**** BEGIN LOGGING AT Sun Mar 05 10:59:56 2006
Mar 05 11:00:12 <koen> zecke: I'm building  against full X11 these days
Mar 05 11:00:14 <zecke> koen: well I can donate disks...
Mar 05 11:00:27 <zecke> koen: well meta-gpe failed on diet-x11
Mar 05 11:00:31 <koen> zecke: u320 disks a re a waste of money
Mar 05 11:02:50 <zecke> | cp: cannot stat `/home/zecke/org.openembedded.dev/packages/xlibs/./errordb-keysymdb-path-fix.patch': No such file or directory
Mar 05 11:04:12 <koen> sounds like a good candidate for bugzilla
Mar 05 11:04:13 <jnc> sata seagate 400gb drives are best reliability
Mar 05 11:04:27 <zecke> okay
Mar 05 11:06:27 <koen> zecke: my cirrus board has 120GB of storage :)
Mar 05 11:07:43 <zecke> nice
Mar 05 11:11:51 <zecke> hmm I should try harder to get a triton board+kit
Mar 05 11:17:37 <koen_> bleh, stupid adsl
Mar 05 11:18:05 <koen_> did I miss anything the last 5 minutes?
Mar 05 11:21:55 <leoncamel> hi, koen. here is a bug in file : packages/linux/linux-openzaurus-2.6.15/defconfig-ipaq-pxa270.
Mar 05 11:22:24 <leoncamel> you should add "#" at line 673 " CONFIG_HWMON is not set"
Mar 05 11:22:25 <NAiL> koen: no, no missing anything 
Mar 05 11:22:35 <leoncamel> koen, please check it . 
Mar 05 11:23:32 <koen> leoncamel: you'd have to talk to RP about that
Mar 05 11:41:48 <leoncamel> hi, RP. 
Mar 05 11:42:48 <leoncamel> RP, packages/linux/linux-openzaurus-2.6.15/defconfig-ipaq-pxa270. seems lose a "#" at line 673 " CONFIG_HWMON is not set". 
Mar 05 11:42:55 <leoncamel> RP, please check it .
Mar 05 11:45:47 <RP> leoncamel: ok, will do, thanks
Mar 05 11:50:43 <[lala_]> RP: hi
Mar 05 11:53:40 <RP> hi [lala] 
Mar 05 11:54:11 <[lala]> RP: just installed a complete new image and now i have some sound problems again :-(
Mar 05 11:54:26 <[lala]> RP: any idea how to get the mic on?
Mar 05 11:56:12 <koen> cute
Mar 05 11:56:24 <koen> an akita is only 230 euro at pricejapan
Mar 05 11:58:09 <RP> [lala]: Its just the mic that isn't working?
Mar 05 11:58:30 <[lala]> yep
Mar 05 11:58:52 <[lala]> RP: first of all, the default settings leave the right mixer off 
Mar 05 11:58:53 <RP> [lala]: Did you change zaurusd to default to a mic instead of headphones?
Mar 05 11:59:15 <reenoo> morning
Mar 05 11:59:19 <RP> hi reenoo 
Mar 05 11:59:23 <[lala]> RP: if i plug a headphone in with your default settings i only get sound on left ear
Mar 05 11:59:47 <[lala]> RP: after switching on the right mixer it is stereo with headset profile
Mar 05 11:59:53 <reenoo> hey RP
Mar 05 11:59:57 <RP> [lala]: ok, I'll need to investigate that...
Mar 05 12:00:02 <[lala]> RP: if i unplug the headset then i get inverted sound on speaker
Mar 05 12:00:21 <[lala]> RP: switching on right out2 playback invert fixes that
Mar 05 12:00:43 <[lala]> RP: now i change the settings to get the headset profile
Mar 05 12:00:52 <[lala]> but mic is not working
Mar 05 12:01:03 <RP> [lala]: ok, we need to make zaurusd aware it needs to set the invert bit for speaker use
Mar 05 12:02:19 <RP> Actually, I'll just change the default setting for right out 2 invert which will take care of that problem
Mar 05 12:03:16 <[lala]> ~lart alsamixer
Mar 05 12:03:17 * ibot throws a AN/M-8 smoke grenade at alsamixer
Mar 05 12:03:57 <RP> [lala]: Just to confirm, this was a totally clean image with not /etc/asound.state file when you started?
Mar 05 12:04:04 <[lala]> yes
Mar 05 12:04:23 <[lala]> RP: the funny thing is that i couldn't get it in that state by simply rebooting
Mar 05 12:04:33 <[lala]> RP: once the settings were ok all was fine
Mar 05 12:05:15 <[lala]> RP: of course with that confusing alsamixer i do not remember how i got it working last time :-(
Mar 05 12:05:27 <CoreDump|home> heh
Mar 05 12:05:27 <RP> [lala]: It only configures alsa once, the first time zaurusd runs
Mar 05 12:05:51 <[lala]> RP: it seems to me that not all state of the sound chip is held in asound.state
Mar 05 12:06:13 <[lala]> RP: yes, but then it takes asound.state after each reboot, doesn't it?
Mar 05 12:06:14 <RP> to simulate a from scratch setup, /etc/init.d/zaurusd stop; rm /etc/asound.state; /etc/init.d/zaurusd start
Mar 05 12:06:22 <koen> zecke: if you want to have a laugh: http://www.oesf.org/forums/index.php?showtopic=18032&hl=
Mar 05 12:06:25 <RP> [lala]: It should do, yes
Mar 05 12:06:52 <[lala]> RP: i do a quick simulation ... hold on :-)
Mar 05 12:07:49 <koen> RP: did you say that you had a hal initscript, or did I imagine that?
Mar 05 12:08:29 <koen> ~lart gnome-python
Mar 05 12:08:30 * ibot installs PocketPC on gnome-python's PDA
Mar 05 12:08:48 <koen> ibot: botsnack
Mar 05 12:08:48 <ibot> koen: thanks
Mar 05 12:10:39 <RP> koen: You imagined it - I just worked out what it might need to contain :)
Mar 05 12:10:54 <[lala]> RP: simulation failed
Mar 05 12:10:56 <koen> damn my imagination
Mar 05 12:11:12 <[lala]> RP: i just did as you said and now the speaker is not muted
Mar 05 12:11:29 <[lala]> as it was after fresh install
Mar 05 12:12:01 <[lala]> RP: as i said - there must be *some* other state for the sound chip as asound.state
Mar 05 12:12:05 <zecke> koen: yes it was nice
Mar 05 12:12:22 <zecke> koen: I wonder the people who trailed Qtopia2...
Mar 05 12:12:44 <koen> zecke: I actually ran qtopia2 for a while
Mar 05 12:12:45 <zecke> koen: I wonder about the +200 patches needed to get the basics things done...
Mar 05 12:12:46 <zecke> koen: I wonder where they did publish them...
Mar 05 12:12:54 <zecke> koen: on SDG System?
Mar 05 12:13:00 <RP> [lala]: The currently loaded setup doesn't neccessarily match that in the state file...
Mar 05 12:13:03 <koen> zecke: yes, for SDG
Mar 05 12:13:31 <[lala]> RP: how can that be? i thought the state is loaded from asound.state?
Mar 05 12:13:43 <zecke> koen: did you use it for real things?
Mar 05 12:14:01 <koen> zecke: for getting annoyed with all the qvga artwork
Mar 05 12:14:39 <RP> [lala]: It is. If the scripts ran the store/restore commands in the wrong order, things could get confused though
Mar 05 12:15:46 <[lala]> RP: well, so i do it again and load the settings manually before starting zaurusd
Mar 05 12:16:49 <[lala]> RP: same thing
Mar 05 12:17:14 <[lala]> RP: let me do a quick test with not running zaurusd at all and only loading the default state via alsactl
Mar 05 12:18:23 <[lala]> RP: now the speaker is on
Mar 05 12:18:33 <[lala]> RP: its kinda weird
Mar 05 12:18:48 <RP> [lala]: I've just realised that to simulate a from scratch startup, you also need to remove the symlinks in /etc/zaurusd to the machine configs
Mar 05 12:18:57 <RP> I'm changing that...
Mar 05 12:19:29 <[lala]> RP: ok, it try that one
Mar 05 12:19:41 <RP> koen: I added c7x0 support to zaurusd btw
Mar 05 12:20:01 <koen> RP: ah, cool
Mar 05 12:20:20 <koen> RP: root on sd is waaaaaaaaaay faster as jffs2 on my c700
Mar 05 12:20:32 <koen> RP: but that might be because I'm not using tmpfs anymore 
Mar 05 12:21:57 <[lala]> RP: well, right mixer is on after that simulation again
Mar 05 12:22:03 <[lala]> RP: i really don't get it
Mar 05 12:22:39 <[lala]> RP: it's impossible to get in the state after fresh install
Mar 05 12:22:59 <RP> [lala]: in which state?
Mar 05 12:23:11 <[lala]> RP: right mixer was off
Mar 05 12:23:45 <[lala]> RP: had to turn it on manually
Mar 05 12:23:54 <[lala]> RP: i do a fresh install again
Mar 05 12:23:55 <RP> [lala]: When the machine powers up, that mixer will be powered off
Mar 05 12:24:16 <RP> [lala]: The scritps should then run alsactl restore and it should be enabled
Mar 05 12:25:10 <[lala]> RP: i do a fresh install now, see if it is at least reproduceable
Mar 05 12:26:37 <[lala]> RP: if the mixer is off in the asound.state - why wouldnt it help to restore the setting with alsactl?
Mar 05 12:27:16 <RP> [lala]: I think I have an idea about why it might do this - alsactl restore should be run before switchevd in the zaurusd init script
Mar 05 12:27:43 <RP> At the moment there is a subtle race :-/
Mar 05 12:28:48 <[lala]> i can try that one if the my borzoi is up again
Mar 05 12:29:54 <[lala]> after that test i have to hurry home or else my wife would go shouting ;-)
Mar 05 12:30:36 <RP> [lala]: I've updated zaurusd svn with somethe changes
Mar 05 12:30:43 <koen> zecke: automerger script courtesy of njs: http://rafb.net/paste/results/UU7IYr89.nln.html
Mar 05 12:31:13 <[lala]> RP: ok, i update my tree then
Mar 05 12:31:32 <RP> koen: jffs2 is really slow on the zaurii btw. spitz and the microdrive is a totally different world!
Mar 05 12:33:10 <koen> RP: I still can't justify buying a Z ;)
Mar 05 12:34:39 <RP> koen: Its only a question of time before you're converted ;-)
Mar 05 12:35:22 <koen> RP: now I have a semi-stable income, yes :)
Mar 05 12:35:38 <koen> but first holiday + diving course
Mar 05 12:36:58 <RP> Given my tendancy towards dangerous things, I sometimes wonder why I haven't tried diving. I guess my other hobbies make up for it though :)
Mar 05 12:37:30 <koen> motocross + driving mg without brakes?
Mar 05 12:38:41 <RP> I had the MG airborne yesterday on a snow covered/icy unclassified road :-/
Mar 05 12:38:49 <[lala]> RP: reinstalled - now speaker (right mixer) is really muted
Mar 05 12:39:02 <[lala]> RP: now i update your zaurusd
Mar 05 12:39:48 <[lala]> RP: have you changed the default of this setting in the newest version?
Mar 05 12:40:15 <RP> [lala]: I changed the inverted setting
Mar 05 12:40:34 <[lala]> but right mixer is off by default?
Mar 05 12:40:59 <RP> [lala]: Not from what I read in the default state file, unless we're talking about different controls
Mar 05 12:41:20 <RP> [lala]: When booting have you the headset connected or disconnected?
Mar 05 12:41:38 <[lala]> RP: i had it disconnected
Mar 05 12:41:41 <[lala]> speaker was muted
Mar 05 12:41:52 <[lala]> connecting the headset only plays left channel
Mar 05 12:42:04 <koen> RP: is the switch debounced in hw?
Mar 05 12:43:00 <RP> koen: on corgi? I noticed the long delays too.  I suspect this means yes although who knows :-/
Mar 05 12:43:11 <RP> [lala]: A headset is always going to me mono
Mar 05 12:43:26 <RP> [lala]: headphones should be stereo though
Mar 05 12:43:37 <[lala]> RP: i meant headset then
Mar 05 12:43:47 <[lala]> doh
Mar 05 12:43:51 <[lala]> headphones of course
Mar 05 12:44:28 <RP> right :). I'll run some tests on spitz and see if I can reproduce the problem...
Mar 05 12:44:44 <koen> later all
Mar 05 12:45:49 <[lala]> RP: the switch in alsamixer is named "Right Mixer" and defaults to off
Mar 05 12:45:59 <[lala]> i have to switch it on to get my left ear going
Mar 05 12:47:22 <[lala]> RP: and the inverted switch is still off - although i just updated zaurusd via ipkg
Mar 05 12:47:47 <[lala]> devmand complains about [: missing] each time i plug or unplug the jack
Mar 05 12:48:49 <[lala]> RP: that would be the first problem ... the second issue is the non-working mic if i change all to headset
Mar 05 12:49:27 <CIA-1> 03nail 07org.oe.dev * rec0208dc... 10/conf/distro/slugos-packages.conf: slugos-packages: Update x11-packages paths
Mar 05 12:49:34 <CIA-1> 03nail 07org.oe.dev * rd6c08dde... 10/conf/distro/slugos-packages.conf: slugos-packages: Fix missing \
Mar 05 12:49:38 <CIA-1> 03nail 07org.oe.dev * r05bec92e... 10/conf/distro/slugos-packages.conf: slugos-packages: Remove xtrans dir
Mar 05 12:49:56 <RP> [lala]: You are using a kernel with the latest ASoC, aren't you?
Mar 05 12:51:13 <RP> [lala]: I can see the problem - I think something has changed within ASoC as the control files have changes format slightly :-/
Mar 05 12:52:11 <[lala]> i have that one which fixed the suspend bug
Mar 05 12:52:21 <[lala]> is there a newer version?
Mar 05 12:53:02 <RP> [lala]: no, that should be the latest  - I'm also seeing the problem
Mar 05 12:53:39 <[lala]> 0.10rc3
Mar 05 12:54:40 <[lala]> ok, im gone
Mar 05 12:54:43 <[lala]> see you later
Mar 05 12:54:57 <RP> [lala]: I'll make sure this gets fixed...
Mar 05 13:11:15 <RP> [lala_]: Its really fixed this time (tested and confirmed working)
Mar 05 13:24:59 <polyonymous> Maybe today I will have better luck trying to satisfy modules dependencies...
Mar 05 13:54:44 <[lala_]> RP: even the mic? :-)
Mar 05 13:55:21 <koen|770> hi all
Mar 05 13:55:40 <NAiL> hi koen-with-a-770-that-I-want
Mar 05 13:55:48 <NAiL> ;)
Mar 05 13:56:06 <[lala_]> i dont want a 770 until i can build my own image ;-)
Mar 05 13:56:08 <koen|770> :)
Mar 05 13:56:30 <NAiL> It's still cool ;)
Mar 05 13:56:39 <[lala_]> agreed :-)
Mar 05 13:59:12 <RP> [lala]: even the mic. It was just that the default alsa state file wasn't getting used correctly
Mar 05 13:59:24 <[lala]> RP: what a bummer ;-)
Mar 05 13:59:44 <RP> [lala]: You can build images for the 770 from OE now...
Mar 05 14:00:03 <[lala]> RP: koen told me that it's not entirely possible atm
Mar 05 14:00:14 <[lala]> RP: at least that's what he told in bruessel
Mar 05 14:00:26 <RP> [lala]: You can't build maemo images but you can build a working xserver
Mar 05 14:00:34 <[lala]> RP: there a some essential bits missing
Mar 05 14:00:53 <RP> It depends how similar to the normal 770 image you want to be
Mar 05 14:01:25 <[lala]> RP: so is it possible to run gpe ore opie on the 770 device?
Mar 05 14:01:56 <RP> law_: I can't see why they wouldn't work (apart from maybe not fully supporting the odd screen size)
Mar 05 14:02:35 <uv1> [lala]: just got a opie-image build for my 870 ;-)
Mar 05 14:02:47 <reenoo> RP: the bits required to control device specifics (power management in particular) are closed source to my knowledge
Mar 05 14:03:00 <RP> koen|train: Do I remember correctly that you said the kernel source was available for the 770? I couldn't find the url...
Mar 05 14:03:36 <koen|train> it should be in the dep repo
Mar 05 14:04:09 <RP> reenoo: Ah, I guess I'd not notice that given mine is lacking working battery so spends its time on mains power :-/
Mar 05 14:04:11 <uv1> [lala]: Though I am not sure, if I will try the image today, as I need my 870 working next week!
Mar 05 14:04:15 <koen|train> the wiki has a 'compile your own' page
Mar 05 14:04:25 <[lala]> uv1: ;-)
Mar 05 14:05:08 <uv1> [lala]: Just had problem with konqe again (but was able to use the old workarround) ...
Mar 05 14:05:26 <[lala]> RP: have you also fixed that one: devmand complains about [: missing] each time i plug or unplug the jack
Mar 05 14:05:37 <reenoo> koen|train: how about wifi drivers?
Mar 05 14:05:51 <uv1> [lala]: Also sharp-aticore-oss did not build, But I send a patch to bugzilla this morning, which helps out.
Mar 05 14:05:59 <koen|train> reenoo: enclosed in the initfs
Mar 05 14:06:11 <[lala]> uv1: interesting
Mar 05 14:06:21 <koen|train> but it's gpl anyway
Mar 05 14:07:16 <RP> [lala]: Yes, I fixed that
Mar 05 14:08:30 * reenoo yawns
Mar 05 14:08:43 <reenoo> there are looooooots of locales...
Mar 05 14:09:08 <reenoo> koen|train: the driver is completely gpl? interesting.
Mar 05 14:09:37 <RP> koen|train: I found it, thanks. I wasn't looking in maemo1.1 for some reason...
Mar 05 14:09:39 <reenoo> s/gpl/open source/
Mar 05 14:10:45 <RP> So how do we get OE to extract the source from a .deb in a SRC_URI? :-/
Mar 05 14:14:32 <uv1> [lala]: Now my box builds e-wm. Up to now just libx11 missing, which has to be build as a single package therefore ...
Mar 05 14:15:27 <koen|train> RP: florian is working on a deb/dsc parser
Mar 05 14:15:39 <uv1> RP: Why do you need it from deb ? Did that with the .orig.tgz from debian mirror for schedutils before and that worked quite well.
Mar 05 14:17:57 <uv1> Apropos : How to provide the package to OE repository ?
Mar 05 14:18:51 <polyonymous> Why is it that eveyone can build *-image but me? ;-)
Mar 05 14:19:28 <uv1> polyonymous: which problems do you have ?
Mar 05 14:20:03 <polyonymous> Sorry ;-)
Mar 05 14:20:14 <polyonymous> I can't satisfy a lot of modules dependencies.
Mar 05 14:20:22 <koen|train> uv1: are all your rpm patches in?
Mar 05 14:20:42 <polyonymous> Actually, it does build task-bootstrap, but failes when it comes to *-image
Mar 05 14:21:17 <uv1> polyonymous: Which version of bitbake do you use ? Which dependencies and what is * in *-image ?
Mar 05 14:22:10 <uv1> koen|train: no new patches since the ones month ago. Just looked into package_rpm.bbclass and did not yet find them !
Mar 05 14:22:25 <polyonymous> uvl: the latest svn, "hostap-modules-cs orinoco-modules-cs kernel-module-net...", I don't think the whole list makes sense, I believe there is one problem. * is bootstrap and opie. Probably gpe as well, but at this time I fell back to bootstraip
Mar 05 14:23:23 <CoreDump|home> polyonymous: what's your bitbake version?
Mar 05 14:23:32 <polyonymous> Maybe the problem is that I'm trying to build 2.4 image for borzoi. But, in fact, I don't need kernel at all at this point, as I'm just building an image I can chroot into and run environment.
Mar 05 14:23:45 <polyonymous> CoreDump|home, head revision as of yesterday.
Mar 05 14:23:53 <koen|train> heh
Mar 05 14:24:10 <koen|train> don't rdepend on modules
Mar 05 14:24:21 <koen|train> rrecommend them
Mar 05 14:24:31 <polyonymous> koen|train, I don't, it's something in the tree that depends on them ;-)
Mar 05 14:24:36 <CoreDump|home> +mhentges@oe-akita-head:~ >bitbake --version
Mar 05 14:24:36 <CoreDump|home> BitBake Build Tool Core version 1.3.3.0, bitbake version 1.3.3
Mar 05 14:25:12 <polyonymous> BitBake Build Tool Core version 1.3.3.3, bitbake version 1.3.3.2
Mar 05 14:25:21 <CoreDump|home> heh that should work
Mar 05 14:25:35 <uv1> CoreDump|home: I agree, that was my problem last week.
Mar 05 14:25:51 <polyonymous> Well, yesterday I've had bitbake-related problems - it didn't build any packages, but not it's something different.
Mar 05 14:25:59 <uv1> polyonymous: Did you clean your tmp dir after updating to 1.3.3.3 ?
Mar 05 14:25:59 <zecke> time for a new bitbake release then
Mar 05 14:26:07 <zecke> (now that meat-opie-all works again)
Mar 05 14:26:15 <uv1> zecke: ;-((
Mar 05 14:26:29 <polyonymous> uvl: I believe I have, but well, I think it worth giving a try once more.
Mar 05 14:27:04 <polyonymous> uvl: do I have to wipe out the whole tmp or just part of it?
Mar 05 14:28:00 <CoreDump|home> all of it
Mar 05 14:28:09 <polyonymous> So it will take ages again... ;-)
Mar 05 14:28:25 * CoreDump|home nods
Mar 05 14:28:34 <polyonymous> Okay, wiped, baking.
Mar 05 14:28:52 <polyonymous> BTW, CoreDump|home, I'm Gin-geR who finally decided to register, if you still remember me ;-)
Mar 05 14:29:06 <CoreDump|home> heh I do
Mar 05 14:29:34 <polyonymous> I've set up a test5 image in chroot environment and played a bit, but then decided to go for newlybuild image.
Mar 05 14:30:02 <polyonymous> newlybuilt too.
Mar 05 14:31:36 <uv1> Just working on my openembedded doc. Any new doc hints for bitbake 1.3.3 ? Where to find them ?
Mar 05 14:32:03 <polyonymous> uvl: suggest that people remove what's left of previous version if using bitbake from svn.
Mar 05 14:32:30 <polyonymous> uvl: it was my problem yesterday - svn bitbake seemed to not work at all, until I figured I have some pieces of 1.3.2 left.
Mar 05 14:32:39 <zecke> uv1: http://developer.berlios.de/project/showfiles.php?group_id=2701
Mar 05 14:34:08 <zecke> I plan to release a bitbake 1.3.3.1 today, as it can now build meta-opie-all again
Mar 05 14:34:26 <polyonymous> BTW, maybe this should be fixed in bitbake itself?
Mar 05 14:34:33 <zecke> what?
Mar 05 14:35:11 * zecke just semi watches this window (again)
Mar 05 14:36:09 <polyonymous> zecke, I've had parts of 1.3.2 in /usr/lib, I think and bitbake failed to do anything when run from ~/svn/bitbake/bin, complaining about wrong number of parameters in some function.
Mar 05 14:36:10 <uv1> polyonymous: Would you send a more detailed description of your problemn and how you solved. Yould be interesting for the FAQ chapter.
Mar 05 14:36:31 <polyonymous> Probably, loading some modules from /usr/lib (I'm not good with python).
Mar 05 14:36:43 <zecke> polyonymous: well this can't be solved in bitbake easily
Mar 05 14:37:04 <polyonymous> zecke, So, in python you can't prepend to whatever path it uses to look for modules?
Mar 05 14:37:04 <zecke> polyonymous: this would mean we would have to hack python's import function to check each version of each module
Mar 05 14:37:14 <zecke> PYTHONPATH
Mar 05 14:37:25 <uv1> polyonymous: I did an update and it worked here. Did not remove anything from 1.3.2.
Mar 05 14:37:41 <polyonymous> Can't you just put local directory with higher priority? Like in front of the list.
Mar 05 14:37:56 <uv1> polyonymous: Just overwrite your old install dir ;-)
Mar 05 14:38:02 <polyonymous> uvl: no, I've had 1.3.2 from gentoo portage and and the other one in svn
Mar 05 14:38:19 <zecke> sys.path.append(os.path.join(os.path.dirname(os.path.dirname(sys.argv[0])), 'lib'))
Mar 05 14:38:23 <uv1> polyonymous: OK, that may cause problems then ...
Mar 05 14:38:29 <zecke> hmm we could try prepend
Mar 05 14:38:31 <polyonymous> uvl: The 'GettingStarted' page recommends runing from svn, but doesn't suggest removal of the other.
Mar 05 14:39:03 <polyonymous> zecke, so there's no way to modify sys.path in any other way?
Mar 05 14:39:22 <zecke> polyonymous: well set $PYTHONPATH before calling bitbake
Mar 05 14:39:34 <zecke> polyonymous: and I look of sys.path is just a list where we could prepend
Mar 05 14:39:39 <polyonymous> zecke, no, it's not a problem for me, I'm talking about possible general solution.
Mar 05 14:40:07 <zecke> polyonymous: either always install bitbake, or always run it from the folder :)
Mar 05 14:40:20 <zecke> everything else confuses us humans
Mar 05 14:40:39 <polyonymous> zecke, you do 'append' which probably puts new path at the end of the list. Isn't it possible to add os.path.join(os.path.dirname(os.path.dirname(sys.argv[0])), 'lib') to the beginning of the list?
Mar 05 14:41:02 <zecke> polyonymous: I think it is using .insert instead of append
Mar 05 14:41:14 <zecke> polyonymous: but I do not know the consequences/implications of that
Mar 05 14:41:17 <polyonymous> zecke,  that is about what I suggest to you ;-)
Mar 05 14:41:18 <zecke> polyonymous: I will consider it
Mar 05 14:41:44 <polyonymous> I do not foresee any, as long as you do not pollute your 'lib' with harmfull stuff, but then again - it's up to you, this is not a problem for me anymore.
Mar 05 14:42:26 <polyonymous> And for uvl the suggestion is to write about this possible problem.
Mar 05 14:43:02 <polyonymous> And my build failed again at the earlier stage now ;-))
Mar 05 14:43:26 <polyonymous> C compiler can not create executables... :)
Mar 05 14:44:03 <zecke> that sounds like OSX, FreeBSD host system to me
Mar 05 14:44:21 <polyonymous> zecke, no, Gentoo, I think I'll figure this one out once I find config.log;)
Mar 05 14:45:18 <uv1> polyonymous: Had the same problem. Your target is arm ?
Mar 05 14:45:24 <polyonymous> yup
Mar 05 14:45:31 <polyonymous> arm-linux-gcc: No such file or directory
Mar 05 14:45:41 <uv1> Do you have a ASSUME_PROVIDED entry in your local.conf ?
Mar 05 14:45:45 <polyonymous> yes
Mar 05 14:45:48 <polyonymous> And renames gcc
Mar 05 14:45:56 <polyonymous> that's clearly what the problem is.
Mar 05 14:46:02 <polyonymous> renamed
Mar 05 14:46:15 <uv1> Change to "ASSUME_PROVIDED +=" instead of "ASSUME_PROVIDED ="!. Thatll do.
Mar 05 14:46:38 <polyonymous> lemme see..
Mar 05 14:46:52 <polyonymous> Well, there are +=
Mar 05 14:47:01 <Eddy> Hey how do you include python in gpe-image, I tried adding 'python' to rdepends and ipkg_install but it gives me 'ERROR: Nothing provides runtime dependency python' even though it compiles python before that.
Mar 05 14:47:33 * koen|train wonders why people would want to use 2.4 on clamshells
Mar 05 14:47:35 <uv1> Hmm. bitbake -e | grep ASSUME_PROVIDED ?
Mar 05 14:47:36 <Eddy> I mean depends not rdepends
Mar 05 14:47:41 <polyonymous> Well, it looks for arm-linux-gcc and I renamed those to arm-linux-gcc-2.95 or something. That is why the problem...
Mar 05 14:48:03 <polyonymous> koen|train, At this point I want to play with this image while running cacko rom - that is the only reason.
Mar 05 14:48:09 <uv1> polyonymous: Did you install the sharp toolchain ?
Mar 05 14:48:21 <zecke> Eddy: look at the deploy directory, how is the python package named?
Mar 05 14:48:33 <polyonymous> uv1, umm... I installed the one, mentioned on wiki/
Mar 05 14:49:03 <uv1> polyonymous: As long as I know until now, thats necessary, if you want to play with the old stuff.
Mar 05 14:49:18 <koen|train> so why build 2.4 if it isn't used?
Mar 05 14:49:48 <polyonymous> koen|train, that is what I started with - I do not need kernel at all. But how do I tell *-image targets to forget about kernel?
Mar 05 14:49:55 <Eddy> python-core, python-image, libpython2.4, python-devel, etc.
Mar 05 14:49:56 <uv1> BTW ASSUME_PROVIDED and your bitbake problem will be documented in www.uv-ac.de/openembedded from later afternoon on ;-). Usage of sharp-toolchain is already.
Mar 05 14:50:06 <Eddy> am I going to have to list all of the ones I want?
Mar 05 14:51:05 <polyonymous> uvl, so far my problem is not the wrong toolchain, but inability to find an ;-
Mar 05 14:51:13 <polyonymous> any ;-)
Mar 05 14:51:41 <polyonymous> uvl, why is it suggested to mv arm-linux-gcc arm-linux-gcc-2.95? Doesn't it prevent gcc from being found?
Mar 05 14:51:49 <polyonymous> (on ZaurusKernels page)
Mar 05 14:52:43 <koen|train> the 2.95 toolchain is broken and shouldn't be used
Mar 05 14:53:08 <uv1> polyonymous: Because a later build arm-linux-gcc (e.g.3.4.4) would not be found.
Mar 05 14:53:25 <uv1> koen|train: Why. It did work a few weeks ago, If I remember right.
Mar 05 14:54:28 <koen|train> uv1: it only works with broken kernels
Mar 05 14:54:29 <polyonymous> uv1, how do I tell the build to find arm-linux-gcc-2.95 earler?
Mar 05 14:55:04 <polyonymous> uv1, Anyway, the recent version of gcc breaks on building kernel or just produces screwed up one?
Mar 05 14:55:29 <koen|train> polyonymous: don't build a 2.4 kernel and you won't need 2.95
Mar 05 14:55:32 <polyonymous> koen|train, And speaking of not building kernel at all: is there a way to prevent, say, opie-image target from putting in kernel.
Mar 05 14:55:45 <polyonymous> I'd love to ;-)
Mar 05 14:56:11 <polyonymous> Then the question is - are other parts of the image built differently depending on the kernel used?
Mar 05 14:56:41 <uv1> polyonymous: what is the result of which arm-linux-gcc ?
Mar 05 14:57:15 <polyonymous> uv1, huh?
Mar 05 14:57:29 <uv1> polyonymous: "which arm-linux-gcc"
Mar 05 14:57:34 <polyonymous> ahh ;-)
Mar 05 14:57:43 <polyonymous> I guess nothing?
Mar 05 14:57:59 <polyonymous> otherwise it would find it.
Mar 05 14:58:04 * koen|train votes for deprecating cxxx 2.4 support
Mar 05 14:58:17 <polyonymous> but shouldn't it be built on the way?
Mar 05 14:58:32 <uv1> polyonymous: Your path is not set. Do you use an environmentscript like mentioned on my doc page ?
Mar 05 14:58:51 <polyonymous> koen|train, So, can you answer any of my questions about kernel and image? I'd love not to mess with 2.4
Mar 05 14:59:08 <uv1> IF not you should consider. Set PATH=/usr/local/arm/2.95.3/bin:$PATH in there.
Mar 05 14:59:22 <polyonymous> uvl: my path is set to point to 2.95 toolchain whcih has arm-linux-gcc renamed
Mar 05 14:59:35 <polyonymous> And this is exactly what my path looks like.
Mar 05 15:00:16 <koen|train> polyonymous: in short: nobody uses the 2.4 stuff in oe anymore, so it's likely to be broken
Mar 05 15:00:53 <polyonymous> koen|train, okay, and how about kernel-dependent functionality of the rest of the image?
Mar 05 15:01:04 <uv1> polyonymous: bitbake -e | grep PATH | grep "2.95" in target environment. What does it say ?
Mar 05 15:02:27 <koen|train> polyonymous: since you're chrooting from 2.4-crappix i shouldn't worry about that
Mar 05 15:03:34 <polyonymous> uv1, there is a staging_bindir in path...
Mar 05 15:04:09 <polyonymous> koen|train, You should not, of course. Actually, you don't HAVE to help me even if I'm using OZ ;-)
Mar 05 15:04:57 * koen|train isn't using oz 
Mar 05 15:05:21 <polyonymous> koen|train, what do you use?
Mar 05 15:05:21 <uv1> polyonymous: is there your /usr/local/arm/2.95.3/bin dir in path ?
Mar 05 15:05:33 <uv1> s/path/PATH/
Mar 05 15:05:48 <polyonymous> uv1, yes, but it has mv-ed gcc, remember? :)
Mar 05 15:05:58 <koen|train> polyonymous: ångström
Mar 05 15:06:26 <polyonymous> On what machine? (anyway, you don't HAVE to help me even if I use it:))
Mar 05 15:06:34 <polyonymous> or whatever. You just don't have to.
Mar 05 15:06:51 <koen|train> c700, hx4700 and h2200
Mar 05 15:07:06 <polyonymous> Aha, it's not for borzoi, is it?
Mar 05 15:07:10 <koen|train> and ep93xx :)
Mar 05 15:07:57 <uv1> polyonymous: so move it back to arm-linux-gcc and give it a try (and send me an email, if my doc page is wrong then (uv1@uv-ac.de)).
Mar 05 15:08:06 <koen|train> i don't have a borzoi
Mar 05 15:08:34 <gremlin[it]> ~flame who write cirrus docs 
Mar 05 15:08:36 <ibot> who write cirrus docs sux0rz d00d!!!!!!!!!!!!!!@@!!!!!!!11
Mar 05 15:08:37 <polyonymous> The thing is that at this point I don't think OE stuff is usable on PDA, but I want to do my best to do something about it, because I'm not happy about cacko either.
Mar 05 15:08:51 <polyonymous> uv1, lemme try..
Mar 05 15:10:30 <polyonymous> But, uv1, maybe what I should do is to bitbake arm-gcc manually?
Mar 05 15:10:33 <koen|train> polyonymous: don't confuse buildsystem with desktop envs
Mar 05 15:10:49 <Eddy> oh boy, I got into this one last night
Mar 05 15:10:51 <Eddy> hehe
Mar 05 15:10:53 <polyonymous> koen|train, well, It's not the buildsystem, it's packages tree.
Mar 05 15:11:12 <Eddy> You might want to formally define the differences on your website or something
Mar 05 15:11:37 <koen|train> Eddy: it's on the frontpage
Mar 05 15:12:09 <uv1> polyonymous: You cannot do that for 2.95.3. No chance to ge a sharp compatible gcc. If you want to for e.g. 3.4.4 no probs. But then youll have problems with the cacko rom.
Mar 05 15:12:28 <Eddy> well, I'll just shut up then
Mar 05 15:12:34 <Eddy> :-)
Mar 05 15:13:14 <polyonymous> uv1, but I think that having ASSUME_PROVIDED gcc-3.4.4 is not built and that is the problem, no?
Mar 05 15:13:23 <koen|train> polyonymous: oe still suffers from the opiezaurus association
Mar 05 15:14:02 <polyonymous> koen|train, well, it has to do with oe. It has more to do with oe than with oz, but even more with opie, of course.
Mar 05 15:14:38 <koen|train> opie stinks
Mar 05 15:14:45 <koen|train> no doubt about it
Mar 05 15:15:02 <koen|train> it's tech from 2001
Mar 05 15:15:06 <uv1> NOt if you have a "ASSUME_PROVIDED += "virtual/arm-linux-gcc-2.95" like me. gcc-3.4.4 is build anyway.
Mar 05 15:15:26 <polyonymous> koen|train, I agree. What env do you use?
Mar 05 15:15:41 <koen|train> zaurus users always go like 'opie=oz=oe, so oe sucks'
Mar 05 15:16:07 <polyonymous> koen|train, I'm not saying = and do not blame any link in this chain.
Mar 05 15:16:07 <zecke> Eddy: right
Mar 05 15:16:21 <koen|train> polyonymous: a custom gpe setup
Mar 05 15:16:30 <Eddy> but I only have to list them in the ipkg_install right?
Mar 05 15:16:41 <Eddy> no wait
Mar 05 15:16:49 <zecke> koen|train: how do you date C?
Mar 05 15:16:52 <Eddy> do I list them in depeds and ipkg_install
Mar 05 15:16:54 <Eddy> ?
Mar 05 15:16:55 <Eddy> depends
Mar 05 15:16:59 <polyonymous> koen|train, so gpe is better? At first I didn't like it either (I've tried both a few times), but sure I will dig more.
Mar 05 15:17:10 <zecke> Eddy: no, not depends
Mar 05 15:17:10 <koen|train> zecke: c70 or c99?
Mar 05 15:17:27 <koen|train> ply
Mar 05 15:17:28 <zecke> koen|train: measuring GPE is not using C99, I'm fro C70
Mar 05 15:17:39 <Eddy> thanx
Mar 05 15:17:40 <zecke> koen|train: wow opie is 30 years more current than GPE
Mar 05 15:17:50 <zecke> Eddy: DEPENDS are only for build time dependencies
Mar 05 15:17:54 <polyonymous> And please, do not misinterpret me, I'm not saying OE sucks and know well what is the difference.
Mar 05 15:18:05 <polyonymous> uv1, I have ASSUME_PROVIDED += "virtual/arm-linux-gcc-2.95"
Mar 05 15:18:38 <polyonymous> Well, okay, I'm going for another build.
Mar 05 15:19:21 <polyonymous> gpe-image this time ;-))
Mar 05 15:19:27 <uv1> polyonymous: With gcc-3.4.4 you have to set PREFERRED_gcc-cross = "3.4.4". Otherwise you get a 4.x
Mar 05 15:19:49 <CIA-1> 03koen 07org.oe.dev * r2d917a97... 10/conf/distro/angstrom-2006.9.conf: angstrom 2006.9: prefer dbus 0.61 to work around some building weirdnesses
Mar 05 15:19:56 <uv1> polyonymous: And comment your ASSUME_PROVIDED entry ...
Mar 05 15:20:08 <uv1> uv1: Takes a nap now ;-))
Mar 05 15:20:15 <zecke> uv1: cya :)
Mar 05 15:20:17 <polyonymous> sweet dreams ;-)
Mar 05 15:21:29 <polyonymous> meanwhile I'll chroot into test5 gpe image ;-)
Mar 05 15:23:07 <leoncamel> hi,folks. would you please have a look at this : http://rafb.net/paste/results/QaVc1T15.html ? 
Mar 05 15:23:25 <leoncamel> it seems there are something wrong while bitbake xset .. :( 
Mar 05 15:26:13 <[lala]> RP: ping
Mar 05 15:27:01 <zecke> leoncamel: were all dependencies build?
Mar 05 15:27:15 <zecke> leoncamel: did you look for the missing header file in the STAGING_INCDIR?
Mar 05 15:28:10 <reenoo> zecke: that version (date actually) of xset shouldn't try to use XPrint in the first place
Mar 05 15:28:32 <leoncamel> zecke,  could you explain more detail ? 
Mar 05 15:29:23 <zecke> okay
Mar 05 15:29:33 <leoncamel> zecke, I can't "find tmp/work -name Print.h".. 
Mar 05 15:29:45 <zecke> leoncamel: this backs up what reenoo said
Mar 05 15:30:46 <leoncamel> zecke, but why we use the xset of 20040817 ? is it too old ? 
Mar 05 15:31:30 <zecke> leoncamel: what is your CVSDATE/SRCDATE pointing to?
Mar 05 15:32:11 <leoncamel> zecke, I don't know .. 
Mar 05 15:32:21 <leoncamel> zecke, how can I get it ? 
Mar 05 15:32:22 <zecke> leoncamel: well then find it out
Mar 05 15:36:28 <leoncamel> zecke, SRCDATE="20060305"
Mar 05 15:36:42 <leoncamel> zecke, but I can't find CVSDATE . 
Mar 05 15:37:34 <zecke> ah well xset is hardcoded to 20040817
Mar 05 15:37:44 <zecke> not being a X maintainer I do not know why this is the case
Mar 05 15:37:59 <zecke> leoncamel: look at the config.log and findout why xset believes Xprint is available?!
Mar 05 15:38:17 <reenoo> xset: because I got tired of rediffing the autotools patches
Mar 05 15:39:26 <zecke> understandable
Mar 05 15:39:40 <pb_> zecke: xset hasn't changed in any useful way since forever, so there isn't much benefit in having a newer one than that
Mar 05 15:39:59 <pb_> as you are only too well aware, x is old technology
Mar 05 15:40:32 <zecke> leoncamel: so to summarize: using xset from 2004 is okay, it shouldn't use Xprint though
Mar 05 15:40:44 <zecke> leoncamel: also it is using autotools so a config.log should exist
Mar 05 15:40:50 <zecke> leoncamel: there might be traces for Xprint
Mar 05 15:40:57 <reenoo> pb_: people will hate us for this patch
Mar 05 15:41:06 <reenoo> pb_: ~10 secs per locale...
Mar 05 15:41:42 <zecke> reenoo: on which assumptions is this patch build?
Mar 05 15:41:51 <leoncamel> zecke, here is the log . http://rafb.net/paste/results/vNrywl18.html
Mar 05 15:42:04 <zecke> leoncamel: well I don't know X, look yourself :)
Mar 05 15:42:08 <leoncamel> zecke, I can't find any print detection trace .. 
Mar 05 15:42:11 <reenoo> zecke: assumptions?
Mar 05 15:42:46 <zecke> leoncamel: well then look at the compiled file and see why it wants to include the header file
Mar 05 15:43:20 <zecke> reenoo: how are these locales generated now? natively? using qemu userspace emulation?
Mar 05 15:43:41 <reenoo> zecke: qemu-arm
Mar 05 15:44:06 <zecke> reenoo: so no userspace emulation - good ?
Mar 05 15:44:49 <reenoo> zecke: qemu-arm does userspace emultion to my knowledge
Mar 05 15:44:59 <reenoo> emulation even
Mar 05 15:45:51 <zecke> reenoo: I have glibc almost building on OSX, so please make that qemu usage optional
Mar 05 15:46:08 <reenoo> it is optional
Mar 05 15:46:14 <zecke> good :)
Mar 05 15:46:18 <koen> reenoo: qemu built by OE or an external one?
Mar 05 15:46:38 <reenoo> qemu-native_cvs
Mar 05 15:46:59 <zecke> ~lart mad_fcl
Mar 05 15:47:00 * ibot whips out a hot clue gun and makes sure that mad_fcl is stuck to the floor
Mar 05 15:47:03 <koen> perfectg
Mar 05 15:48:57 <koen> reenoo|afk: you rock!
Mar 05 15:49:17 <zecke> right
Mar 05 15:51:05 <Arthur> could anyone take a look http://rafb.net/paste/results/EQZqKF49.html please?
Mar 05 15:51:52 <koen> Arthur: which branch are you building from?
Mar 05 15:52:01 <koen> if it's from .dev, bitbake is too old
Mar 05 15:52:33 <Arthur> oz354fam083
Mar 05 15:55:12 <Arthur> so?
Mar 05 15:57:22 <zecke> Arthur: how many versions of bitbake do you have installed?
Mar 05 15:58:52 <Arthur> one
Mar 05 15:59:06 <pb_> reenoo|afk: heh, oh well.  those people should get faster computers, or build glibc less often.
Mar 05 15:59:16 <zecke> Arthur: wait a second
Mar 05 15:59:18 <polyonymous> Isn't it an old bitbake?
Mar 05 15:59:54 <Arthur> 1.3.3.2 revision 384 from cvs
Mar 05 16:00:00 <polyonymous> revision seems right...
Mar 05 16:00:24 <polyonymous> build tool core version is lower than mine... I'm running svn version as well.
Mar 05 16:00:57 <koen> pb_: when does localegen run? in do_rootfs or during glibc build?
Mar 05 16:01:06 <pb_> during glibc build
Mar 05 16:01:37 <pb_> running it during rootfs construction would be tricky and/or wasteful
Mar 05 16:01:53 <pb_> and, also, wouldn't actually solve the underlying problem
Mar 05 16:03:13 <koen> localegen only needs the bits from glibc, not foo-locale-de?
Mar 05 16:03:58 <pb_> indeed
Mar 05 16:04:42 <koen> I see, I thougt it also needed the other bits
Mar 05 16:04:46 <koen> hence my confusion
Mar 05 16:06:55 <zecke> Arthur: svn up your lib directory as well
Mar 05 16:07:23 <Arthur> how can I do that?
Mar 05 16:08:46 <zecke> Arthur: svn up lib?
Mar 05 16:09:07 <zecke> hey minipanda 
Mar 05 16:09:10 <zecke> hey mickeyl 
Mar 05 16:09:28 <zecke> mickeyl: do you know if it is perfectly legal to prepend a path to sys.path?
Mar 05 16:09:40 <koen> hey mickeyl 
Mar 05 16:10:35 <mickeyl> hi guys!
Mar 05 16:10:53 <mickeyl> zecke: yeah, that's valid
Mar 05 16:11:18 * mickeyl just returned from 4h squash ... yes it's ACHING everywhere
Mar 05 16:11:51 <koen> heh
Mar 05 16:30:04 <hrw> morning
Mar 05 16:30:09 <zecke> hey
Mar 05 16:30:45 <hrw> time to do some keylaunch related hacks
Mar 05 16:37:41 <Arthur> zecke, svn up lib said "skipped lib"
Mar 05 16:37:51 <koen> hey hrw 
Mar 05 16:38:03 <zecke> Arthur: google for a subversion documentation
Mar 05 16:38:14 <zecke> Arthur: I could ask from where you called svn up lib
Mar 05 16:38:28 <zecke> Arthur: but I think your are smarter than a turing machine
Mar 05 16:38:37 <Arthur> /home/arthur/oe/
Mar 05 16:38:54 <zecke> Arthur: I could ask, but I don't :)
Mar 05 16:39:05 <zecke> Arthur: and oe is a subversion checkout?
Mar 05 16:39:11 <hrw> ~seen law_
Mar 05 16:39:14 <ibot> law_ is currently on #gpe (3h 34m 58s) #oe (3h 34m 58s) #openzaurus (3h 34m 58s) #opie (3h 34m 58s), last said: 'is /usr/bin/aplay available ?'.
Mar 05 16:40:04 <reenoo> pb_: happen to remember how aux_files_pattern is supposed to work with do_split_packages?
Mar 05 16:40:09 <Arthur> nope.
Mar 05 16:40:27 <zecke> Arthur: so is it obvious why it doesn't work?
Mar 05 16:40:34 <law_> hi hrw
Mar 05 16:40:35 <zecke> Arthur: update your complete bitbake
Mar 05 16:40:48 <Arthur> so I should go to /home/arthur/oe/bitbake?
Mar 05 16:40:52 <zecke> Arthur: I have just tried with a current bitbake (from svn trunk) and it is working
Mar 05 16:40:59 <polyonymous> Arthur, and svn update there
Mar 05 16:41:12 <reenoo> pb_: do_split_packages does a the_files.append(aux_files_pattern % on)
Mar 05 16:41:26 <Arthur> doesn't my system config (I mean of my linux distro) doesn't interfere with oe building? 
Mar 05 16:41:46 <reenoo> pb_: where "on" is the (sanitized) package name
Mar 05 16:41:49 <pb_> reenoo: right, yeah. 
Mar 05 16:41:50 <Arthur> sorry for the English.
Mar 05 16:41:52 <polyonymous> Arthur, do you have a separate bitbake installed?
Mar 05 16:41:57 <zecke> Arthur: it shouldn't, when it does it is a bug in OE
Mar 05 16:42:19 <Arthur> I have only one bitbake on /home/arthur/oe/bitbake.
Mar 05 16:42:33 <reenoo> pb_: for the glibc locale packages I'd need that to be the non-sanitized name...
Mar 05 16:42:34 <Arthur> and I have a export PATH on my .bashrc to point to the bin dir.
Mar 05 16:42:41 <hrw> law_: I will try to put your collie changes from #722 into OE
Mar 05 16:42:51 <zecke> Arthur: well, trunk version of bitbake is version
Mar 05 16:42:57 <zecke> Arthur: so update your version to trunk?
Mar 05 16:42:59 <polyonymous> and you haven't install the one with distro?
Mar 05 16:43:12 <reenoo> pb_: i.e. en_GB instead of en-gb
Mar 05 16:43:15 <Arthur> Guys, just a minute. There are a lot of things I can't understand. 
Mar 05 16:43:31 <pb_> reenoo: ah yes, I suppose you would.
Mar 05 16:43:31 <Arthur> I have get bitbake from cvs and it is installed on /home/arthur/oe/bitbake.
Mar 05 16:43:40 <Arthur> I have to enter that dir and run svn up lib/
Mar 05 16:43:41 <Arthur> ?
Mar 05 16:43:47 <Arthur> then svn update?
Mar 05 16:43:53 <pb_> reenoo: I guess you'll have to patch package.bbclass to allow that.
Mar 05 16:44:18 <polyonymous> Arthur, enter that dir and see the output of svn info | grep '^\(Rev\|URL\)'
Mar 05 16:44:26 <zecke> Arthur: svn up == svn update
Mar 05 16:44:36 <reenoo> pb_: aux_files_pattern_verbatim ?
Mar 05 16:44:36 <zecke> Arthur: go to the bitbake dir and update
Mar 05 16:44:36 <Arthur> URL: svn://svn.berlios.de/bitbake/trunk/bitbake
Mar 05 16:44:36 <Arthur> Revisão: 384
Mar 05 16:44:36 <Arthur> Revisão da Última Mudança: 384
Mar 05 16:44:44 <pb_> reenoo: sounds good to me
Mar 05 16:44:46 <polyonymous> sounds good
Mar 05 16:44:51 <reenoo> pb_: ok
Mar 05 16:45:01 <law_> hrw: ok, perhaps you could take a look at cancel and contacs key, i cant find any solution
Mar 05 16:45:05 <Arthur> so what should I do now?
Mar 05 16:45:25 <polyonymous> Where does bitbake get two versions of --version output from?
Mar 05 16:45:27 <hrw> law_: for me cancel should stay as esc on press and suspend on hold
Mar 05 16:45:31 <zecke> Arthur: then run bitbake as python -v /home/arthur/oe/bitbake/bin/bitbake
Mar 05 16:45:42 <zecke> Arthur: which files does it load, and please use pastebin to paste it
Mar 05 16:46:29 <law_> hrw: but it doesnt work as esc. if you push it it closes the application
Mar 05 16:46:55 <law_> hrw: try to open rxvt an push cancel... very bad with vi
Mar 05 16:47:20 <Arthur> http://rafb.net/paste/results/d6XE8Z90.html
Mar 05 16:47:41 <reenoo> law_: get rid of the bogus collie /etc/matchbox/kbdconfig
Mar 05 16:47:47 <zecke> import bb.parse.parse_py # precompiled from /usr/lib/python2.4/site-packages/bb/parse/parse_py/__init__.pyc
Mar 05 16:47:52 <zecke> Arthur: you lied to me!
Mar 05 16:48:01 <Arthur> what?
Mar 05 16:48:09 <CoreDump|home> lol
Mar 05 16:48:10 <Arthur> sorry but what does it mean?
Mar 05 16:48:23 <polyonymous> maybe he has setup installed the older version?
Mar 05 16:48:33 <zecke> polyonymous: yes, again :)
Mar 05 16:48:40 <zecke> polyonymous: humans are error prone
Mar 05 16:48:46 <Arthur> could anyone explain me what is happening, please?
Mar 05 16:48:47 <zecke> Arthur: export PYTHONPATH as well
Mar 05 16:48:51 <hrw> reenoo: thx
Mar 05 16:48:58 <polyonymous> so the solution is to install it again ;-)
Mar 05 16:49:00 <law_> reenoo: thanks, i?ll have a look
Mar 05 16:49:00 <zecke> Arthur: well instead of /home/arthur/oe/bitbake/lib/bb
Mar 05 16:49:12 <Arthur> Nothing has happened with export PYTHONPATH
Mar 05 16:49:15 <zecke> Arthur: /usr/lib/python2.4/site-packages/bb is used
Mar 05 16:49:30 <zecke> Arthur: to what did you set PYTHONPATH?
Mar 05 16:49:46 <Arthur> I haven't set it. The Getting Started didn't mentioned it.
Mar 05 16:50:03 <polyonymous> Arthur, try ./setup.py install again
Mar 05 16:50:05 <zecke> Arthur: did the Getting Started mention to install a bitbake
Mar 05 16:50:18 <polyonymous> zecke, I believe it did ;-)
Mar 05 16:50:20 <zecke> Arthur: and then use the one in /home/arthur/oe/bitbake?
Mar 05 16:50:23 <Arthur> it mentioned to install but said that the best way was using it without installing...
Mar 05 16:50:30 <reenoo> zecke: why don't you finally get rid of setup.py?
Mar 05 16:50:40 <zecke> reenoo: why should I?
Mar 05 16:50:44 <Arthur> I have removed the bitbake package installed on my distro...
Mar 05 16:50:51 <Arthur> I don't know why it is there..
Mar 05 16:50:56 <zecke> reenoo: specially ones mithro binds the c-parser we will need setup.py for compiling
Mar 05 16:51:04 <polyonymous> Arthur, then just remove wha's left of it.
Mar 05 16:51:12 <Arthur> how? with rm ?
Mar 05 16:51:19 <Arthur> what should I remove?
Mar 05 16:51:19 <polyonymous> Precompiled stuff isn't getting removed?
Mar 05 16:51:26 <Arthur> yes.
Mar 05 16:51:40 <zecke> Arthur: set PYTHONPATH to /home/arthur/oe/bitbake/lib
Mar 05 16:51:41 <polyonymous> Arthur, /usrl/lib/python2.4/site-packages/bb/ seems to be a good target, but let zecke confirm it ;)
Mar 05 16:51:54 <reenoo> zecke: because people try to use it and fail
Mar 05 16:52:11 <polyonymous> reenoo, but shouldn't bitbake be installable in the end?
Mar 05 16:52:28 <zecke> reenoo: that would be a nice research topic, but it is unrealistic
Mar 05 16:52:39 <zecke> reenoo: e.g. at ROAD we only used packaged bitbake
Mar 05 16:53:00 <zecke> reenoo: and for compiling the c-parser module we need setup.py as well
Mar 05 16:53:42 <Arthur> zecke, should I remove with rm /usrl/lib/python2.4/site-packages/bb/ ?
Mar 05 16:53:49 <zecke> Arthur: no need to
Mar 05 16:53:57 <zecke> Arthur: set the PYTHONPATH in your bashrc as well
Mar 05 16:54:02 <Arthur> ok.
Mar 05 16:54:11 <polyonymous> zecke, what the use of leftovers scattered around hard disk?
Mar 05 16:54:35 <zecke> polyonymous: he might want to use his bitbake for other things
Mar 05 16:54:47 <zecke> polyonymous: and I don't want to be responsible for a sudo rm -rf /
Mar 05 16:55:07 <polyonymous> zecke, he DID uninstall it -it's precompiled stuff that wasn't removed.
Mar 05 16:55:11 <polyonymous> that's the point ;-))
Mar 05 16:55:32 <zecke> polyonymous: ah right, hmm
Mar 05 16:55:37 <zecke> mickeyl: still there?
Mar 05 16:55:40 <reenoo> zecke: then put some big fat warnings in there stating that python's deployment stuff is inherently broken, or add a shell wrapper around bitbake, or...
Mar 05 16:55:40 <mickeyl> kind of
Mar 05 16:56:13 <zecke> reenoo: ?? what about shared libraries and rpath usage? how many big fat warnings have you seen until today?
Mar 05 16:56:15 <law_> hrw: remove Escape=close in /etc/matchbox/kbdconfig and cancel=escape
Mar 05 16:56:18 <Arthur> everytime to bitbake a file it has to parse all the bb files? It takes too much time. Ins't there a fast way to test it?
Mar 05 16:56:25 <hrw> law_: I did it already
Mar 05 16:56:29 <zecke> mickeyl: do you now if the dist stuff should remove the pyc files?
Mar 05 16:56:34 <koen> Arthur: bitbake -i
Mar 05 16:56:47 <zecke> Arthur: bitbake -b to bake one file (wihtout dependencies)
Mar 05 16:57:10 <zecke> Arthur: http://bitbake.berlios.de/manual/ <- if you spot errors, please report them
Mar 05 16:57:17 <law_> hrw: found a solution for contacts already
Mar 05 16:57:26 <hrw> law_: cool
Mar 05 16:57:40 <Arthur> but if I try to bitbake nano it will be needed to have the dependencies, huh?
Mar 05 16:57:40 <hrw> law_: I do not like 'control on datebook' solution...
Mar 05 16:57:41 <law_> hrw: this was a question :-)
Mar 05 16:57:57 <zecke> reenoo: "If you are a DFA python might smart you out?" would be a sane warning?
Mar 05 16:58:00 <mickeyl> zecke: i think it doesn't, because they don't belong to the installation. it shouldn't hurt to leave them there though, because Python always checks the timestamps of the .py and the .pyc to find out if recompiling is necessary
Mar 05 16:58:09 <polyonymous> Arthur, bitbake -i and then bitbake whatever bbs you want.
Mar 05 16:58:19 <mickeyl> zecke: as long as you don't do stuff with PYTHONPATH
Mar 05 16:58:32 <mickeyl> zecke: in that case everything can happen
Mar 05 16:58:35 <reenoo> zecke: well. the current "wait for people to run into issues and spam #oe. then make fun of them." doesn't scale very well
Mar 05 16:58:35 <zecke> mickeyl: the issue with the pyc files were the .py equivalents were removed
Mar 05 16:58:42 <hrw> law_: I think that it should be: datebook/contacts/desktop/menu/tasks + keypad/space + return + cancel=espace
Mar 05 16:59:53 <reenoo> zecke: I guess setup.py could mention that in order to upgrade bitbake you'll need to make sure to set PYTHONPATH or wipe any old versions
Mar 05 17:00:13 <zecke> reenoo: it hasn't to do with upgrade
Mar 05 17:00:19 <zecke> reenoo: it has to do installing one copy
Mar 05 17:00:33 <zecke> reenoo: and then wondering that python applies python specification (look up order)
Mar 05 17:00:34 <polyonymous> the real problem here is not bitbake upgrade, but upgrading svn version while having crap resting /usr/
Mar 05 17:01:02 <reenoo> zecke: also, bitbake could check it's not trying to run against incompatible modules
Mar 05 17:01:12 <polyonymous> So, what about prepending path? Have you figured if it is harmless?
Mar 05 17:01:25 <reenoo> zecke: and output a reasonable error message
Mar 05 17:01:59 <mickeyl> since we can check the file path of any loaded module, we could alert if the prefixes are not matching
Mar 05 17:02:15 <zecke> reenoo: I could start distributing static builds of bitbake as well :}
Mar 05 17:02:47 <mickeyl> i'm not sure if it's worth the hassle though - isn't this the same with LD_LIBRARY_PATH? if you are installing stuff into non-standard directories, you're basically responsible to tell the system where the latest stuff is
Mar 05 17:03:10 <hrw> law_: mail holded = keycode 118
Mar 05 17:03:20 <zecke> reenoo: I somehow agree, but I find it tough to workaround peoples non understanding of technology
Mar 05 17:03:33 <zecke> mickeyl: worse, if you use rpath, LD_LIBRARY_PATH will be ignored
Mar 05 17:03:45 <mickeyl> eek! i wasn't aware of that
Mar 05 17:04:15 <reenoo> zecke: I guess you'll have to live with that if you publish your software
Mar 05 17:04:27 <law_> hrw: http://helmsdeep.homelinux.net/~pat/collie_keys.txt
Mar 05 17:04:49 <zecke> mickeyl: you would have to move the original lib away to make LD_LIBRARY_PATH work
Mar 05 17:04:55 <hrw> law_: thx
Mar 05 17:06:33 <zecke> well bitbake prepend the lib directory instead of appending
Mar 05 17:06:51 <zecke> this should fix the issues we have seen today, and maybe is creating new ones
Mar 05 17:12:10 <florian> i all
Mar 05 17:12:14 <florian> +h
Mar 05 17:12:15 <zecke> hey
Mar 05 17:12:19 <mickeyl> i florian
Mar 05 17:12:30 <reenoo> hey florian 
Mar 05 17:12:34 <hrw> hi florian 
Mar 05 17:14:02 <hrw> heh.. my wifi card decides to die soon ;9
Mar 05 17:14:06 <hrw> ;(
Mar 05 17:14:21 <mickeyl> darn
Mar 05 17:14:54 <mickeyl> do you need one?
Mar 05 17:14:58 <hrw> ~lart cf cards for being fragile
Mar 05 17:14:58 * ibot strangles cf cards with a 9-pole serial cable for being fragile
Mar 05 17:15:23 <pb_> florian: good afternoon
Mar 05 17:16:59 <hrw> mickeyl: this one will work for some time - added more paper into case to get it more squezzed so it goes back to 11Mb/s from 1Mb/s
Mar 05 17:17:58 <hrw> mickeyl: I will think about other one when this one will die
Mar 05 17:18:01 <mickeyl> hrw: heh, ok. when you are in desperate need for a new one, come back to me
Mar 05 17:18:11 <hrw> mickeyl: ok
Mar 05 17:18:19 <koen> mickeyl: didn't you use to collect wifi cards?
Mar 05 17:18:40 <mickeyl> koen: indeed, i had to - in order to make Wellenreiter / libopienetwork working on different kinds
Mar 05 17:18:54 <mickeyl> i just need one of every kind though
Mar 05 17:18:55 <mickeyl> not more
Mar 05 17:20:03 <pb_> "just"
Mar 05 17:20:04 <pb_> heh
Mar 05 17:20:17 <mickeyl> :)
Mar 05 17:20:23 <RP> [lala]: pong
Mar 05 17:20:34 <hrw> ~curse sharp for lack of ctrl on collie
Mar 05 17:20:37 <ibot> May the fleas of a thousand camels infest your most sensitive regions, sharp for lack of ctrl on collie !
Mar 05 17:20:53 <koen> mickeyl: http://dominion.kabel.utwente.nl/koen/OE/people/Michael%20Lauer-1.jpg
Mar 05 17:21:22 <zecke> data mining
Mar 05 17:21:25 <[lala]> RP: just to go sure - what is the expected situation after a fresh image install without anything plugged in? speaker should be on?
Mar 05 17:21:33 <RP> [lala]: yes
Mar 05 17:21:48 <zecke> koen: I'm sure you will hear from my lawyers :}
Mar 05 17:21:55 <mickeyl> koen: hehehe, this is the one and only event I had a beer in my hand.
Mar 05 17:22:00 <mickeyl> and you have captured it
Mar 05 17:22:10 <koen> heh
Mar 05 17:22:10 * mickeyl smiles innocently
Mar 05 17:22:14 <koen> 'one and only'
Mar 05 17:22:15 <koen> ;)
Mar 05 17:22:21 <zecke> mickeyl: how much lefe do you have left?
Mar 05 17:22:40 <mickeyl> zecke: all 6 bottles. didn't open the pack yet
Mar 05 17:22:42 <zecke> koen: and where is your image?
Mar 05 17:22:58 <zecke> mickeyl: pb_ believes you should have local leffe resellers as well?
Mar 05 17:23:22 <mickeyl> zecke: i'm still looking for any
Mar 05 17:23:34 <[lala]> RP: can you translate the meaning of those controls you are modifying in zaurus-mixer-callback? whats that right-mixer-left-playback for?
Mar 05 17:24:04 <reenoo> hrm
Mar 05 17:24:26 <reenoo> waiting for locales to be built is well.. boring
Mar 05 17:25:00 <RP> [lala]: mono audio is always supposed to be on the left channel. With a headset the mic is on the left channel and the single output channel is the right speaker. We therefore have to feed the left playback into the right mixer when using a headset
Mar 05 17:25:08 <koen> zecke: near the 'k'
Mar 05 17:26:32 <RP> reenoo|kitchen: Can we have a setting to say which locale's we care about?
Mar 05 17:26:52 <koen> RP: IMAGE_LINGUAS?
Mar 05 17:27:41 <[lala]> RP: so left-mixer-playback off means that this channels is in record mode then?
Mar 05 17:27:48 <koen> RP: or run a few of them in parallel
Mar 05 17:27:54 <hrw> koen: http://www.hrw.one.pl/img/avatar.png - or I can provide better one
Mar 05 17:28:47 <RP> [lala]: Its in some kind of input mode, yes
Mar 05 17:29:35 <[lala]> RP: and which would be the control to turn up the mic volume? left mixer?
Mar 05 17:29:50 * RP notes koen is missing his own picture ;-)
Mar 05 17:30:01 <hrw> RP: he is not
Mar 05 17:30:28 <RP> hrw: He's not now :)
Mar 05 17:30:37 <RP> [lala]: Mic Boost?
Mar 05 17:31:08 <RP> [lala]: or some of the ALC controls
Mar 05 17:31:28 <[lala]> RP: i totally hate that alsa mixer stuff ... :-/
Mar 05 17:31:33 <koen> I tried to get usefull heads from the fosdem pics, but not everyone one had one
Mar 05 17:31:43 <RP> [lala]: ALC Capture Target is probably a good one to fiddle with...
Mar 05 17:32:01 <hrw> koen: I have to create good one
Mar 05 17:32:01 <RP> [lala]: Unfortunately these are mic dependent...
Mar 05 17:32:27 <RP> koen: Who'd forgotten their head?
Mar 05 17:33:00 <mickeyl> hehe
Mar 05 17:33:10 <pb_> I guess too few heads makes a refreshing change.
Mar 05 17:33:44 <RP> [lala]: Also turn all the Bypass volumes down - you probably don't want the sidetone but turning that mixer off causes ASoC to break for some reason I can't work out
Mar 05 17:34:26 <koen> the pictures of pb_ I have either figure pb with his eyes closed or in red-eyed borg mode
Mar 05 17:34:50 <mickeyl> heh
Mar 05 17:35:02 <mickeyl> there are some cool red eye removal programs
Mar 05 17:35:08 <koen> mickeyl: http://dominion.kabel.utwente.nl/Gallery/bordeaux-stad2004/IMG_5883
Mar 05 17:35:37 <mickeyl> scary!
Mar 05 17:35:38 <[lala]> RP: i'll give it a try tomorrow morning, see if i can work it out finally
Mar 05 17:38:09 <CSMan> bordeaux
Mar 05 17:38:13 <RP> [lala]: ok
Mar 05 17:40:38 <hrw> smacznego mickey
Mar 05 17:40:44 <koen> ~bon appetit
Mar 05 17:40:46 <ibot> [bon appetit] smacznego.  Guten Appetit.  Eet Smakelijk.  God Appetitt.  Buon Appetito.  Buen apetito  Bom Apetite. buen apetito
Mar 05 17:43:01 <CSMan> do you guys know the actual current of wpa_supplicant on the H555x?
Mar 05 17:43:52 <CSMan> *current state
Mar 05 17:44:10 <ashak> Hmmm, how comes when I try to build something with bitbake today after updating my copy of oe, I get: NOTE: Not using a cache. Set CACHE = <directory> to enable. ERROR: no files to build.
Mar 05 17:45:27 <koen> ashak: your BBFILES and/or BBPATH is probably wrong
Mar 05 17:45:51 <Arthur> guys, when I finish to run bitbake nano... will it give me an ipk file?
Mar 05 17:46:45 <ashak> koen: Hmm, I havent' changed them since about a week ago when I last built something... so has something else changed that could affect it?
Mar 05 17:47:07 <ashak> koen: Doh... i'm being an idiot :)
Mar 05 17:48:03 <ashak> I wrote a script to setup BBFILES, I was running ./script instead of . ./script
Mar 05 17:48:16 <Bernardo> guys, I have a friend with both hentges and 3.5.4test5 who can't install any ipk I generate with fam083oz354 branch, any idea on what I am doing wrong? All images I generate install my ipks without any problem
Mar 05 17:48:51 <hrw> Bernardo: you build with machine=akita and he has a collie?
Mar 05 17:48:56 <Bernardo> yes
Mar 05 17:48:59 <Bernardo> ooops
Mar 05 17:49:23 * Bernardo somehow thought that since those weren't machine dependent apps...
Mar 05 17:49:23 <hrw> Bernardo: ;)
Mar 05 17:49:30 <Bernardo> subversion and apr
Mar 05 17:49:33 <law_> hrw: hmm i dont understand why the contacts-key  doesnt work :-(
Mar 05 17:49:44 <koen> Bernardo: armv5 vs armv4
Mar 05 17:50:00 <hrw> law_: I will flash new gpe-image to my collie to check it again from scratch (escape disabled)
Mar 05 17:50:08 <Bernardo> koen, hrw, you're both right
Mar 05 17:50:25 <Bernardo> I just forgot I don't have a collie anymore and I was building for a more recent cpu
Mar 05 17:51:38 <hrw> happens
Mar 05 17:51:39 <Bernardo> brb, time to boot into a gcc 4.0.2 compiled kernel
Mar 05 17:51:45 <Bernardo> yep
Mar 05 17:51:53 <Bernardo> thanks
Mar 05 17:53:06 <ade|desk> koen: which version of gallery do you use 
Mar 05 17:53:20 <koen> ade|desk: a very ancient 1.x one
Mar 05 17:53:56 <ade|desk> looks good
Mar 05 17:54:15 <ade|desk> as do the pictures 
Mar 05 17:54:18 <koen> that's just a few hours of fiddling with the css files
Mar 05 17:56:49 <CSMan> that's 2.x
Mar 05 17:57:15 <hrw> ARHGGG
Mar 05 17:57:21 <hrw> ~lart linux
Mar 05 17:57:21 * ibot hauls linux up by the scruff of the neck and spanks him until he waddles
Mar 05 18:00:11 <hrw> flashing 3.5.4-test6
Mar 05 18:01:33 <Arthur> florian, are you out there?
Mar 05 18:04:33 <florian> Arthur: yes
Mar 05 18:05:37 <CSMan> once i get a small app , how can i build it with the bitbake system ?
Mar 05 18:09:27 <Arthur> was you who posted an sdk image on the gpe list?
Mar 05 18:11:55 <jnc> koen|away: is gcc-4.1 in OE supposed to work?
Mar 05 18:12:06 <emte> jnc, not fully
Mar 05 18:12:40 <emte> they are ~ 1month behind on gcc actually
Mar 05 18:13:06 <jnc> oh hmm
Mar 05 18:13:07 <jnc> how so?
Mar 05 18:13:36 <emte> few problems by the looks
Mar 05 18:14:39 <zecke> jnc: gcc4.1 was not released last weak (I think it isn't released now as well)
Mar 05 18:14:43 <emte> http://gcc.gnu.org/ml/gcc/2006-01/msg00476.html
Mar 05 18:17:11 <hrw> law_: after matchbox-wm setting we need to fix keys&buttons behaviuor?
Mar 05 18:18:08 <CIA-1> 03hrw 07org.oe.oz354fam083 * rc02f9f04... 10/packages/matchbox-wm/ (matchbox-wm/collie/kbdconfig matchbox-wm_svn.bb): 
Mar 05 18:18:13 <CIA-1> matchbox-wm: removed escape=close setting from collie config
Mar 05 18:18:15 <CIA-1> - now vi is usable in rxvt
Mar 05 18:18:17 <CIA-1> - escape button is usable
Mar 05 18:18:33 <law_> hrw: yes, only add my *clamshell.patch into gpe-conf
Mar 05 18:19:05 * emte wonders ... if familiar is going to scratchbox is gpe as well ...
Mar 05 18:19:15 <law_> hrw: i found some strange thing with the contacts key, have you got your collie here?
Mar 05 18:19:24 <hrw> law_: have
Mar 05 18:19:39 <florian> emte: eh?
Mar 05 18:19:58 <law_> try xev on the contacts key. there is some strange sign in...
Mar 05 18:20:12 <law_> take a screenshot .....
Mar 05 18:20:25 <emte> florian, http://www.handhelds.org/hypermail/familiar-dev/current/0735.html
Mar 05 18:20:59 <hrw> law_: matchbox-wm is faster on grabbing key then xev
Mar 05 18:21:35 <law_> hrw i turn off the keys in kbdconfig..
Mar 05 18:22:32 <hrw> ok. disabled, restarted
Mar 05 18:23:22 <florian> emte: interesting... i wonder how long that mail will take to show up here :-/
Mar 05 18:23:51 <emte> no idea
Mar 05 18:24:04 <emte> but from the next post 0.8.3 is ready to ship
Mar 05 18:24:27 <emte> no point in filing another bug against it i guess
Mar 05 18:25:31 <law_> hrw: http://helmsdeep.homelinux.net/~pat/contactskey_collie.png
Mar 05 18:26:38 <hrw> law_: hm
Mar 05 18:27:16 <law_> hrw: could you try this?
Mar 05 18:28:00 <hrw> why konqueror 3.5.1 has to freeze on RMB??? WHY?
Mar 05 18:28:03 <hrw> ~lart konq
Mar 05 18:28:04 * ibot tries to shut konq up
Mar 05 18:28:35 <hrw> law_:     XLookupString gives 0 bytes:
Mar 05 18:28:36 <law_> hrw: if you remove F2=!matchbox-remote -mbmenu from kbdconfig you also could configure menu-key in .keylaunchrc
Mar 05 18:28:48 <hrw> law_: I know
Mar 05 18:29:43 <hrw> law_: we need collie picture anyway
Mar 05 18:30:01 <jnc> i guess i'll work on gcc 4.2
Mar 05 18:30:22 * hrw will kill any konq/kde developer which will get into my hands
Mar 05 18:30:24 <emte> jnc, ?
Mar 05 18:30:26 <law_> hrw: for gpe-conf ? 
Mar 05 18:30:30 <hrw> law_: yes
Mar 05 18:30:37 <emte> gcc4.2 is based on gcc4.1 ,,,
Mar 05 18:30:37 <jnc> emte: koen has 4.1, i'll take 4.2
Mar 05 18:30:47 <jnc> they are parallel branches though
Mar 05 18:31:07 <emte> yeah but isnt gcc4.2 only accepting proposals ?
Mar 05 18:31:15 <jnc> look at gcc.gnu.org if you have any questions
Mar 05 18:32:13 <emte> i am/have
Mar 05 18:32:55 <law_> hrw: http://handhelds.org/scap/port.18379.png
Mar 05 18:34:24 <emte> jnc, from what i am reading there will be little/no difference from 4.1->4.2
Mar 05 18:35:07 * emte shrugges
Mar 05 18:35:17 <hrw> law_: patches ready? :)
Mar 05 18:38:06 <jnc> what is multilib and why do we disable it
Mar 05 18:38:33 <florian> aha? i seee reasons...
Mar 05 18:38:34 <florian> ~lart france
Mar 05 18:38:35 * ibot throws france's poor little doggy off a cliff
Mar 05 18:38:50 <emte> florian, ?
Mar 05 18:39:00 <law_> hrw: push them to a webspace...
Mar 05 18:39:11 <hrw> law_: key=???PressedXF86Calendar:-:gpe-calendar
Mar 05 18:39:24 <hrw> law_: should not it be s/?/. rather?
Mar 05 18:40:08 <_law_> hrw: yes
Mar 05 18:41:41 <hrw> argh... 
Mar 05 18:42:48 <hrw> florian: can gpe-conf be able to configure keys by "I press key, you recognise which I pressed and let me configure what it has to do" rather then "edit /etc/gpe/key-layout, provide pixmap and try to get it working"?
Mar 05 18:44:16 <_law_> hrw: http://members.a1.net/patrick.steiner/collie-keylayout/collie-keylayout.tar.gz
Mar 05 18:45:42 <Zero_Chaos> anyone know how to fix:
Mar 05 18:45:43 <Zero_Chaos> NOTE: exceptions.SyntaxError:EOL while scanning single-quoted string (line 1) while evaluating:
Mar 05 18:45:46 <Zero_Chaos> ?
Mar 05 18:47:43 <florian> re
Mar 05 18:52:02 <jnc> Zero_Chaos: it sounds exactly like it sounds, though
Mar 05 18:53:22 <Zero_Chaos> jnc: as helpful as I'm sure you are trying to be, can you give me a little bit better of a pointer?
Mar 05 18:56:14 <hrw> _law_: http://pastebin.com/585425 is my .keylaunchrc and it is not working ;(
Mar 05 18:57:35 <hrw> hm
Mar 05 18:57:52 <_law_> hrw: remove XF86PowerDown doesnt suspend if you add this. XF86Calendar is mapped to ctrl or?
Mar 05 18:58:45 <hrw> _law_: calendar is calendar.
Mar 05 18:58:59 <hrw> _law_: mail and home works. calendar and contacts are not
Mar 05 18:59:00 <_law_> hrw: and ctrl ?
Mar 05 18:59:37 <hrw> _law_: there is no ctrl
Mar 05 18:59:40 <_law_> removed stuff from key kbdconfig
Mar 05 19:00:01 <hrw> _law_: removed from /etc/matchbox/kndconfig everything collie related
Mar 05 19:02:47 <_law_> hrw: did you use my configs?
Mar 05 19:04:31 <hrw> brb
Mar 05 19:04:34 <_law_> hrw: if you use my collie.xmodmap calendar is mapped to ctrl
Mar 05 19:06:18 <hrw> _law_: http://pastebin.com/585436
Mar 05 19:08:49 <hrw> _law_: and your key-layout with enabled calendar key
Mar 05 19:10:50 <_law_> hrw: mom i?ll try to reenable calendar key on my xmodmap, i dont know why contacts key doesnt work, perhaps this (http://helmsdeep.homelinux.net/~pat/contactskey_collie.png) could be the problem....
Mar 05 19:11:44 <_law_> hrw: key=...Held telephone:-:rxvt and key=...Released telephone:-:rxvt  should work
Mar 05 19:12:02 <hrw> will check
Mar 05 19:13:46 <hrw> _law_: held works. released not
Mar 05 19:17:23 <_law_> hrw: same problem with calendar-key held works....
Mar 05 19:17:27 <_law_> hmm
Mar 05 19:18:39 <_law_> hrw: try to launch keylauch from rxvt (kill running one before) and you will get some output
Mar 05 19:18:50 <hrw> I 'love' way how x11 handle keyboards...
Mar 05 19:19:18 <_law_> so the strange sign in my screenshot could be the problem i think....
Mar 05 19:20:25 <_law_> did noticed the calendar key problem because i always mapped it to ctrl
Mar 05 19:20:46 <hrw> _law_: without keylaunch running I do not that that strange sign which you got
Mar 05 19:21:19 <_law_> ahh contacts is mapped to alt ! try calendar + contacts + backspace
Mar 05 19:21:31 <_law_> kill the xserver
Mar 05 19:22:09 <hrw> _law_: I know - but I want calendar/contacts as keys...
Mar 05 19:23:12 <_law_> hrw: but this could be the problem ctrl + alt are "special-keys" for keylaunch
Mar 05 19:24:11 <hrw> _law_: so here we came to start.
Mar 05 19:24:27 <hrw> ~lart x11 for lack of normal keyboard handling
Mar 05 19:24:27 * ibot throws x11's poor little doggy off a cliff for lack of normal keyboard handling
Mar 05 19:24:58 <hrw> all those xmodmap files, lack of kernel keymap support...
Mar 05 19:25:06 <_law_> hrw: look at the example_rc in the source dir
Mar 05 19:25:38 <hrw> if machine has keyboard then you need to make kernel keymap, x11 keymap and there always will be something (a key) which will not work...
Mar 05 19:26:44 <hrw> _law_: in other words - we will not get those keys working under x11 probably
Mar 05 19:27:27 <_law_> hrw: we can try to change stuff in kbdconfig, but gpe-conf doesnt work after that ...
Mar 05 19:27:58 <hrw> _law_: kbdconfig is matchbox-wm stuff which I prefer to avoid because of its bugs
Mar 05 19:28:28 <_law_> or someone could write a patch for keylaunch do disable "special-keys" for collie
Mar 05 19:28:44 <hrw> fsck. let someone shoot all those 'project people' which reiinvent wheel
Mar 05 19:28:56 <zecke> koen|away: I consider buying the Tomados dubble friteuse from blokker
Mar 05 19:29:21 <hrw> _law_: and then someone will connect external keyboard...
Mar 05 19:29:37 <_law_> hrw: right
Mar 05 19:31:29 <_law_> the easiest way will be to only make Home/Menu/Mail buttons configureable in gpe-conf (keylaunch) and "hardcode" calendar and contacts to ctrl and alt
Mar 05 19:31:30 <ashak> Hmm, konqueror-embedded just failed to compile.
Mar 05 19:31:51 <hrw> I will probably NEVER understood how X11 manage to get keys working... for kernel Calendar is keycode 96 which can be changed to F24 for example. for X11 we set it to XF86Calendar and I expect X11 to handle it as XF86Calendar. but it prefer to handle it as Ctrl (which is hardcoded somewhere inside of it)
Mar 05 19:31:57 <hrw> ARGH
Mar 05 19:32:15 <zecke> ashak: pcre issue?
Mar 05 19:32:36 <ashak> zecke: Just trying to find the error in the log file, hold on...
Mar 05 19:32:50 <zecke> ashak: (I guess you need to install pcre dev files on your host)
Mar 05 19:33:54 <hrw> _law_: I'm going to IGNORE that problem and do that fscking 3.5.4 release
Mar 05 19:34:10 <_law_> hrw: or we set calendar to gpe-calendar and contacts to gpe-contacts in kbdconfig
Mar 05 19:34:36 <hrw> info in ReleaseNotes: do now even TRY to use ;keys&buttons; on zaurus or you will have to do some hard work.
Mar 05 19:35:00 <hrw> _law_: we have it like this already iirc
Mar 05 19:35:44 <hrw> reflashing back..
Mar 05 19:35:45 <ashak> zecke: pcre doesn't appear anywhere in the file... 
Mar 05 19:36:02 <_law_> hrw: if you add my keylaunch_clamshell.patch the problem wound be there
Mar 05 19:36:07 <zecke> ashak: what is the error, feel free tzo pastebin it
Mar 05 19:36:43 <ashak> zecke: I could be wrong, this is the last line I can find before it tells me Error: /openembedded/build/tmp/staging/arm-linux/qt2/include/qrangecontrol.h:52: warning: 'class QRangeControl' has virtual functions but non-virtual destructor, make[5]: *** [kjs_navigator.lo] Error 1
Mar 05 19:36:51 <ashak> pastebin?
Mar 05 19:36:52 <_law_> and also add different config-sets (clamshell/collie) for gpe-conf 
Mar 05 19:36:53 <hrw> _law_: you had r/w powers - why not added it yet?
Mar 05 19:37:24 <zecke> ~pastebin
Mar 05 19:37:27 <ibot> pastebin is probably a place to paste your stuff without flooding the channel - try http://pastebin.com/, or http://pastebin.ca, or http://channels.debian.net/paste
Mar 05 19:37:33 <hrw> _law_: the problem is the same as always: lack of time + lack of gpe/x11 keyboard managing
Mar 05 19:37:38 <hrw> knowledge
Mar 05 19:37:41 <zecke> ashak: is it compiling with -Werror?
Mar 05 19:37:49 <zecke> ashak: which version of gcc are you using?
Mar 05 19:37:54 <_law_> hrw: dont have r/w lost my private monotone key (hd crash) have to build a now one...
Mar 05 19:38:29 <ashak> zecke: Erm, bitbake built gcc-4.0.2 a moment ago i think... so would it be using that?
Mar 05 19:38:38 <zecke> ashak: right
Mar 05 19:38:39 <hrw> _law_: could we fix that keyboard problems after release?
Mar 05 19:39:38 <_law_> hrw: yes i?ll have a look at it again this/next week (and build some .bb incl patches) for OE so we can add it do updates feed asap.
Mar 05 19:40:29 <hrw> _law_: good. if you will find time with florian to get it nice packaged for gpe-conf then it would best
Mar 05 19:40:31 <_law_> we need differnt gpe-conf version
Mar 05 19:40:33 <ashak> zecke: -Werror doesn't appear in the log file either.
Mar 05 19:41:07 <hrw> _law_: no - we need gpe-conf-data
Mar 05 19:42:01 <_law_> hrw: ok gpe-conf-data with config files for collie, huky...?
Mar 05 19:42:14 * _law_ looks at florian
Mar 05 19:43:05 <hrw> _law_: and default for ipaq with things which it contain now
Mar 05 19:43:39 <hrw> _law_: so gpe-conf-data_arm will have ipaq stuff inside like gpe-conf has now and we will get collie/c7x0 version too
Mar 05 19:44:18 <_law_> hrw: ok but i should we handle the keylaunch_clamshell.patch ?
Mar 05 19:44:47 <hrw> _law_: thats thing which should be discussed not with me but with florian
Mar 05 19:45:49 <_law_> is someone here who can test keylaunch with an Ipaq ? i dont have one :-(
Mar 05 19:46:19 <hrw> monotone:    1,4 M |   228,9 k |     1269 |     296 |            3
Mar 05 19:48:51 <ashak> zecke: Hmm, trying to compile again it tells me: konq-embed/kdesrc/khtml/ecma/../../kjs/ustring.h: In member function 'unsigned char& KJS::UCharReference::low() const':
Mar 05 19:49:06 <ashak> konq-embed/kdesrc/khtml/ecma/../../kjs/ustring.h:157: error: cannot bind packed field '(+ KJS::UCharReference::ref())->KJS::UChar::lo' to $
Mar 05 19:49:24 <ashak> 'unsigned char&'
Mar 05 19:49:33 <zecke> ashak: okay I think others have seen this error before
Mar 05 19:49:41 <zecke> ashak: I sadly lack the time to look into it now
Mar 05 19:49:48 <zecke> ashak: use gcc 3.4.2 for konqueror :}
Mar 05 19:49:57 <_law_> hrw: ok perhaps florian finds some time this week.
Mar 05 19:50:25 <ashak> zecke: Heh, ok, that's fine, silly question though, how can I tell it to use a different gcc?
Mar 05 19:50:40 <zecke> ashak: using PREFERRED_VERSION for the cross compiler
Mar 05 19:50:55 <zecke> ashak: e.g. take a look at the conf/distro/* files
Mar 05 19:51:00 <ashak> zecke: ok, cool thanks
Mar 05 19:52:55 <florian> _law_, hrw: Good thinking... that might be faster for now.
Mar 05 19:54:26 <_law_> florian: could you take a look at my keylaunch_clamshell.patch? doest keylaunch work on ipaq after that ?
Mar 05 19:55:36 <florian> _law_: oops, yes i should do that asap
Mar 05 19:57:45 <hrw> CoreDump|bbl: altboot should be bumped >1.0 - we had altboot 1.0 already
Mar 05 20:05:42 <hrw> http://handhelds.org/scap/port.27372.png - any ideas why I cant have 3 applets on right side?
Mar 05 20:08:00 <_law_> hrw: tried to move it with panel manager?
Mar 05 20:08:09 <hrw> lardman: yes
Mar 05 20:08:16 <hrw> _law_: always two
Mar 05 20:09:27 <hrw> added more then usual: http://handhelds.org/scap/port.28078.png
Mar 05 20:09:47 <_law_> hrw: cant reproducte it http://handhelds.org/scap/port.18379.png (i have installed your test5 image) 
Mar 05 20:10:38 <_law_> but i think you have got smaller font size or?
Mar 05 20:10:48 <_law_> and icon size
Mar 05 20:11:01 <chouimat> AAAAARRRRR!!!
Mar 05 20:11:12 <hrw> _law_: when there are lots of them then its ok. it looks like 'if there is XX pixels ocupied on left then allow to more on right'
Mar 05 20:11:45 <hrw> _law_: icons:24, font:6
Mar 05 20:14:33 <chouimat> hi hrw 
Mar 05 20:14:39 <hrw> hi chouimat 
Mar 05 20:14:57 <_law_> hrw: http://handhelds.org/scap/port.28821.png havent changed anything
Mar 05 20:15:18 <hrw> _law_: current gpe-image works quite good on collie. keys works, escape is fixed
Mar 05 20:15:25 <hrw> _law_: so I'm out of luck
Mar 05 20:16:05 <hrw> time to go - cu
Mar 05 20:16:12 <_law_> hrw: cu
Mar 05 20:16:41 <_law_> i also leaving n8 all
Mar 05 20:21:00 * france is back (gone 18:30:33)
Mar 05 20:51:11 <Eddy> Hi everybody I'm still on my quest to include python in my gpe-image.  I made a copy of the gpe-image.bb added 'python' to the depends and added all the 'python-*' modules to the ipkg_install.  My problem is that 'python-curses' rdepends on 'libncurses5' and I keep getting the error message:  'ERROR: Nothing provides runtime dependency libncurses5', anyone know a way to rectify this situation?
Mar 05 20:55:47 <johnX> the package name shows up as ncurses on my zaurus
Mar 05 20:56:16 <Eddy> Exactly and I have an ncurses .ipkg I don't know why its saying this
Mar 05 20:56:53 <johnX> I don't know, but it could be a bug
Mar 05 20:57:26 <Eddy> yeah I had a similar problem with python-zlib
Mar 05 20:57:32 <Eddy> looking for dependency libz1
Mar 05 20:59:57 <johnX> things that are far outside the scope of gpe-image and opie-image get much less testing I think
Mar 05 21:00:40 <Eddy> probably
Mar 05 21:04:00 <Eddy> crap and if I keep in python-db it bombs out because it can't find libgdm3
Mar 05 21:04:08 <Eddy> libgdbm3
Mar 05 21:04:58 <jnc> erg, compiler not working
Mar 05 21:25:48 <zecke> good nite
Mar 05 21:27:19 <jnc> Zero_Chaos: if it says unterminated string, maybe it is an unterminated string?
Mar 05 21:39:52 <CoreDump|home> hrw|gone: well it was only "1.0" because no PV was defined
Mar 05 21:40:12 <polyonymous> my bitbake died on sylpheed ;-)
Mar 05 21:40:53 <polyonymous> undefined references to SSL stuff...
Mar 05 21:41:28 <johnX> polyonymous: I posted a bug about that...must not have been acted on
Mar 05 21:41:44 <polyonymous> johnX, including solution?
Mar 05 21:41:47 <johnX> Anyways, I also posted a .bb for a newer version of sylpheed that doesn't have the issue
Mar 05 21:42:10 <polyonymous> Aha, thank you, so I'll check bugzilla
Mar 05 21:42:43 <johnX> the solution to make that old version of sylpheed work involves patching the new version of gnutls that got into OE to support some openssl functions
Mar 05 21:43:08 <polyonymous> I'm fine with going for a newer version
Mar 05 21:45:22 <johnX> great, if it gets some more testing maybe some kind soul will add it to OE
Mar 05 21:46:10 <polyonymous> at this time I can't give it much testing as I'm testing the whole gpe thing in chrooted environment. I want to change from cacko, but I want to have all the stuff I want to have ;-)
Mar 05 21:46:46 <polyonymous> I mean, I will test it, of course, but I wouldn't rely on this test.
Mar 05 21:46:52 <johnX> that's fine
Mar 05 21:48:11 <johnX> I haven't been testing it exhaustively either, and this is the .dev branch so hopefully people have more sense than to keep all their important work email on the Zaurus with no backups
Mar 05 21:49:14 <polyonymous> Well, I normally just check imap from pda, leaving the real thing for desktop.
Mar 05 21:50:01 <polyonymous> baking some bits again now.
Mar 05 21:57:12 <polyonymous> johnX, so far sylpheed compiled okay, baking more bits now ;-)
Mar 05 21:57:19 <johnX> good
Mar 05 21:58:16 <polyonymous> Oh well, bath time now  ;-)
Mar 05 22:28:37 <florian> re
Mar 05 22:56:41 <mithro> welp, i'm no longer in EUrope
Mar 05 23:14:29 <florian> re
Mar 05 23:53:02 <jnc> errgg
Mar 06 00:29:43 <ggilbert> I hear there are problems with treke.net?
Mar 06 00:29:46 <ggilbert> What's going on?
Mar 06 00:30:12 <johnX> ggilbert: From what I can see it works now
Mar 06 00:30:22 <ggilbert> When was it down?
Mar 06 00:30:39 <johnX> The day before yesterday I think
Mar 06 00:30:49 <ggilbert> hmmm. 
Mar 06 00:30:53 <johnX> or possibly the day before that, I don't properly recall.  It's been a busy week
Mar 06 00:31:20 <ggilbert> interesting. uptime is a day and a half
Mar 06 00:31:24 <ggilbert> never saw it go down :p
Mar 06 00:33:20 <ggilbert> ~kergoth
Mar 06 00:33:22 <ibot> well, kergoth is Chris Larson, an open source developer, often in the embedded Linux arena.  Currently unemployed.  Contact at kergoth@handhelds.org.  Core dev / architect on the BitBake, OpenZaurus, and OpenEmbedded projects.  Maintainer of the tslib project.  Often the resident BOFH, and often an ass.  Newbies and the clueless beware.
Mar 06 00:34:07 <ggilbert> ~mickeyl
Mar 06 00:34:08 <ibot> somebody said mickeyl was our local python guru, opie/openzaurus/openembedded/bitbake/opensimpad developer, and generally in charge of way too many things
Mar 06 00:34:14 <emte> pretty sure kergoth works now :)
Mar 06 00:34:18 <ggilbert> hehe
Mar 06 00:34:24 <ggilbert> Just trying to dig up his email address :p
Mar 06 00:52:54 <jnc> ehhh... help, competent OE devs
Mar 06 00:53:06 <jnc> what is the purpose of our gcc builds?
Mar 06 00:53:14 <jnc> gcc, gcc-cross, gcc-cross-initial
Mar 06 00:53:37 <emte> host host->device device specific
Mar 06 00:54:09 <champ> anyone selling a zaurus for cheap?
Mar 06 00:54:15 <emte> i think its only things like glibc/uclibc that use all three
Mar 06 01:13:25 <florian> good night
Mar 06 01:13:38 <champ> night
Mar 06 01:17:54 <jnc> champ: i'm selling my C3000 soon
Mar 06 01:18:01 <jnc> i think i will sell it to a co-worker though
Mar 06 01:22:20 <champ> jnc: how much are you asking?
Mar 06 01:29:20 <jnc> i think $300usd-ish for the unit
Mar 06 01:30:06 <jnc> i'm replacing it with a C3100, which i am waiting to receive
Mar 06 01:30:31 <champ> i am waiting to see if they come out with a ram upgraded c3200
Mar 06 01:30:39 <jnc> if the C3100 is in good shape (i bought it used at auction), then maybe i will hang on to the C3000 until bugs are fixed in the kernel, and sell it
Mar 06 01:53:56 <jnc> oh
Mar 06 01:54:00 <jnc> i don't know about c3200
Mar 06 01:54:01 <jnc> hmm
Mar 06 03:08:02 <douglee> I tried to install monotone-0.25 rpm for Fedora Core, I got:
Mar 06 03:08:07 <douglee>         libstdc++.so.6(GLIBCXX_3.4.6) is needed by monotone-0.25-1.i386
Mar 06 03:08:41 <douglee> I also tried to use http://venge.net/monotone/downloads/monotone-0.25-linux-x86.bz2 on my Ubunto box
Mar 06 03:09:02 <douglee> monotone-0.25-linux-x86: /usr/lib/libstdc++.so.6: version `GLIBCXX_3.4.6' not found (required by monotone-0.25-linux-x86)
Mar 06 03:09:09 <douglee> Did I miss something?
Mar 06 03:38:33 <njs> douglee: it sounds like you're running an old distro
Mar 06 03:39:14 <njs> douglee: hmm, and it doesn't look like anyone has contributed a fc3 rpm for 0.25
Mar 06 03:40:10 <njs> douglee: http://www.ipd.uni-karlsruhe.de/~moschny/monotone/monotone-0.25-x86-semistatic.bz2 may or may not work for you
Mar 06 03:40:57 <njs> douglee: oh, on an ubuntu giving that error, http://venge.net/monotone/downloads/monotone_0.25-0sarge1_i386.deb may also work...
Mar 06 03:52:49 <douglee> I'm running Ubunto 5.10 and Fedora Core 4
Mar 06 03:53:21 <Zero_Chaos> douglee: sorry about your luck
Mar 06 03:54:03 <njs> douglee: hrm, that's weird, the rpm is specifically for fc4
Mar 06 03:55:51 <douglee> I'm trying to use monotone-0.25-x86-semistatic.bz2 as you mentioned
Mar 06 03:56:45 <njs> worst comes to worse, there are source packages for both rpm and deb systems there, that should not be too hard to build.
Mar 06 03:59:54 <douglee> $ sudo dpkg -i monotone_0.25-0sarge1_i386.deb
Mar 06 03:59:55 <douglee> (Reading database ... 89739 files and directories currently installed.)
Mar 06 03:59:55 <douglee> Preparing to replace monotone 0.25-0sarge1 (using monotone_0.25-0sarge1_i386.deb) ...
Mar 06 03:59:58 <douglee> Unpacking replacement monotone ...
Mar 06 04:00:00 <douglee> dpkg: dependency problems prevent configuration of monotone:
Mar 06 04:00:03 <douglee>  monotone depends on libboost-date-time1.32.0; however:
Mar 06 04:00:05 <douglee>   Package libboost-date-time1.32.0 is not installed.
Mar 06 04:00:08 <douglee>  monotone depends on libboost-filesystem1.32.0; however:
Mar 06 04:00:10 <douglee>   Package libboost-filesystem1.32.0 is not installed.
Mar 06 04:00:13 <douglee>  monotone depends on libboost-regex1.32.0; however:
Mar 06 04:00:15 <douglee>   Package libboost-regex1.32.0 is not installed.
Mar 06 04:00:18 <douglee> dpkg: error processing monotone (--install):
Mar 06 04:00:20 <douglee>  dependency problems - leaving unconfigured
Mar 06 04:00:23 <douglee> Errors were encountered while processing:
Mar 06 04:00:26 <douglee>  monotone
Mar 06 04:00:28 <douglee> $ dpkg -l|grep libboost
Mar 06 04:00:31 <douglee> ii  libboost-date-time1.33.0               1.32.0+1.33.0-cvs20050727-1ubuntu1 set of date-time libraries based on generic
Mar 06 04:00:34 <douglee> ii  libboost-dev                           1.32.0+1.33.0-cvs20050727-1ubuntu1 Boost.org libraries development files
Mar 06 04:00:37 <douglee> ii  libboost-filesystem1.33.0              1.32.0+1.33.0-cvs20050727-1ubuntu1 filesystem operations (portable paths, itera
Mar 06 04:00:40 <douglee> ii  libboost-regex1.33.0                   1.32.0+1.33.0-cvs20050727-1ubuntu1 regular expression library for C++
Mar 06 04:00:43 <douglee> ii  libboost-thread-dev                    1.32.0+1.33.0-cvs20050727-1ubuntu1 portable C++ multi-threading
Mar 06 04:00:46 <douglee> ii  libboost-thread1.33.0                  1.32.0+1.33.0-cvs20050727-1ubuntu1 portable C++ multi-threading
Mar 06 04:01:02 <njs> douglee: hmm, right, looks like breezy does have boost 1.33
Mar 06 04:01:27 <njs> douglee: what does 'dpkg -s libboost-date-time1.33.0 | grep Depends' say?
Mar 06 04:01:45 <Zero_Chaos> douglee: --force-deps or how ever you do it in debian
Mar 06 04:01:50 <njs> Zero_Chaos: no
Mar 06 04:02:09 <njs> Zero_Chaos: these dependencies really are needed :-)
Mar 06 04:02:18 <Zero_Chaos> njs: yeah, and he has them...
Mar 06 04:02:30 <Zero_Chaos> but hey, what do I know....
Mar 06 04:02:55 <njs> Zero_Chaos: no, he has a different version of them, that the binary in that package is not linked against
Mar 06 04:03:40 <Zero_Chaos> njs: and if he has 1.33-cvssnapshot, but it needs 1.32, it really should work, don't you think?
Mar 06 04:05:12 <njs> Zero_Chaos: according to packages.ubuntu.com, that package contains usr/lib/libboost_date_time-gcc-mt-1_33.so.1.33.0
Mar 06 04:05:48 <Zero_Chaos> njs: maybe I'm a fool, but that can be fixed with ln.  but hey, don't argue with me, I'm a nobody, help the kid.
Mar 06 04:10:57 <njs> douglee: anyway, hmm, I guess ubuntu has an unlucky mix of packages, that doesn't quite match anything debian ever had...
Mar 06 04:45:36 <douglee> anyway, monotone-0.25-x86-semistatic works for me.
Mar 06 04:45:40 <douglee> thanks njs :)
Mar 06 04:45:49 <njs> douglee: ah, cool :-)
Mar 06 04:47:06 <douglee> since I switch buildroot to oe, hope oe config file is not hard for me
Mar 06 05:22:36 <douglee> Cool, oe config file is easy to modify. Now I'm starting building opie-image for h1940
Mar 06 05:29:03 <douglee> Err, it seems no such file http://www.oesources.org/source/current/quilt_cvs.savannah.nongnu.org_VER_0_42_.tar.gz
Mar 06 06:04:29 <kergoth> douglee: that isnt an error
Mar 06 08:15:57 <alan|home> morning all
Mar 06 08:15:59 <Cockroach-> morning
Mar 06 08:22:57 <hrw|work> morning
Mar 06 08:30:49 <hrw|work> meeting
Mar 06 08:38:43 <[lala]> RP: ping
Mar 06 08:44:53 <_law_> morning hrw|work 
Mar 06 09:01:14 <zecke> moin
Mar 06 09:05:03 <XorA> morning
Mar 06 09:05:44 <koen|770> Hi all
Mar 06 09:06:10 <RP> morning all
Mar 06 09:06:13 <RP> [lala]: pong
Mar 06 09:06:20 * koen|770 abuses the unprotected wifi at the policestation
Mar 06 09:06:31 <[lala]> RP: morning, your patch was successfull :-)
Mar 06 09:06:39 <[lala]> RP: all is working as expected now
Mar 06 09:06:43 <RP> [lala]: excellent :)
Mar 06 09:06:58 <[lala]> RP: that was quite a relief for me :-)
Mar 06 09:07:25 <[lala]> RP: much thanks for your support
Mar 06 09:07:49 <RP> [lala]: I had it all working so I knew it shouldn't be anything major :)
Mar 06 09:11:36 <katossi_uni> morning all!
Mar 06 09:12:37 <katossi_uni> [lala]: if RP got a cent everytime he helped with some patch or something, we could have another Bill Gates :)
Mar 06 09:15:48 <koen|sleep> nick koen|work
Mar 06 09:16:35 <[lala]> katossi_uni: :-)
Mar 06 09:17:25 <obergix[work]> hi
Mar 06 09:18:14 <koen|work> hey obergix[work] 
Mar 06 09:37:57 <greentux> morning
Mar 06 09:39:18 <RP> hi greentux 
Mar 06 09:39:28 <greentux> Hi RP, all waorks fine... 
Mar 06 09:39:51 <RP> greentux: great! I like it when people say that :)
Mar 06 09:51:39 <hrw|work> that reminds me that I had to test 2.6.15-mmc-fix 
Mar 06 09:52:33 <hrw|work> fsck.. forgot to add g_ether module to rootfs
Mar 06 10:03:42 <reenoo> morning
Mar 06 10:13:46 <Bernardo> good morning
Mar 06 10:14:08 <hrw|work> hi Bernardo 
Mar 06 10:15:44 <katossi_uni> what module do I need to use sd in akita?
Mar 06 10:16:16 <Bernardo> katossi_uni: it should work automatically...
Mar 06 10:16:25 <Bernardo> let me see which module is loaded on mine
Mar 06 10:16:56 <Bernardo> sharp_mmcsd_m
Mar 06 10:17:02 <katossi_uni> kernel 2.6
Mar 06 10:17:28 <Bernardo> ah, ok
Mar 06 10:17:50 <hrw|work> under 2.6 we have SD in kernel
Mar 06 10:18:01 <hrw|work> otherwise you need: mmc_core, mmc_block, pxamci
Mar 06 10:18:43 <Bernardo> hrw|work: about the opie sound on the akita, it isn't related to libxine, you were right. It just works half of the time
Mar 06 10:18:56 <hrw|work> Bernardo: under 2.4 or 2.6?
Mar 06 10:19:02 <Bernardo> 2.4
Mar 06 10:19:07 <katossi_uni> but sd is not automatically mounted, where is it in /dev?
Mar 06 10:19:20 <hrw|work> katossi_uni: vi /etc/fstab then
Mar 06 10:19:37 <hrw|work> katossi_uni: /dev/mmcblk0pX
Mar 06 10:19:51 <Bernardo> couldn't build 2.6 this weekend, had too many complaints about missing dependencies (starting with quill) and not enough time to see what was wrong
Mar 06 10:20:46 <katossi_uni> hrw|work:  only /dev/mmcda is showing up
Mar 06 10:21:00 <hrw|work> katossi_uni: you sure that you have 2.6 not 2.4?
Mar 06 10:21:06 <hrw|work> katossi_uni: udev working?
Mar 06 10:21:11 <katossi_uni> hope so
Mar 06 10:21:24 <katossi_uni> 2.6.15-rc7
Mar 06 10:21:43 <hrw|work> katossi_uni: -rc7?
Mar 06 10:22:49 <katossi_uni> it's what uname said
Mar 06 10:23:54 <katossi_uni> dmesg shows mmcblk0, so it's weird
Mar 06 10:24:23 <hrw|work> looks like udev is not running
Mar 06 10:26:08 <surinder> hi
Mar 06 10:27:04 <katossi_uni> hrw|work: I restarted it. Could it be a prblem with udev after suspend?
Mar 06 10:27:22 <hrw|work> rather not
Mar 06 10:27:41 <hrw|work> hmm.. at work my builddir is nearly 0.5 year old...
Mar 06 10:27:42 <hrw|work> 0915
Mar 06 10:41:39 <schurig> ~seen mickey
Mar 06 10:41:41 <ibot> mickey <n=mickey@deneb.tm.informatik.uni-frankfurt.de> was last seen on IRC in channel #oe, 21d 15h 44m 37s ago, saying: 'ups'.
Mar 06 10:41:47 <schurig> ~seen mickey|dinner 
Mar 06 10:41:49 <ibot> mickey|dinner is currently on #oe #opie #handhelds.org, last said: 'lunch, even'.
Mar 06 10:42:00 <hrw|work> hi Holger
Mar 06 10:42:17 <hrw|work> 06 11:46 [FreeNode] -!- mickey|dinner is idle for 17 hours, 2 minutes and 24 seconds.
**** ENDING LOGGING AT Mon Mar 06 10:59:56 2006