.07RPAs XorA says, filesystem mounting doesn't have ant place in udev
11:37.08kergoth`zzzand waits on the daemon to create the node before hte agent exits?
11:37.19RPIf it doesn't its a bug
11:37.39kergoth`zzzi didnt say to use udev to mount devices. i said to use a dev.d script to call out a tool to mount devices.
11:38.40RPThat amounts to the same thing...
11:38.55RPI'll try and find out how synchronous udev/hotplug is...
11:39.15kergoth`zzz<PROTECTED>
11:39.15kergoth`zzz#! /bin/bash
11:39.15kergoth`zzzUSBMOUNT=/usr/local/bin/usb-mount
11:39.15kergoth`zzz# Unfortunately udev only creates the device nodes AFTER this script is run.
11:39.15kergoth`zzz# We sleep for 2 seconds in the background and then run usb-mount. This will
11:39.17kergoth`zzz# mount the device.
11:39.20kergoth`zzz(sleep 2 && $USBMOUNT) &
11:39.27kergoth`zzzquoted from the usb-storage mounting project, linked to by the udev page
11:40.04*** join/#oe mithro (~tim@dsl1-83.gw1.adl1.airnet.com.au)
11:40.07kergoth`zzzyes, please do.
11:41.12RP<PROTECTED>
11:41.14CIA-803koen 07 * r1.3133.1.1 10openembedded/packages/matchbox-themes-extra/matchbox-themes-extra_0.3.bb: matchbox-themes-extra_0.3.bb: hh.org -> o-hand.com
11:44.31*** join/#oe [g2] (~g2@g2.nslu2-linux)
11:49.54alakdanwhich bb file should I put in local.conf if I want to build a kernel for poodle, I need also to reconfigure the kernel as I want to include keybdev module
11:55.32*** join/#oe marcansoft (1337@80.26.245.253)
11:56.48schurigalakdan: you don't put BB-files into your local.conf
11:57.09schurigalakdan: look at conf/local.conf.sample for an example what could be inside
12:01.10hrw|workre
12:02.26*** join/#oe Laibsc1 (~leggewie@G17d5.g.pppool.de)
12:03.39CoreDump|home!test
12:05.50CIA-803koen 07 * r1.3138 10openembedded/packages/kbdd/ (4 files in 2 dirs): kbdd_cvs.bb: Apply patch from Paul Eggleton
12:07.18*** join/#oe stealth (~stealth@user-0c6sdm3.cable.mindspring.com)
12:08.07stealthhi
12:08.11stealthwould it be possible to compile klogic http://www.a-rostin.de/klogic/ for opie?
12:11.06zeckeschurig: I#m confident that one can
12:14.18*** join/#oe stevenh (~xev@65.167.23.2)
12:15.43bluelightningthere is also KTechLab: http://ktechlab.fadedminds.com/
12:15.50bluelightning(slightly different focus)
12:16.18stealthcool
12:17.32stealthso whats the first step, svn bitbake? where do i put the klogic/ktechlab sources? /stuff/build ?
12:19.29zeckestealth: minimal porting will be required
12:21.35schurigstealth: I'd get a staging area, e.g. the libs and headers and cross-compiler. I don't know if you can download that from somewhere. If not, use OE to create this yourself.
12:21.37schurigstealth: it
12:21.42schurigstealth: it's somewhat time-consuming
12:22.18schurigstealth: once you can compile Qt programs for your target, you can get the source code of klogic/ktechlab and try to compile it, opie-fy it etc
12:22.46schurigstealth: once you have patches to opie-fy one of the programs, you (or one of us) can create an bitbake recipe for OpenEmbedded
12:33.22*** join/#oe cdbot (~supybot@hentges.net)
12:33.33CoreDump|homecdbot: config channel supybot.reply.whenAddressedBy.chars ""
12:33.34cdbotCoreDump|home: The operation succeeded.
12:33.40CoreDump|home~lart cdbot
12:33.40cdbotCoreDump|home: Don't know, don't care
12:33.47CoreDump|homesigh
12:33.57CoreDump|home~config channel supybot.reply.whenAddressedBy.chars ""
12:34.35CoreDump|home~lart cdbot
12:34.35cdbotCoreDump|home: ??
12:34.35zecke~botfight ?
12:34.47ibotGeorge- stop it!
12:34.47CoreDump|home......
12:34.47CoreDump|homecdbot: part
12:34.47*** part/#oe cdbot (~supybot@hentges.net)
12:34.47*** join/#oe cdbot (~supybot@hentges.net)
12:34.47CoreDump|home~lart cdbot
12:34.48cdbotCoreDump|home: ??
12:34.49CoreDump|homedamn thing
12:35.13bluelightningeh? what was that last bit?
12:35.37cdbotouch!
12:36.43schurigI'm getting nasty
12:36.48schurig/kick cdbot
12:36.48cdbotschurig: Dunno?
12:40.42CoreDump|homecdbot: config channel supybot.reply.whenAddressedBy.chars ""
12:40.43cdbotCoreDump|home: The operation succeeded.
12:40.46CoreDump|home~lart cdbot
12:40.47cdbotCoreDump|home: I have no frelling idea!
12:41.03CoreDump|homebleh, kinda bugged
12:41.32CoreDump|homeups
12:41.41bluelightning@lart test
12:41.52bluelightningtoo many bots... ;)
12:41.58CoreDump|homeibot: config channel supybot.reply.whenAddressedBy.chars ~
12:43.35schurigCoreDump|home: create some new channel, e.g. #botarena :-)
12:43.43CoreDump|homeschurig: heh
12:43.52CoreDump|homei just want it to shut up in this channel
12:44.06CoreDump|homethe same command worked in two other channels grr
12:45.00CoreDump|homecdbot: config channel supybot.reply.whenAddressedBy.chars !
12:45.01cdbotCoreDump|home: The operation succeeded.
12:45.14CoreDump|home!test
12:45.15cdbottest is BS
12:45.19CoreDump|home~lart me
12:45.24CoreDump|homeahh :)
12:46.11Kristoffernothing provides kernel-modules
12:46.23Kristofferargh
12:47.30schurig!lart CoreDump|home
12:48.14*** join/#oe schurig (~schurig@p54B28BEE.dip0.t-ipconnect.de)
13:20.59hrw|work<PROTECTED>
13:21.04hrw|worklibxineopie failed
13:21.22mickeyldid it? let me try
13:23.00bluelightningargh who broke it this time
13:23.09hrw|workmickeyl: maybe it;s because my builddir is over 3 weeks old and have many older versions..
13:23.24bluelightningtime to rm -rf TMPDIR?
13:24.04hrw|workbluelightning: give me machine which will build meta-opie in time shorter then 24h...
13:24.33koenewi546.ewi.utwente.nl
13:24.51koenthat should do it in approx 6 hours
13:26.06bluelightningnight all
13:26.50hrw|workn8 bluelightning
13:27.20alakdanhow long does it take to build a kernel using bitbake?
13:28.53hrw|work| xine_decoder.c: In function `vorbis_decode_data':
13:28.53hrw|work| xine_decoder.c:222: error: too many arguments to function `vorbis_synthesis'
13:28.53hrw|work| make[3]: *** [xine_decoder.lo] Error 1
13:28.53hrw|work| make[3]: Leaving directory `/home/hrw/zaurus/bb/build/0301/tmp/work/libxine-opie-1.0-r0/xine-lib-1.0/src/libvorbis'
13:29.04mickeylcrumbs
13:29.06mickeylsame here :/
13:29.07hrw|workprobably need newer/older vorbis..
13:29.27hrw|workwe have two versions of vorbis in OE and some stuff want one, some another
13:31.10hrw|workanyway - time for lunch (disabled mediaplayer2 and met-opie build goes on)
13:32.07*** join/#oe univac^ (~univac@ami175.neoplus.adsl.tpnet.pl)
13:32.12univac^hi
13:32.14koen/home/koen/OE/build/tmp/familiar//rootfs/usr/lib/ipkg/info/qte-font-smoothsans-110.postinst: line 5: /usr/sbin/update-qtfontdir: No such file or directory
13:32.27koen(probably harmless, but annoying)
13:32.55univac^what is the good mp3/ogg player, divx player, www browser, jabber client, txt editor for _opie_?
13:33.07koenask #opie
13:33.17univac^ok
13:34.35mickeylkoen: when does that happen?
13:34.43mickeylkoen: on image building time?
13:34.54koenmickeyl: yes
13:35.08koenwith poodle
13:35.24koenand c7x0 judging from the missing rootfs's
13:35.47mickeyli'll take a look at next image building time
13:37.44CIA-803RP 07 * r1.3134.1.1 10openembedded/packages/linux/linux-openzaurus_2.6.11.bb: openzaurus-2.6: Compile fixes from Dirk for tosa
13:45.11zeckeschurig: GPIO5 behaves friendly now
13:45.22*** join/#oe cbrake (~cbrake@mail.accelent.com)
13:47.16schurigzecke: on MNCI Ramses GPIO5 is 'nc' :-)
13:47.19greg72Did anybody use http://www.bitmover.com/bk-client.shar successfully?
13:52.10zeckegreg72: yes
13:53.12greg72I've compiled it and tried to use update bk://openembedded.bkbits.net/openembedded
13:53.13greg72<PROTECTED>
13:53.46koenshouldn't that be bk pull?
13:54.10KristofferHOW can i just make the modules end up in the rootfs
13:54.16Kristofferi cant imagine why this would be so damn hard
13:54.41zeckeKristoffer: in your machine depend on the modules?
13:54.56Kristofferi know, and its very good if i just need a couple
13:55.01Kristofferbut i want all the modules
13:55.18Kristofferdo i need to make a list of all the modules i want included?
13:55.24koenRDEPENDS on kernel-modules
13:55.35koenthat should be a catch-all meta-package
13:55.49Kristofferah, thx ill try that
13:55.59mickeylthat doesn't work w/ 2.6 yet though
13:56.39koen2.6-c7x0 or 2.6 in general?
13:57.00mickeyl2.6 in general
13:57.11KristofferEh
13:57.20mickeylwe should use objdump to gather the intermodule dependencies on 2.6
13:57.32Kristofferoki, so i need to specify each module by itself?
13:57.34mickeylnot very high priority atm. though
13:57.39koendidn't pb work on that last month?
13:57.52mickeylKristoffer: if on 2.6, yes. if not, depending on kernel-modules should be sufficient.
13:57.55mickeylkoen: did he?
13:58.24koenmickeyl: He did some work on kernel.bbclass and was working on 2.6
13:58.32greg72$ ls
13:58.32greg72demo.sh  Makefile  sandbox  sfio  sfioball  sfioball.c  sfio.c  system.h  update  update.c
13:58.36koenIdidn't check the actuall diffs
13:58.39Kristoffer2.4 is quite broken for superh so it will never be used. I'll just specify each and every module...
13:58.41mickeylkoen: aha. must've missed that
13:58.47greg72sfio sfioball and update are the executables
13:58.58mickeylKristoffer: that's better anyway, since otherwise you end up with a pretty huge rootfs
13:59.07hrw|workgreg72: does demo.sh works for you?
14:00.36greg72it does something...
14:00.42greg72but nothing is on the display
14:00.50hrw|workhmm.. worked for me
14:00.54greg72Press return to get the audiofile repo as of cset 1.100
14:01.12CIA-803CoreDump 07 * r1.3140 10openembedded/packages/netbase/netbase_4.20.bb: Update netbase from 4.19 to 4.20 since 4.19 vanished upstream
14:01.20hrw|workgreg72: and?
14:01.33hrw|workgreg72: press 'return' or 'enter' key
14:01.39greg72hrw|work: are you behind a firewall? I am.
14:01.55greg72hrw: I did
14:02.00hrw|worknot
14:03.05*** join/#oe xep (misha@porter-kresge-apts-59-178.resnet.ucsc.edu)
14:06.12*** join/#oe cyn (~cyn@adsl-068-209-224-046.sip.mem.bellsouth.net)
14:06.14*** join/#oe qfh (~qfh@ppp-62-245-163-223.mnet-online.de)
14:06.29hrw|workgreg72|break: ./sfioball bk://openembedded.bkbits.net/oe.deprecated oe
14:06.47hrw|workgreg72|break: works - but I'm not firewalled
14:07.39greg72hrw|work: but why oe.deprecated?
14:08.38greg72bk://audiofile.bkbits.net/audiofile/: Connection timed out
14:08.44greg72demo.sh told me
14:09.28hrw|workgreg72: first bk repo which I used and it is small enough to be test one
14:12.50RPmickeyl: I've tried a recent c7x0 opie-image. It all seems to work fine but I did notice all the fonts, icon sizes and taskbar seems small though. Is that by design?
14:13.46mickeylno
14:14.05mickeylcat /home/root/Settings/qpe.conf
14:15.00chouimatmorning
14:15.25RPDoes c7x0 do something special with qpe.conf? My /home/root was a symlink to a none existant /root when I installed this image which may have screwed things up...
14:15.56mickeylyes, we need a special qpe.conf for c7x0
14:16.01mickeylsee opie-taskbar/c7x0/qpe.conf
14:16.05mickeyli.e. we need
14:16.29mickeyluseBigPixmaps = 1
14:16.29mickeylSmallIconSize = 24
14:16.29mickeylBigIconSize = 48
14:16.29mickeylFontFamily = smoothsans
14:16.29mickeylFontStyle = Regular
14:16.29mickeylFontSize = 12
14:16.29mickeylto make a fairly sane default appearance
14:17.07RPNone of that is set. This will have been that symlink problem. I'll reflash as who knows what else I broke...
14:17.36mickeylok
14:17.37mickeyl*phew*
14:17.53RPThis is a gotcha for anyone with a symlink in /home/root. I don't know if it can be worked around?
14:18.19RPmickeyl: Sorry to cause a panic. I'm trying to help with some testing :)
14:18.24mickeylwell, the default qpe.conf from /opt/QtPalmtop/skel is always only installed when there is not one existing
14:18.26mickeylRP: sure :)
14:20.25*** join/#oe hufnus (~slonsiki@m8b1836d0.tmodns.net)
14:20.42*** join/#oe cdbot (~supybot@hentges.net)
14:21.10RPmickeyl: /opt/QtPalmtop/skel doesn't exist...
14:23.09mickeylyes, it's /opt/QtPalmtop/etc/skel
14:23.59RPmickeyl: Somehow that file wasn't used... :-/
14:24.36mickeylcan you take a look at the opie init script`?
14:24.46mickeylit should install it when there's is no one in $HOME/Settings/
14:25.54RPmickeyl: It does indeed look correct...
14:26.13RPWhat's the betting, I removed the symlink with opie running :)
14:26.35mickeylah heh
14:27.55zeckemickeyl: we could simply copy the sekeleton to $OPIEDIR/etc
14:28.53*** join/#oe hufnus (~slonsiki@m8b1836d0.tmodns.net)
14:30.24RPmickeyl: I wiped out /home/root/Settings and its looking a lot better :)
14:30.50RPmickeyl: With the mmc.agent change I commited, automounting of SD cards works under opie
14:31.35RPAuto screen rotate also works nicely. It doesn't suspend when I close the clamshell though which would be the final touch :)
14:31.42RPOtherwise it loos great :)
14:32.03XorARP: eeep, dont suspend when closed
14:32.11RPlooks
14:32.25XorARP: I use mine closed
14:33.56RPXorA: So do I. I'd settle for turning the backlight off :)
14:34.17zeckeRP: good job
14:34.21XorARP: screen off would work for me
14:34.22mickeylRP: It should suspend
14:34.28mickeylRP: if you tell it to do that
14:34.32mickeylRP: light-and-power --> hinge action
14:35.06RPmickeyl: aha - "display off" is what I want :)
14:35.15mickeylyep
14:35.19RPThat should probably be the default...
14:35.20mickeylon kernel 2.4 we had a race condition
14:35.26mickeylwhen you chose "suspend"
14:35.41mickeylthe device woke up and fell into sleep again
14:35.46mickeyli hope that's vanished w/ 2.6
14:36.38RPmickeyl: "suspend" also works as expected here
14:37.20RP(no race)
14:37.34XorAscreen off works here
14:37.52mickeylgreat
14:38.09mickeylis there any reason why we should ship 2.4 with oz 3.5.3 ?
14:38.14mickeyl(the LEDs don't count)
14:38.48XorAmickeyl: GPE
14:39.03mickeylno, serious reasons only
14:39.05mickeyl(just kidding)
14:39.21florian_kcmickeyl: !
14:39.31XorAmickeyl: for opie I am finding 2.6 so much more solid than 2.4
14:39.41RPGPE is broken under 2.6. As far as opie is concerned, 2.6 wins in every way
14:40.07RPThere are no missing items of 2.6 functionality now...
14:40.12zeckeflorian_kc: do you know how much work it would be to fix GPE to cope with a 2.6 kernel?
14:40.21zeckeRP: really good job
14:40.51mickeylok, we ship both then and recommend 2.6 for opie and 2.4 for gpe.
14:40.52RPzecke: thanks :)
14:41.00florian_kczecke: Basically i don't see a reason why it shouldn't work with 2.6. Do you know what is going wrong?
14:41.14mickeylunless someone fixes the issues until friday
14:41.21zeckeflorian_kc: No I'm asking because I don't see a reason as well :)
14:41.35RPI couldn't figure out what was going wrong with gpe. It looks like some kind of init race...
14:41.37zeckeflorian_kc: probably one needs to adjust the xmodmap or such
14:41.53XorAzecke: nothing wrong with the xmodmap
14:42.05XorAzecke: X will not yeild the VT for suspend
14:42.24XorAgpe-dm and windowmaker argue over the root window so you cant actually log in
14:42.28RPThe keymap is fine. The problem was the window manager wasn't loading as it claimed one was already loaded when it wasn't
14:42.36hrw|workmickeyl: atleast for collie there is no such problem ;)
14:42.58XorAesd is unkillable and unfreezable so you can never suspend anyway
14:43.20florian_kcXorA: Usually noone uses windowmaker with gpe...
14:43.22koenhow enlightened to be unkillable
14:44.18XorAflorian_kc: Im a mong, matchbox is what I meant to type
14:44.26florian_kcah
14:44.46XorA~larts fingers for not following instructions
14:44.51XorA~lart fingers for not following instructions
14:45.11florian_kcXorA: i don't think that's a 2.6 issue... nothing should run matchbox while gpe-dm is running.
14:45.31mickeylhrw|work: well, john has been doing more work lately.
14:45.57RPhrw|work: collie-2.6 is slowly getting there. It will catch up...
14:46.02XorAflorian_kc: just going through all the problems I have with GPE today
14:46.44RPHas anyone tried a recent 2.4 gpe image? Maybe that has the same issues?
14:47.11hrw|workanyway I'm sceptical one in collie/2.6 subject
14:50.13mickeyltrue, usb and SD will probably be never working
14:50.22mickeyland without these two, it's hard to convince users to use 2.6
14:51.00CoreDump|homeindeed
14:52.10hrw|workand when someone will try to use wifi card to connect collie with net (instead of usbnet) then where will save something?
14:53.03XorAmickeyl: I wonder if the guys behind Z emulator have the access to write SD/USB drivers
14:53.58mickeylXorA: i don't think they have source to the SD.
14:56.21zeckeRP: do you know if mmcblock needs to be compiled as module?
14:58.58RPzecke: No, it works fine compiled in
15:02.24*** join/#oe prpplague (~dave@mailhost.amltd.com)
15:03.29*** join/#oe BlackRain (~camillo@213-140-6-126.fastres.net)
15:06.50zeckeRP: hmm maybe I'm mistaken but if I do a mmc_detect_change after getting the removal interrupt
15:06.56CIA-803pb 07 * r1.3138.1.1 10openembedded/packages/gtk+/ (gtk+-1.2_1.2.10.bb gtk+-1.2-1.2.10/timezone-fix.patch): apply patch from Michael Kropfberger to fix timezone problem
15:07.15zeckeRP: udev does not remove the device, probably I should some time until calling mmc_detect_change?
15:08.52RPzecke: You need to debounce the switch if the hardware doesn't do that for you.
15:08.58*** join/#oe webmind (~webmind@217-195-236-172.dsl.esined.net)
15:09.49zeckeRP: yeah, I've spied at your code :)
15:09.57koenI guess hiring a programmer to fix it in software was cheaper than putting to flipflops in the design...
15:10.05RPzecke: With the way mmc/sd slots are built, the switch says the card has left whilst there is still electrical contact with the card. It can half init the card in the time it takes to break electrical contact, hence the timer in the corgi code...
15:10.09koens/to/two/
15:10.59zeckeRP: but mmc_detect_change schedules_work as well, but it sounds realistically that the switch is triggered
15:11.05zeckeRP: but the card is still present
15:12.06RPThe schedule_work doesn't delay the work. It can happen instantly after that call is made
15:12.32zeckethat is right though
15:12.51RPWhen I said above is that yes, the switch triggers and the card is ejecting but can still be in electrical contact
15:13.38zeckeRP: thank you
15:14.14zeckeRP: C7xx: Would you declare the keyboard layout final
15:14.20zeckeRP: or do we've any short comings?
15:14.46RPI'm not aware of any and I can't see it changing much
15:14.53*** join/#oe webmind (~webmind@217-195-236-172.dsl.esined.net)
15:15.16RPzecke: Do you mean the kernel driver or the userspace keymap?
15:25.57zeckeRP: Admitingly I don't know who loads it
15:27.13RPThe kernel driver has a basic set of keys. The kernel's default keymap is incorrect though. We therefore load one at runtime which corrects it. Its done by initscripts in oe.
15:27.53zeckeRP: My question basicly is do we've some modifiers we don't use yet?
15:31.33RPzecke: I don't think there are any more
15:34.16chouimathi zecke
15:48.06*** join/#oe Cwiiis (~cwiiis@user-214-207-151-83.e7even.com)
15:50.46*** join/#oe dkey (~dkey@L0007P08.dipool.highway.telekom.at)
15:50.56schurigoh no, http://www.treke.net/oe/source/libc_sources.redhat.com__20040726.tar.gz doesn't fetch and NOTE: Update cvs://anoncvs@sources.redhat.com/cvs/glibc;module=libc;date=20040 doesn't work either
15:51.04schurigcvs [update aborted]: end of file from server (consult above messages if any)
15:51.06schurig:-(
15:56.03koenschurig: http://ewi546.ewi.utwente.nl/OE/libc_sources.redhat.com__20040726.tar.gz
15:56.29*** join/#oe marcansoft (1337@80.26.245.253) [NETSPLIT VICTIM]
15:56.55*** join/#oe _drak0_ (~rob@ip68-98-148-202.dc.dc.cox.net) [NETSPLIT VICTIM]
16:00.20*** join/#oe mrz80 (1000@caledonia.cns.ufl.edu)
16:08.06*** join/#oe [g2] (~g2@g2.nslu2-linux)
16:08.49*** join/#oe cbrake (~cbrake@mail.accelent.com) [NETSPLIT VICTIM]
16:12.08alakdanwierd, after the long compile I am getting this make[2]: arm-linux-gcc-2.95: Command not found
16:12.09alakdanbut when I echo $PATH
16:12.09alakdan/usr/local/arm/2.95.3/bin:/stuff/bitbake/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/bin/X11:/usr/games
16:12.58*** join/#oe hufnus (~slonsiki@m8b1836d0.tmodns.net)
16:13.42*** join/#oe Toi (~pleemans@dD577A871.access.telenet.be)
16:16.00XorAalakdan: you probably started bitbake before you changed the path?
16:16.26XorAalakdan: or you didnt rename the arm-linux-gcc file
16:20.01alakdanXorA: the first one is true (started bitbake before realizing that I forgot to add the 2.95.3/bin to PATH ). How do I remedy it?
16:20.17hrw|work<PROTECTED>
16:20.28alakdanok
16:21.02alakdanthanks :)
16:23.26hrw|work~lart multisynk
16:23.54*** join/#oe Crofton (~balister@h80ad91b7.dhcp.vt.edu)
16:25.45XorAhrw|work: heh heh multisync, otherwise known as multiscreup
16:25.51XorAmultiscrewup
16:26.36hrw|workforget it - opie can be used without pim
16:27.00hrw|workI just wanted to check does opie-contacts finally sorts when click on headers
16:39.21*** join/#oe jbs (~jbs@10001200047.0000006537.acesso.oni.pt)
16:40.43alakdankernel build is now complete. Is it possible to just ipkg install the kernel ipk files or I have to build an image or something?
16:41.05alakdanand reflash
16:41.28Bernardohi
16:41.50hrw|workyou can flash kernel and then install newer modules
16:41.53hrw|workhi Bernardo
16:49.33*** join/#oe reenoo (~r@p5489D81A.dip.t-dialin.net)
16:50.50reenooafternoon
17:04.13*** part/#oe Laibsc1 (~leggewie@G17d5.g.pppool.de)
17:09.24*** join/#oe Timelord (~TL@64-147-65-66.access-3.intertex.net)
17:10.03*** join/#oe Cwiiis (~cwiiis@user-214-207-151-83.e7even.com)
17:15.54*** join/#oe zap (~zap@217.170.93.196)
17:21.44CosmicPenguinbk is sloww today
17:27.41*** part/#oe ade|desk (~adavey@194.200.143.249)
17:38.09CroftonHow can I get diff to ignore files only in one of the directories?
17:44.40*** join/#oe treke|laptop (~ggilbert@69-175-32-221.ventca.adelphia.net)
17:51.20alakdanweird, Im getting this while installing a module ERROR: Cannot satisfy the following dependencies for kernel-module-keybdev:
17:51.20alakdan<PROTECTED>
17:51.50*** join/#oe zap (~zap@217.170.93.196)
17:54.06reenooalakdan: from the topic: "This is not a distribution support channel"
17:55.11alakdanreenoo: Im trying to build a kernel with keybdev support, as kbdd is not supported in oz :)
17:57.49*** join/#oe prpplague^2 (~dave@mailhost2.amltd.com)
18:02.02*** join/#oe xep (misha@porter-kresge-apts-57-219.resnet.ucsc.edu)
18:02.30*** join/#oe OSS542 (~jhart@l215080.ppp.asahi-net.or.jp)
18:06.53koenmickeyl: does OZ ship kbdd?
18:07.50mickeyldoubtful. i don't even know what that's for
18:09.23koenah, it only shows up in ipaq-common
18:09.37koenmickeyl: it's a userspace driver for serial keyboards
18:09.45mickeylaha
18:09.50mickeyldo we need that?
18:09.58mickeylif so, put it into handheld-common instead of ipaq-common
18:10.13koenif you want to support IR keyboards and some bt keyboards
18:11.15koenalakdan wants it :)
18:11.26mickeylwho am i to disobey :D
18:12.24koenthe reason why I asked is kbdd depends on 2 kernel modules: uinput and keybdev
18:12.33mickeylhmm
18:12.37mickeylnot good
18:12.46mickeylbetter go w/ standalone install then
18:12.46koenand if alakdan is correct, those are disabled in the OZ kernels
18:13.32mickeyli see. alakdan - want to cook up a patch enabling those?
18:13.32alakdanyes its disabled
18:14.05alakdanmickeyl: created the /dev/misc/uinput via mknod. but the kbdd still will not run and outputs failed to open uinput device: No such file or directory
18:14.23alakdangrrr :)
18:16.14alakdani just modified the defconf-poodle by hand and enabled INPUT support UINPUT and KEYBDEV
18:20.49*** join/#oe treke|laptop (~ggilbert@69-175-32-221.ventca.adelphia.net)
18:20.58alakdanis devfs disabled for zaurus? how do I enable it?
18:21.06treke|laptopmornin
18:21.08treke|laptopit's disabled
18:21.12mickeylwe don't support devfs for the zaurii
18:21.23treke|laptopkergoth`zzz: wake up. you have a meeting
18:23.16alakdanmickeyl: seems kbdd needs /dev/devfs/misc/uinput or /dev/input/uinput . problem is I tried creating manually /dev/input/uinput via mknod but I still get failed to open uinput device
18:23.45mickeylyeah. i don't think devfs will improve things though. you have to find out what happens there
18:23.47*** join/#oe treke|laptop (~ggilbert@69-175-32-221.ventca.adelphia.net)
18:25.56alakdanmickeyl: do I have to rebuild the kernel with debug options or something? do I have to uncomment the following on local.conf? # DEBUG_BUILD = "1"
18:25.56alakdan# INHIBIT_PACKAGE_STRIP = "1"
18:26.08treke|laptopto do what?
18:26.59alakdantreke|laptop: I dont know where to start on tracking down the problem of uinput module no logs or something
18:27.09mickeylalakdan: nothing in dmesg ?
18:27.16mickeylsure that your device nodes are right?
18:27.23mickeylstrace relevals nothing?
18:27.26mickeylreveals even
18:27.45alakdanhangon
18:30.15alakdannothing in dmesg, ls -al /dev/input/uinput crw-r--r--    1 root     root      10, 233 Jan  1 00:30 /dev/input/uinput
18:31.25*** join/#oe [cc]smart (~smart@gw.ptr-62-65-149-158.customer.ch.netstream.com)
18:38.20*** join/#oe andersee (~andersee@codepoet.org)
18:40.06*** join/#oe treke|laptop (~ggilbert@69-175-32-221.ventca.adelphia.net)
18:46.44alakdankoen: mickey_away: ok, I got it now. my next problem is debugging modification I made for kbdd (to make my IR keyboard work) :) Thanks a lot! Gotta sleep now
18:54.23*** part/#oe alakdan (~alakdan@210.213.196.113)
18:58.50*** join/#oe jh_ (~user@69.144.85.206)
19:03.34*** join/#oe treke|laptop (~ggilbert@69-175-32-221.ventca.adelphia.net)
19:09.37*** join/#oe _alwin_ (alwin@cable-81-173-164-147.netcologne.de)
19:14.56CosmicPenguinI like xfce, but its hard to keep up when you are the package maintainer
19:16.10florian_kcbbl
19:16.52*** join/#oe _law_ (~law@213.173.86.202)
19:35.30*** join/#oe gremlin[it] (~gremlin@194-185-170-166.f4.ngi.it)
19:35.59gremlin[it]good evening ...
19:36.18koenhey gremlin[it]
19:36.26gremlin[it]hi koen ;)
19:36.38gremlin[it]ppp problem is knows ?
19:36.42gremlin[it]ppp problem is known ?
19:39.01koenppp doesn't build?
19:39.32gremlin[it]yep
19:39.47gremlin[it]and others
19:39.53gremlin[it]<PROTECTED>
19:43.18gremlin[it]mhhh crypt, setkey, encrypt not found and are all libc functions :( ...
19:43.39gremlin[it]or a missing include in pppd os something wrong with libc
19:47.30gremlin[it]mhh >>>
19:47.34gremlin[it]#ifdef HAVE_CRYPT_H
19:47.35gremlin[it]#include <crypt.h>
19:47.35gremlin[it]#endif
19:47.36gremlin[it]<<<
19:47.47gremlin[it]in chap_ms.c
19:48.18gremlin[it]it's a clue ... maybe configure step fail something ;(
19:48.39koenwe'll see
19:49.01gremlin[it]£define HAVE_CRYPT_H=y .. it's ok
19:50.05*** join/#oe ar_ (~ar@port-ip-213-211-231-76.reverse.mdcc-fun.de)
19:50.19*** join/#oe darkschneider (~gab@213-140-22-64.fastres.net)
20:00.30*** join/#oe pb_ (~pb@2002:5168:d330:1:a00:1fff:fe06:93c)
20:00.50pb_hi all
20:00.57reenoohey pb_
20:02.55*** join/#oe dkey (~dkey@193.170.62.146)
20:03.58gremlin[it]koen, gotcha !!!
20:04.18gremlin[it]i found on Makefile of pppd >>>
20:04.26gremlin[it]ifneq ($(wildcard /usr/lib/libcrypt.*),)
20:04.27gremlin[it]LIBS += -lcrypt
20:04.27gremlin[it]endif
20:04.27gremlin[it]<<<
20:04.34gremlin[it]and fail ...
20:04.46pb_heh, that's no good
20:04.50pb_~lart pppd
20:05.41gremlin[it]pb_ are u connected via ipv6 ?
20:07.39koenI thought I debugged the sixxs software when I got ipv6 connectivity yesterday....
20:07.54pb_gremlin[it]: always.
20:07.55koen... only to find out my uni vpn also assings ipv6 :)
20:08.02pb_heh
20:08.12gremlin[it]pb_, i notice just now ;)
20:09.57koenactually my tunnel wasn't working because I'm not italian
20:10.08koenitalian people have access to avaya
20:10.21gremlin[it]koen, really ?!?!?!
20:10.23koenwith works behind a NAT firewall
20:11.05koenhttp://www.sixxs.net/tools/ayiya/
20:11.06gremlin[it]but ADSL and broadband conenctions are really poor in italy ;( and really not cheap as in others country :( ...
20:11.46pb_I hear that GPRS is very cheap in Italy, though.  I guess you can't win them all.
20:12.29gremlin[it]mhh mybe isn't expensive.. i think in general cells fee in italy ar cheap ...
20:13.27koenfrom scratch to glibc packaging: 24 minutes
20:14.02koenwhich would be less when using raid
20:17.34Croftonif you list a .patch file in the SRC_URI, does the ;patch=1 mean patch -p1?
20:17.48koenno
20:17.52koenpnum=1
20:18.11Croftonhmm, what does the patch=1 mean?
20:18.16koenpatch=1 marks it as a patch
20:18.19Croftonah
20:18.28Croftonwould pnum=1 be the default?
20:18.45koeniirc, yes
20:18.54Croftonthat would make sense
20:19.19CroftonThanks
20:47.15*** join/#oe darkschneider (~gab@213-140-22-64.fastres.net)
20:56.20Kristoffer<PROTECTED>
20:56.20Kristoffer|        kernel-image-2.6.11 kernel-module-3c589 kernel-image-2.6.11 kernel-image-2.6.11 kernel-image-2.6.11 kernel-image-2.6.11 kernel-image-2.6.11 kernel-image-2.6.11 kernel-image-2.6.11 kernel-image-2.6.11 kernel-image-2.6.11 kernel-image-2.6.11
20:56.20Kristoffer| Nothing to be done
20:56.42Kristofferim going nutters :D
20:58.07Kristofferi cant imagine why it doesnt work
20:58.31pb_well, do you have kernel-image-2.6.11.ipk on hand?
21:03.53KristofferI've used the linux-epia-2.6.11.bb for template
21:04.28pb_so, is that a yes or a no?
21:05.10Kristofferwell, ... it builds the kernel,but doesnt package it
21:05.20pb_okay, so that's a "no".
21:05.30Kristofferye, kinda.
21:05.39pb_you need to fix it so that the kernel is packaged, or fix your bootstrap dependencies to not require kernel-image.
21:05.46pb_depending on whether you want the kernel in the rootfs or not.
21:06.07pb_heh.  there's no "kinda" about it.  either you have the file or you don't, and in your case you don't.
21:08.20Kristofferi know :P just realised what i must have done wrong when you asked
21:22.41*** join/#oe chouimat (~dieu@r2351064.cidc.net)
21:43.17pb_hi mickeyl
21:43.50mickeylgood evening phil
21:44.32*** join/#oe bipolar (~bipolar@146.145.26.90)
21:50.37gints|sleepmickeyl: recent changes in oe qpf dir causes fonts not apear while install as package
21:52.56mickeylthat's sad
21:53.48gints|sleepshould ask hrw?
21:54.10mickeylyeah. i didn't touch the fonts since adding the autosplit feature months ago
21:54.13gints|sleepwill try to check tomorow
21:54.16mickeylk
22:16.06*** join/#oe qfh (~qfh@ppp-62-245-163-223.mnet-online.de)
22:21.12Bernardomickeyl: gpsd init.d script isn't working
22:21.49Bernardoit is trying to run a /etc/rc.d/init.d/functions
22:22.04BernardoI'll try to fix it today or tomorrow
22:22.08mickeylvery good
22:24.08BernardoI don't know if the latest gpsd releases still work with qpegps
22:24.30mickeylthey should
22:24.35mickeylsince they sent their changes upstream
22:24.47*** join/#oe hufnus (~slonsiki@mbf1d36d0.tmodns.net)
22:24.52mickeyli'm going to test after your fix
22:25.47Bernardook
22:25.59Bernardotime to restart my desktop - kde 3.4
23:04.22*** join/#oe ibot (ibot@apt.bot.TimRiker.active.supporter.pdpc)
23:04.22*** topic/#oe is OpenEmbedded Developer Lounge | http://bugs.openembedded.org/ (l/p guest/guest) | This is not a distribution support channel | The OE mailing list is back, but everyone needs to resubscribe. | We have a bitbake-dev mailing list for discussions on the core. If you're interested in BitBake-Ng, please subscribe. | LCA 2005 Embedded Miniconf - http://www.openembedded.org/miniconf
23:07.02w3rwolfkoen: doesn't work...
23:07.25koenwith an up to date OE repo?
23:08.07w3rwolfhm... about one week old..
23:08.13w3rwolf3-4 days.. mom
23:09.08koenw3rwolf: in any case update, I pushed a new mozconfig for minimo this week
23:09.31w3rwolfkoen: i placed the CVSDATE in the .bb file, now it works
23:10.04w3rwolf*sigh* all the compiling stuff once again :(
23:10.34koenif you are really lazy: http://familiar.handhelds.org/snapshots/v0.8.2/20050321/gpe-image-simpad-20050321123457.rootfs.jffs2
23:10.57w3rwolfui
23:12.00w3rwolfthx, but I want to try building my own image first...
23:12.34w3rwolfbtw.. what's this minimo-stuff?
23:14.07koenmini mozilla
23:14.38w3rwolfso i won't need firefox at all?!
23:14.51w3rwolfor should i preferr firefox?
23:16.01koenit's a stripped down firefox
23:16.21koenworks pretty wel
23:16.38koendougt is pressing it really hard
23:16.54CosmicPenguincept the UI is depressing for anyone not on a 320x200
23:17.30koenCosmicPenguin: have you seen the new UI?
23:17.40*** join/#oe stealth (~stealth@user-0c6sdm3.cable.mindspring.com)
23:18.05CosmicPenguinkoen: when you say new....
23:18.14koenlast week....
23:18.47CosmicPenguinOh, no I haven't
23:18.50CosmicPenguinURLS?
23:19.15koenhttp://handhelds.org/~florian/gallery/view_photo.php?set_albumName=Applications&id=port_27393
23:19.44CosmicPenguinWell, now, isn't that purty
23:20.55koenCosmicPenguin: and  http://handhelds.org/scap/port.4710.png
23:21.14koenthe little bar on the bottom is a small version of that menu
23:21.26CosmicPenguinYeah, that looks real nice
23:21.41*** join/#oe Crofton (~balister@66-207-66-26.black.dmt.ntelos.net)
23:21.42koenbtw, there is a wince version of minimo too
23:27.22w3rwolfCosmicPenguin: arrrrrrrgh
23:27.31w3rwolfah koen
23:27.55w3rwolfwatched the last URL in my console and now I've a kind of smurf-vision
23:29.27*** join/#oe darkschneider (~gab@213-140-22-64.fastres.net)
23:33.37w3rwolfhm... just an idea.... : does anyone know if there's a linux-on-a-windows-mobile-cellphone project? :D
23:39.00koen'night all
23:46.44Philippew3rwolf, well I know of some people that tried. Saw some guy that was hacking his phone at the UKUUG summer conference last year.
23:47.01PhilippeDon't know how far it got though.
23:51.19*** join/#oe W8TVI (~noel@69.39.75.252)
23:54.52CosmicPenguinholy crap - rdesktop cvs compiled
23:55.01CosmicPenguinDidn't expect that

Generated by irclog2html.pl by Jeff Waugh - find it at freshmeat.net! Modified by Tim Riker to work with blootbot logs, split per channel, etc.