**** BEGIN LOGGING AT Mon Aug 01 23:59:56 2005 Aug 02 00:02:04 is there an arm devel. channel out there? Aug 02 00:02:23 my drivers keep bugging me Aug 02 00:28:02 morning Aug 02 00:32:29 hi XorA Aug 02 00:53:36 luzik: ping Aug 02 00:53:42 morning people Aug 02 00:55:34 morning hrw|work Aug 02 00:57:23 morning hrw|work Aug 02 00:57:45 bah, bitbake doesnt automatically delete pr0n to make space for code :-) Aug 02 00:57:58 XorA: write bbclass for it Aug 02 00:58:11 * hrw|work is checkouting 2.6.12-mm1 kernel Aug 02 00:58:31 hrw|work: that works apart from irda and RNDIS Aug 02 00:59:29 XorA: I need usbnet and wifi working - 2.6.13-rc3-mm1 == no wifi/hostap Aug 02 01:00:03 hrw|work: both those work with 2.6.12 Aug 02 01:01:29 I know Aug 02 01:01:36 and thats why I'm reverting Aug 02 01:02:13 * XorA is just installing gimp Aug 02 01:06:09 hrw|work: do you have any irda kit? Aug 02 01:07:33 XorA: nokia 6310i, palm m105, zaurus sl-5500 Aug 02 01:08:47 hrw|work: would be nice if you could double check irda on 2.6.12-mm1 as its perfectly possible I did it wrong Aug 02 01:09:08 XorA: if find time then can look Aug 02 01:13:50 morning all Aug 02 01:17:00 morning RP Aug 02 01:17:32 reverting to 2.6.12-mm1 ? :-/ That sounds bad. I wish I could spend more time and fix this... Aug 02 01:20:24 RP: hostap does not work with 2.6.13-rc3-mm1 Aug 02 01:20:37 btw does anyone think it would be a good idea to change module naming to debian style? that way kernels can be swapped by a quick reflash Aug 02 01:20:41 hrw|work: I know :-( Aug 02 01:20:45 RP: and I have too much work to look more into it Aug 02 01:21:10 hrw|work: You and me both... Aug 02 01:22:04 * RP was up until goodness knows what time yesterday just trying to help the kernel devs find out why -rc3-mm3 and -rc4-mm1 lockup when X starts Aug 02 01:22:27 good morning all Aug 02 01:22:45 and spent most of last night trying fixes that didn't make any difference or made things worse... Aug 02 01:22:48 morning koen Aug 02 01:23:04 we need some VC to start up an Angstrom company then some of us can work full time Aug 02 01:25:01 XorA: Its a nice idea Aug 02 01:27:59 how long will initial checkout take? Aug 02 01:29:19 hrw|work: veeeeeeeeeeery long Aug 02 01:29:27 hrw|work: use the snapshot db Aug 02 01:29:37 ok Aug 02 01:30:01 could you add url to int into MonotonePhraseBook? Aug 02 01:30:36 http://ewi546.ewi.utwente.nl/tmp/OE.db.bz2 ? Aug 02 01:30:37 It's already in GettingStarted Aug 02 01:30:40 aha Aug 02 01:30:46 that's the url Aug 02 01:30:56 if you wait 2mins I'll update it Aug 02 01:31:04 done Aug 02 01:31:04 * XorA beats mmcqd to write faster Aug 02 01:35:47 hrw|work: pong Aug 02 01:35:51 hrw|work: hi :) Aug 02 01:37:06 RP: Do you have CONFIG_MMC_BULKTRANSFER enabled? Aug 02 01:37:33 hi! Aug 02 01:38:28 luzik: where you want to mirror OZ? Aug 02 01:39:07 hrw|work: in task Aug 02 01:39:24 hrw|work: ftp://ftp.task.gda.pl Aug 02 01:39:44 luzik: official one? Aug 02 01:39:54 luzik: sf.net does not allow rsync Aug 02 01:40:09 hrw|work: :/ Aug 02 01:40:26 hrw|work: any another ideas? Aug 02 01:42:15 luzik: scp from shell.sf.net account Aug 02 01:42:29 thats the method which I use to update upgrades feed Aug 02 01:43:12 hrw|work: ok i'll ask my ftpadmins for scp Aug 02 01:43:17 hrw|work: with keys yes? Aug 02 01:43:33 hrw|work: it must be full automatic Aug 02 01:45:08 luzik: other option is talk with one of our mirrors admin - maybe someone allow rsync Aug 02 01:47:23 hrw|work: If you can login, you can rsync over ssh Aug 02 01:50:52 koen: didnt know.. Aug 02 01:51:08 hrw|work: yes .. my ftpadmins just tell me the same Aug 02 01:51:31 hrw|work: you have shell there? Aug 02 01:51:59 hmm .. but no one give us ssh keys :/ Aug 02 01:52:30 luzik: anyone can create sf.net account Aug 02 01:52:45 hrw|work: with shell? Aug 02 01:55:22 yes Aug 02 01:59:06 hrw|work: where i can find mirrors admins list? Aug 02 01:59:11 hrw|work: how are you :) Aug 02 02:03:15 03jbowler 07org.oe.nslu2-linux * r5f72a960... 10/packages/ccxstream/ (4 files in 2 dirs): ccxstream package from kolla Aug 02 02:03:19 03ccsmart 07org.oe.nslu2-linux * r10517186... 10/.mt-attrs: Committing executable flag for cyrus startup script. Aug 02 02:03:23 03rwhitby 07org.oe.nslu2-linux * r3b2d465c... 10/classes/tinderclient.bbclass: Reinstated freyther's patch which I inadvertently reverted during a spurious monotone merge :-( Aug 02 02:03:26 03kolla 07org.oe.nslu2-linux * r08db957f... 10/packages/ccxstream/ccxstream_1.0.15.bb: Fixed to use proper variables and install -d instead of mkdir Aug 02 02:03:31 03rwhitby 07org.oe.nslu2-linux * r9ec15bcf... 10/packages/meta/openslug-packages.bb: Added more development packages to OpenSlug feed Aug 02 02:03:35 03ccsmart 07org.oe.nslu2-linux * rb1ad57ce... 10/packages/cyrus-sasl/cyrus-sasl_2.1.19.bb: Make call to saslpasswd2 more conformant to documentation (adding -p). Aug 02 02:03:39 03kolla 07org.oe.nslu2-linux * rd5ca2d90... 10/packages/ccxstream/ccxstream_1.0.15.bb: Wasnt aware of PR... bear with me Aug 02 02:03:43 03jbowler 07org.oe.nslu2-linux * ra0c6e9d6... 10/classes/tinderclient.bbclass: Aug 02 02:03:43 To avoid further errors this is a manual copy of the current Aug 02 02:03:43 classes/tinderclient.bbclass (which includes the de-reverted change) from Aug 02 02:03:43 org.openembedded.dev Aug 02 02:03:47 03mickeyl 07org.oe.dev * rc05a9375... 10/packages/flex/ (4 files in 2 dirs): flex: add patch for flex bug 720976 Aug 02 02:03:51 03mickeyl 07org.oe.dev * r5ae8db07... 10/packages/ (6 files in 5 dirs): add teTeX - a complete TeX distribution for UNIX compatible systems Aug 02 02:03:55 03freyther 07org.oe.dev * rea784b81... 10/classes/tinderclient.bbclass: Manually redo: http://monotone.vanille.de/viewmtn/revision.psp?id=2c3b323f578c3e3f368933af2777b403d390143d Aug 02 02:03:59 03nail 07org.oe.nslu2-linux * r7e76b182... 10/packages/bwmon/ (bwmon_1.3.bb files/makefile.patch): Added package bwmon (Linux bandwidth monitor) Aug 02 02:04:03 03nail 07org.oe.nslu2-linux * r224d1821... 10/conf/distro/openslug-packages.conf: Add bwmon Aug 02 02:04:08 03nail 07org.oe.nslu2-linux * r918d376e... 10/packages/meta/openslug-packages.bb: Add ccxstream and bwmon Aug 02 02:04:12 03nail 07org.oe.nslu2-linux * r9ef371d5... 10/conf/distro/openslug-packages.conf: Added ccxstream Aug 02 02:04:16 03daka 07org.oe.nslu2-linux * rd24be3dd... 10/packages/perl/ (perl-5.8.7/config.sh-i386-linux perl_5.8.7.bb): Add untested i386 config.sh for perl for OE-Bug:#196 Aug 02 02:05:18 good morning mickey|TV Aug 02 02:08:53 ok i found mirrors list Aug 02 02:10:37 hrw|work: I listed the monotone backupservers to in GettingStarted Aug 02 02:12:37 ok Aug 02 02:21:19 do13: No, that would probably help wouldn't it :) Aug 02 02:22:02 do13: I also can't get the SD code from 2.6.13-rc4-mm1 to work... Aug 02 02:23:38 RP: This option should boost the write speed Aug 02 02:24:11 RP: What's the problem with 2.6.13-rc4-mm1 SD code? Aug 02 02:25:31 do13: I don't know. It doesn't seem to show my SD card though. I need to investigate... Aug 02 02:43:53 Hi, how can I make sure the glibc headers is installed properly? Aug 02 02:44:03 in fc3 Aug 02 02:46:17 I got glibc-devel-2.3.5-0.fc3.1, does this mean I am okay with the glibc headers? Aug 02 03:25:56 03nslu2-linux 07org.oe.nslu2-linux * r846ca47e... 10/conf/distro/ucslugc-packages.conf: Added bwmon to ucslugc-packages.conf Aug 02 03:26:00 03nslu2-linux 07org.oe.nslu2-linux * r0aca9d0e... 10/conf/distro/ucslugc-packages.conf: Added ccxstream to ucslugc-packagaes.conf Aug 02 04:13:55 hey Aug 02 04:14:00 rwhitby-away: hey ;) Aug 02 04:29:45 rwhitby-away: At least I now know I actually pushed the change Aug 02 04:34:06 03xora 07org.oe.dev * rd16bc244... 10/packages/gqview/ (gqview-2.1.1/include-path-fix.patch gqview_2.1.1.bb): gqview: added version 2.1.1 of this graphics viewer. Aug 02 04:42:16 03rw 07org.oe.dev * r8893b764... 10/packages/gtk-webcore/osb-jscore_20050430.bb: osb-jscore: zap CPPFLAGS. fixes bug #198. thanks to pb_ for the hint. Aug 02 04:43:33 hail zecke Aug 02 04:43:43 zecke_, too Aug 02 04:44:47 pb_: fear the power of clone(2) Aug 02 04:45:29 the world thinks we're two ;) but we share the same memory and genetic program Aug 02 04:45:31 seems more like vfork() in effect Aug 02 04:45:39 zecke is paused while zecke_ runs Aug 02 04:45:55 pb_: the matrix has one processor per process ;) Aug 02 04:46:01 heh Aug 02 04:46:17 pb_: they make you believe zecke is currently paused Aug 02 04:46:29 oh dear Aug 02 04:46:32 I guess the matrix has me. Aug 02 04:51:06 ibot, botmail for reenoo: regarding bug 198 again, I wonder if we should just have autotools.bbclass stomp on CPPFLAGS as standard. Aug 02 05:02:55 pb_: would you happen to know anything about a company called Sangoma? Aug 02 05:07:55 Twiun: the telephony people? I've heard of them, but that's about as far as it goes. Aug 02 05:08:30 pb_: more than I had ;) Aug 02 05:08:37 pb_: well... as of the weekend anyway Aug 02 05:09:13 pb_: I'll be buying their entry-level adsl pci card ;) Aug 02 05:09:30 ah, right, cool Aug 02 05:09:32 pb_: their linux support looks to be very good Aug 02 05:09:49 * Twiun compares entry-level with the 115GBP price tag and sighs. Aug 02 05:09:53 yeah, they're pretty well established in the WAN/ATM/DSL market. There are linux drivers for most of their boards. Aug 02 05:10:55 they are indeed not a budget solution, though. You can buy a cheap DSL PCI card for about 10GBP these days. Aug 02 05:11:30 I imagine the linux driver situation for those is a bit sketchy though. Aug 02 05:12:56 yeah, I'm also looking at durability - I've gone through 2 routers in 2 years Aug 02 05:13:02 I'm getting tired of off-the-shelf shite Aug 02 05:20:14 right Aug 02 05:20:35 we're using d-link dsl routers here at the office, which seem to be more-or-less ok. Aug 02 05:20:45 My parents use one of the PCI ADSL cards with Linux. Aug 02 05:20:54 There's a big binary blob, but by and large it seems to work. Aug 02 05:21:41 I've even uncovered a linux distro specifically for the job and for that card - for ADSL bonding ;) Aug 02 05:21:41 When we first started using these routers they needed rebooting about once a week, but that seems to have gotten better now. I don't think I've restarted either of them for a couple of months. Aug 02 05:22:06 I use a USB Speedtouch. None of the off the shelf affordable routers do native IPv6. Aug 02 05:22:09 03florian 07org.oe.dev * r0eaad42e... 10/packages/gpe-dm/ (files/Xserver-spitz.patch gpe-dm_0.47.bb): Add patch to support Sharp Spitz. Aug 02 05:22:26 yeah, that sucks, but our ISP doesn't do native IPv6 either. Aug 02 05:22:31 Mine does. ;) Aug 02 05:22:40 I guess I should write to them and complain or something. Aug 02 05:22:41 Twiun: Does the card you're looking at have open source drivers? Aug 02 05:29:10 BigAl: yeah, as a set of patches Aug 02 05:29:49 BigAl: Oh, and also the raw source Aug 02 05:29:59 BigAl: straight off their ftp server (and not just for linux) Aug 02 05:31:15 Twiun: Cool.. I've been looking for an ADSL card with open source drivers for a while.. Binary only drivers tend not to work on a Sparc. Aug 02 05:32:07 hmm.. I think I should start to look for ipv6 support from my isp too Aug 02 05:32:21 BigAl: the speedtouch has open source drivers, it's only the firmware that is binary Aug 02 05:34:49 Bernardo|meeting: yeah, same with the sangoma, the firmware comes in the tarball but also looks like a blob Aug 02 05:35:15 5mb tarball... Aug 02 05:35:46 Hopefully one of those will work in australia. Aug 02 05:38:00 BigAl: maybe get the sources and see what they look like before shelling out too much on the cards? Aug 02 05:39:23 BigAl: ftp://ftp.sangoma.com/Solaris/ Aug 02 05:39:33 BigAl: dunno about the speedtouch ones though Aug 02 05:40:03 The Speedtouch driver is in the stock kernel. Aug 02 05:42:07 Twiun: Thanks, I'll check it out. Aug 02 05:42:44 BigAl: hopefully I'll be able to test mine out tomorrow and let you know how it goes Aug 02 05:42:46 * XorA used the userspace speedtouch driver for years without a hitch Aug 02 05:48:37 hi mickeyl Aug 02 05:50:12 morning pb_ Aug 02 05:53:43 hi mickeyl Aug 02 05:57:02 03nail 07org.oe.nslu2-linux * re90da84b... 10/packages/openntpd/openntpd_3.7p1.bb: Removed unneccessary creation of /shared/empty, added timezones DEPENDS Aug 02 05:57:06 03nail 07org.oe.nslu2-linux * rbceb17b4... 10/packages/bwmon/bwmon_1.3.bb: Make it use , and actually create the dir before installing. (duh) Aug 02 06:04:43 XorA: I had to switch to the kernel drivers one month ago, after one year using them I couldn't get the userspace drivers to work one day Aug 02 06:05:27 Bernardo: :-( Aug 02 06:05:27 I still can't get gcc-cross-initial to load the source needed from savannah Aug 02 06:05:41 whst is this CVS_TARBALL_STASH way? Aug 02 06:05:46 Bernardo: now I have the newer Speedtouch that ive never taken out of box and an X-Modem which I use Aug 02 06:07:17 CVS_TARBALL_STASH is a SRC_URI mirror that - when enabled - tries to download things from a mirror instead of fetching from cvs Aug 02 06:08:01 hmm Aug 02 06:08:15 So I would need to know where to find the offending file Aug 02 06:08:21 * Crofton|laptop hates savannah ... Aug 02 06:09:07 or shoudl I just prefer a different version maybe Aug 02 06:44:05 03mickeyl 07org.oe.dev * r3638859f... 10/packages/tetex/tetex-native_3.0.bb: tetex-native: fix typo Aug 02 07:06:47 03nail 07org.oe.nslu2-linux * rc18bb85d... 10/conf/distro/ucslugc-packages.conf: Add timezones Aug 02 07:09:06 hey kergoth Aug 02 07:09:11 hey Aug 02 07:09:19 ~mornings Aug 02 07:09:20 Mornings MUST be destroyed! (see also http://www.destroymornings.com/) Aug 02 07:09:25 ~botsnack Aug 02 07:09:26 aw, gee, kergoth Aug 02 07:09:35 hey kergoth Aug 02 07:09:53 it feels like a monday Aug 02 07:09:56 ~mondays Aug 02 07:09:56 Sounds like _somebody_'s got a case of the Muhhndays! Aug 02 07:10:06 * koen gets some more coffee Aug 02 07:11:25 hi Aug 02 07:11:44 hey CoreDump|home Aug 02 07:15:25 hi CoreDump|home Aug 02 07:16:00 hey koen & hrw|busy Aug 02 07:28:21 afternoon Aug 02 07:41:33 I am trying to compile OE for x86...got the following error: Aug 02 07:41:37 NOTE: multiple providers are available (glibc, glibc-intermediate); Aug 02 07:41:44 NOTE: consider defining PREFERRED_PROVIDER_virtual/i386-linux-libc-for-gcc Aug 02 07:41:53 NOTE: package glibc-2.3.5+cvs20050627: started Aug 02 07:42:00 NOTE: package glibc-2.3.5+cvs20050627-r0: task do_fetch: started Aug 02 07:42:06 ~pastebin Aug 02 07:42:06 well, pastebin is a place to paste your stuff without flooding the channel - try http://pastebin.ca Aug 02 07:42:09 NOTE: package glibc-2.3.5+cvs20050627-r0: task do_fetch: completed Aug 02 07:42:16 startoe: use that for logs Aug 02 07:42:22 NOTE: package glibc-2.3.5+cvs20050627-r0: task do_compile: started Aug 02 07:42:31 ERROR: function do_compile failed Aug 02 07:42:55 okay sorry Aug 02 07:45:50 http://pastebin.ca/raw/19087 is my error. Aug 02 07:47:37 03nail 07org.oe.nslu2-linux * r4b5d4afc... 10/packages/ccxstream/ccxstream_1.0.15.bb: Make ccxstream depend on readline Aug 02 07:59:34 anyone tried building openobex-apps lately? Aug 02 07:59:47 it compiles, but no ipkg is created Aug 02 08:09:23 anyone can kindly post his/her conf/local.conf file for me to have a look? Aug 02 08:19:13 sox: User options overriding size read in .wav header Aug 02 08:19:13 sox: bad input format for file /opt/QtPalmtop/sounds/keysound.wav: data size 254 Aug 02 08:19:26 any idea what this is about? Aug 02 08:19:54 don't do that Aug 02 08:20:48 startoe: http://hentges.net/tmp/local.conf Aug 02 08:21:33 2 lines is hardly a flood Aug 02 08:23:18 koobla: Packaged contents of openobex-apps into ../openobex-apps_1.0.0-r0_armv5te.ipk Aug 02 08:24:27 CoreDump|home: thanks Aug 02 08:24:36 np Aug 02 08:30:15 CoreDump|home: can I ask what does this line ASSUME_PROVIDED = "virtual/arm-linux-gcc-2.95" do for you? Aug 02 08:30:38 need that for collie Aug 02 08:31:06 and any zaurus 2.4 kernel IIRC Aug 02 08:31:09 if I compile for x86 I don't need this line , right? Aug 02 08:31:23 i'd think so Aug 02 08:31:38 cheers Aug 02 08:32:41 wasn't there a sound format you could cat into /dev/dsp and it would sound right? Aug 02 08:33:54 CoreDump|home: wav, depending on the drivers and settings of the dsp Aug 02 08:35:10 ah, no luck w/ wav here :\ There must be a small shell app to play simple .wav Aug 02 08:35:28 CoreDump|home: cat /dev/urandom >/dev/dsp Aug 02 08:35:43 hrw|busy: heh nice try :) Aug 02 08:36:06 I remember times when 'cat /dev/urandom >/dev/tty1' was often used.. Aug 02 08:36:14 I usually use /boot/bzImage to torture someone Aug 02 08:39:23 CoreDump|home: if it is sharp zImage then it is torture++ Aug 02 08:39:33 ;) Aug 02 08:48:00 03kolla 07org.oe.nslu2-linux * r1aebe30c... 10/packages/pine/ (12 files in 2 dirs): Aug 02 08:48:01 Pine - A tool for reading, sending and managing electronic messages. Aug 02 08:48:01 With chappa-patches, transparent and more. LDAP- and SSL-support. Aug 02 08:48:44 that's a bit scary Aug 02 08:49:14 OE is losing its innocence Aug 02 08:49:55 yeah, it seems Aug 02 08:50:31 I guess we should add some kind of DISALLOWED_LICENSES variable that you can set in local.conf. Aug 02 08:51:56 have base.bbclass throw SkipPackage if LICENSE matches that regexp, or something. Aug 02 08:53:20 good idea Aug 02 08:53:22 cu Aug 02 08:57:08 is there a OE-howto guide? Aug 02 08:58:54 only the GettingStarted page on the wiki Aug 02 09:00:42 The getting started page is good. It helped me a lot to get started. But once I hit problem don't know where to get more information. Aug 02 09:08:03 so, what does monotone want to tell me by opening two temp files in vimdiff? I mean, it doesn't care whether I save either of those... Aug 02 09:09:52 it has a conflict and wants you to merge it by hand Aug 02 09:10:05 * koen uses meld for that Aug 02 09:10:13 that's nice and all, but how? Aug 02 09:10:20 meld doesn't work here Aug 02 09:10:56 all I can say is that 'apt-get install meld' gives me a way to visually merge stuff Aug 02 09:11:06 albeit slow (X over cable)r Aug 02 09:11:06 ~stab koen Aug 02 09:11:07 * ibot runs at koen with an origami Swiss Army knife, and inflicts a nasty paper cut. Aug 02 09:11:28 this is still a fscking broken gentoo box Aug 02 09:11:50 ~lart gentoo Aug 02 09:11:51 * ibot cats /dev/urandom into gentoo's ear Aug 02 09:12:58 koen: how do you merge stuff in meld? merge and then save the result where/how? Aug 02 09:13:38 monotone says "merge to LEFT" or "merge to CENTER" Aug 02 09:14:13 so I click the arrows and check if LEFT or CENTER is correct and click 'save' Aug 02 09:15:03 hmm Aug 02 09:15:08 monotone: help required for 2-way merge Aug 02 09:15:08 monotone: [ left] conf/distro/ucslugc-packages.conf Aug 02 09:15:08 monotone: [ right] conf/distro/ucslugc-packages.conf Aug 02 09:15:08 monotone: [ merged] conf/distro/ucslugc-packages.conf Aug 02 09:15:08 executing external 2-way merge command Aug 02 09:15:09 2 files to edit Aug 02 09:15:47 another way is to pick an ancestor by hand using explicit_merge r1 r2 ancestor Aug 02 09:15:52 I take it it wants me to save to conf/distro/ucslugc-packages.conf? or what? Aug 02 09:16:29 in meld you can save to the temp file Aug 02 09:16:33 doubtful. last time I cancelled the merge at this stage I was left with a broken db Aug 02 09:16:45 afaik monotone merges temp files to stop it from messing up the local copy Aug 02 09:22:01 yup. it screwed itself over again Aug 02 09:23:29 I'm fine with merging stuff by hand (even though there are no conflicts whatsoever). but it should tell me where it wants the result Aug 02 09:49:42 03florian 07org.oe.dev * re352ade7... 10/conf/machine/nokia770.conf: Install gpe-mini-browser in gpe images, some minor bugfixes. Aug 02 09:59:46 mickeyl: blech....pine..... Aug 02 10:30:39 grrr, I can't build glibc for omap5912osk machine ... Aug 02 10:31:11 http://pastebin.ca/19109 Aug 02 10:31:53 03rw 07org.oe.dev * r4f4974b1... 10/packages/gpe-life/ (gpe-life_0.1.bb gpe-life_0.2.bb): gpe-life: version bump to 0.2. Aug 02 10:33:24 Crofton: looks like do_fetch failed Aug 02 10:34:19 finally Aug 02 10:35:44 hmm, so like I have a partial copy of glibc? That might make sense Aug 02 10:42:20 koen: fwiw, the file you have to save in vimdiff is the leftmost one Aug 02 10:42:31 ok Aug 02 10:42:59 koen: the problem is just that you have to edit it before vim will actually save it (even with :w!) Aug 02 10:43:36 ah, it's trying to be smart.... Aug 02 10:44:23 yeah, vim and monotone "outsmart" each other there Aug 02 10:44:38 03nail 07org.oe.nslu2-linux * r3187ed0a... 10/packages/openntpd/openntpd_3.7p1.bb: Change ownership of /var/shared/empty, so that openntpd doesn't freak out Aug 02 10:44:42 03nail 07org.oe.nslu2-linux * r89feede1... 10/packages/openntpd/ (files/init openntpd_3.7p1.bb): Updated initscript to use -s to work around adjtime() bug Aug 02 10:45:08 koen: IMHO, monotone should just accept the file. modified or not. Aug 02 10:45:29 yeah Aug 02 10:46:07 I was thinking about using lua to create some restrictions Aug 02 10:46:20 like disallowing divergence Aug 02 10:47:10 hmm Aug 02 10:47:52 what worries me a bit lately is that there appear to be two largely idependent "sub-graphs" that I have to merge all the time Aug 02 10:48:25 how do I find the nice graph I've seen with viewmtn? Aug 02 10:48:54 I see you people paste a link to the current head once in a while, but... how do you get there? ;) Aug 02 10:49:15 http://monotone.vanille.de/ Aug 02 10:50:00 viewmtn/headofbranch.psp?branch=org.openembedded.dev will give you the latest head Aug 02 10:51:02 oh... that was.. complex ;) Aug 02 10:51:37 thanks Aug 02 10:52:39 someone should tell that kolla guy his pine.bb leads to a package that violates the pine license btw Aug 02 10:52:55 reenoo_: it does? Aug 02 10:53:05 no shit Aug 02 10:53:07 it does Aug 02 10:54:00 reenoo_: the package doesn't violate the pine license, redisributing the package does Aug 02 10:54:17 but yes, that should be noted with big letters somewhere Aug 02 10:54:38 stuff like that should *never ever* make it into the OE repo Aug 02 10:54:50 Maybe he should send a mail to them... " Other licensing terms are available by mutual agreement." <-- ;) Aug 02 10:55:07 guys. this isn't funny Aug 02 10:56:18 17:50 < pb_> I guess we should add some kind of DISALLOWED_LICENSES variable that you can set in local.conf. Aug 02 10:56:21 17:51 < pb_> have base.bbclass throw SkipPackage if LICENSE matches that regexp, or something. Aug 02 10:56:23 reenoo_: disapprove his pine commit, then Aug 02 10:56:40 I don't think its funny, and if OE doesn't want it in the feed it should be removed Aug 02 10:56:42 And didn't mickeyl just commit a pine bb as well? Aug 02 10:57:02 pine isn't worth the trouble anyway Aug 02 10:57:05 mutt is just better Aug 02 10:58:55 JustinP: looking at the merging chaos I don't think a disapprove will work Aug 02 10:59:01 true.... Aug 02 10:59:21 well just remove it and commit that then Aug 02 10:59:41 and yes, I agree that divergence should nto be allowed. If we can fix that server-side that woul dbe great Aug 02 10:59:55 or set DEFAULT_PREFERENCE="-1" and hang on a bit Aug 02 10:59:57 force people to pull, merge, push/sync Aug 02 11:01:37 I wished that would do it (as a script) Aug 02 11:01:56 NAiL: I seem to remeber an EXCLUDE_FROM_WORLD variable Aug 02 11:02:22 pull/merge is so slow that a subsequent push can lead to two heads as well Aug 02 11:02:40 XorA|gone: I guess you're referring to BROKEN Aug 02 11:03:09 true Aug 02 11:03:19 mark it as BROKEN. Whatever will make it not compile until licencing issues are worked out.. Aug 02 11:05:43 actually compiling it is ok....as long as you don't distribute the binary Aug 02 11:05:49 (right?) Aug 02 11:05:52 yep Aug 02 11:05:54 darn Pine license Aug 02 11:06:10 moron license Aug 02 11:06:12 well. for distributors that nitpicking doesn't work Aug 02 11:06:43 you can't check 2000+ packages+patches Aug 02 11:07:11 LICENSE has to reflect the state of a .bb Aug 02 11:07:34 (that's why it was added after all) Aug 02 11:09:00 couldn't the bb check for some variable and quit with a message if its not set, stating that if you want pine you have to set that variable and that you are not allowed to redistribute the ipk? not that I actually care about pine.. Aug 02 11:12:32 hey guys Aug 02 11:13:08 can anyone tell me if the Intel Xscale PXA225 is supported by oe? Aug 02 11:15:28 oe doesn't support processors Aug 02 11:15:45 it supports boards and devices Aug 02 11:15:50 I think there are machine configurations in OE which compile for that processor but I'm not sure Aug 02 11:16:41 225? does such a beast even exist? google only finds some 200 hits, wouldn't that rather be 255? Aug 02 11:17:11 hmmm Aug 02 11:17:20 drupal stink Aug 02 11:17:30 any suggestions for a decent CMS? Aug 02 11:18:38 DaKa2: I guess you could just as well package the whole thing as a separate package that complies with the pine license Aug 02 11:19:13 http://pastebin.ca/19112 Aug 02 11:19:20 looks like fecth did succedd Aug 02 11:19:28 DaKa2: i.e. changelog in the docs or something and an L in the version string Aug 02 11:21:35 reenoo_: yes, that actually has to be done, forgot about that, but even if you make local changes you can not redistribute the binary Aug 02 11:22:22 03rw 07org.oe.dev * r2653a4b4... 10/packages/pine/ (12 files in 2 dirs): disapproval of revision 1aebe30c46f7b39a3a73dcd30e1227c1a403f1bd Aug 02 11:22:26 03rw 07org.oe.dev * r4e0dabf4... 10/packages/pine/pine_4.63.bb: pine: re-add Mickey's version of the pine .bb to avoid licensing issues. (somehow monotone disapprove decided to remove it altogether.) Aug 02 11:24:33 DaKa2: ah, ok. I missed that only local modifications are allowed Aug 02 11:24:57 anyways. surprisingly the disapprove worked (sort of). Aug 02 11:26:32 the pine license is a bit retarded, but since they made it they can set what license they want.. and they did :( Aug 02 11:38:45 03nail 07org.oe.nslu2-linux * r9c9506ff... 10/packages/cron/cron_3.0pl1.bb: Added update-rc.d in postinst Aug 02 11:38:49 03nail 07org.oe.nslu2-linux * reb290743... 10/packages/thttpd/thttpd_2.25b.bb: Added update-rc.d in postinst Aug 02 11:38:53 03nail 07org.oe.nslu2-linux * re300bd8e... 10/packages/openntpd/openntpd_3.7p1.bb: Added update-rc.d in postinst Aug 02 11:38:57 03nail 07org.oe.nslu2-linux * r1d5f3ae4... 10/packages/samba/samba_3.0.14a.bb: Added update-rc.d in postinst Aug 02 11:40:15 thanxs, yeah i did mean 255 Aug 02 11:41:22 also, when i try and sync i get this error monotone: read from fd 4 (peer monotone.vanille.de) failed, disconnecting Aug 02 11:42:53 you can only sync if you have write permission Aug 02 11:44:08 ahhh Aug 02 11:44:25 koen how do i go about getting my changes upstream? Aug 02 11:45:33 attach them to the bugzilla Aug 02 11:46:08 kk Aug 02 11:53:36 03nail 07org.oe.nslu2-linux * r8d70b750... 10/packages/ (4 files in 4 dirs): Use update-rc.d bbclass instead of manually doing update-rc.d Aug 02 11:53:40 03nail 07org.oe.nslu2-linux * re7146baa... 10/packages/ (4 files in 4 dirs): And remember, PR is everything (Or something) Aug 02 12:47:37 Hi, we are still having weird X coredumps after calibrating the touschscreen and login-in from the gpe-login app with the ipaq h6340. To eliminate the possible sources of problems, we would like to build the gpe-image by using existing gcc 3.3.2 based toolchain from the TI's omap webpage. I have added line: ASSUME_PROVIDED = "virtual/arm-linux-gcc virtual/libc" to local.conf? Do I need to take anything out like lines PREFERRED_PROVIDERS += " virtual/${ Aug 02 12:48:11 re Aug 02 12:49:49 hey zecke_ & pb_ Aug 02 12:49:55 hi koen Aug 02 12:51:02 no kidding did someone undo my tinderclient.bbclass changes again? Aug 02 12:51:54 zecke_: http://ewi546.ewi.utwente.nl/tmp/viewmtn/getfile.py?id=6f537dab213f85b8e27126e37a027ee6f81bf3c6&path=classes/tinderclient.bbclass Aug 02 12:51:59 that's what is in the repo Aug 02 12:52:38 koen: anyway we've negative times in the tinderbox again Aug 02 12:53:07 Our builds are so fast, they change the direction of time Aug 02 12:54:45 I'm really getting pissed off... Aug 02 12:54:48 messing with the time-space continuum sounds dangerous Aug 02 12:55:06 angry zecke_ sounds even more dangerous Aug 02 12:56:39 check the diffs on the commit list....although it doesn't show merges so it's not a full picture Aug 02 12:57:10 we could convince mickeyl to include merges too Aug 02 12:58:08 http://pastebin.ca/19122 Aug 02 12:58:26 cannot compute suffix while building glibc Aug 02 12:59:15 is there a guide on how to set up your build for tinderbox? Aug 02 12:59:51 Crofton: closes is looking into conf/tinder.conf Aug 02 12:59:52 TZ="Europe/Berlin"; USER="zecke"; INHERIT="tinderclient" Aug 02 12:59:54 something like that Aug 02 13:00:12 in local.conf? Aug 02 13:00:23 koen: I think that would be a good idea, although they should have a special subject prefix so that they can be easily filtered. It will add extra traffic but IMHO it's worth it as merges are increasingly causing problems. Aug 02 13:00:33 hehe yeah TINDER_TZ either Europe/Berlin or offset in minutes Aug 02 13:00:52 koen: are you currently building? Aug 02 13:01:24 zecke_: not that I know off Aug 02 13:01:31 I aborted the build a few hours ago Aug 02 13:01:49 I see tinder.conf Aug 02 13:01:56 ah so it is possible nobody builds? and the tinderbox isn't broken?! Aug 02 13:02:52 So how do I add it to my build process? Aug 02 13:03:07 zecke_: correct Aug 02 13:04:00 Ciao all Aug 02 13:05:46 hey Pigi Aug 02 13:06:01 hi koen| Aug 02 13:06:07 s/|/! Aug 02 13:07:19 grr, can someone jump start me with tinderbox now? Aug 02 13:07:27 hey Pigi Aug 02 13:08:22 zecke_, do you had some time to test the patch ? Aug 02 13:15:30 Pigi: no Aug 02 13:15:38 ok, np. Aug 02 13:15:41 Pigi: I'm not feeling too good now ;) Aug 02 13:16:43 I'm sorry about that. Nothing important I hope Aug 02 13:16:56 zecke_: do you want me to start a new tinderbuild? Aug 02 13:18:06 koen: I did ;) Aug 02 13:18:09 *works* relief Aug 02 13:18:14 congrats Aug 02 13:18:22 bbl Aug 02 13:18:23 unbelievable how your guts control your mood Aug 02 13:18:29 heh Aug 02 13:18:37 zecke_: http://ewi546.ewi.utwente.nl/drupal/ Aug 02 13:20:22 one more time, where do I start adding the lines to do the tinerbox thing? Aug 02 13:20:37 local.conf Aug 02 13:21:02 copy tinder.conf, or is there a way to include it? Aug 02 13:21:19 or copy and include :) Aug 02 13:21:27 I have this in local.conf: http://pastebin.com/327810 Aug 02 13:21:39 thanks Aug 02 13:21:59 koen: Not too bad ;) Aug 02 13:22:19 koen: PR and documentation is needed for *ström* to be successfull Aug 02 13:22:29 yep Aug 02 13:22:43 after an hour of clicking I gave up and read the drupal docs Aug 02 13:22:54 after that it was straight forward Aug 02 13:23:33 * zecke_ goes to oesf.org and writes about his horrible ängström experience Aug 02 13:23:41 :) Aug 02 13:24:13 ? Aug 02 13:24:31 ! Aug 02 13:24:40 koen: I think Ewi could host some resources for '*ström' QA as well Aug 02 13:25:01 I think so too Aug 02 13:25:16 koen: Do you've tomcat installed on ewi? Aug 02 13:25:17 but my boss will only think that after I get cvm + rxtx working Aug 02 13:25:40 what is rxtx? Aug 02 13:25:51 native serialstack for java Aug 02 13:26:04 http://rxtx.org/ Aug 02 13:26:57 koen: now where is the difficuilty? Aug 02 13:27:38 it needs to be crosscompiled (again a java classpath) and packaged neatly Aug 02 13:27:52 probably ~3 hours of work Aug 02 13:28:13 that's on my TODO for this week Aug 02 13:28:27 next week I'll try to hookup an ipaq to the server Aug 02 13:30:48 OK, so as I understand it, at some point my build will appear on the tinderbox web page? Aug 02 13:31:06 possibly after it has finished the cache stuff? Aug 02 13:31:37 if you have the VERBOSE part it will send info every task Aug 02 13:31:48 if not it will send it at the end of the complete build Aug 02 13:31:54 (correct me if I'm wrong) Aug 02 13:32:07 aaargh Aug 02 13:32:11 error 500 Aug 02 13:32:27 koen|tv: " it currently just uses the standard "/dev/cua?" names" Aug 02 13:32:39 koen|tv: rxtx runs on FreeBS as well? Aug 02 13:33:09 why not? Aug 02 13:33:33 I doubt it that does more than reading from/writing to a character device Aug 02 13:34:27 do you care which I use? Aug 02 13:34:46 reenoo_: yeah I would have expected /dev/ttyS0 instead of cua though Aug 02 13:34:52 zecke_: 500 Internal Server Error Aug 02 13:35:02 koen|tv: that is okay ;) Aug 02 13:35:07 zecke_: I think the serverside has some probs Aug 02 13:35:22 koen|tv: it simply does not print anything to stdout Aug 02 13:35:37 zecke_: on linux, yeah. on the BSDs cua was standard last time I checked Aug 02 13:36:03 reenoo_: right (on my 5.4 it is) Aug 02 13:36:38 TONDER_TZ is the timezone of the *server* or my my local box? Aug 02 13:36:41 TINDER_TZ Aug 02 13:37:01 JustinP: your offset from UTC in minutes Aug 02 13:37:15 JustinP: no sorry Aug 02 13:37:24 JustinP: offset of the server from UTC Aug 02 13:38:13 hmmm, should I have left it alone then? Aug 02 13:39:43 Crofton: no set it to Europe/Berlin or +0120 Aug 02 13:41:40 ugh...so not *my* building machine, it's the TINDER server? Aug 02 13:41:52 so this should never be touched by anyone unless we set up our own tinder server? Aug 02 13:42:16 hmm, I just got some internal server errors Aug 02 13:42:31 re Aug 02 13:45:45 wb Aug 02 13:45:52 * JustinP starts a supposed tinderbox build Aug 02 13:46:30 Crofton: what version of bitbake do you've? Aug 02 13:46:39 Crofton: you might need to set TINDER_START Aug 02 13:47:07 I updated from svn this week Aug 02 13:47:17 yay, it broke Aug 02 13:47:18 How do i set TINDER_START? Aug 02 13:47:59 it isn't in tinder.conf Aug 02 13:49:35 nor tinder.bbclass Aug 02 13:50:29 * JustinP sets it to "1" and starts a build Aug 02 13:52:04 zecke_: It wasn't me this time :-) Aug 02 13:52:42 hmmm, just saw updates to bitbake with svn update ........... Aug 02 13:54:59 Crofton: update bitbake if possible Aug 02 13:55:04 rwhitby-asleep: don't worry Aug 02 13:55:18 zecke_, just did Aug 02 13:55:20 my stomach/coltis is givng me a had time though Aug 02 13:55:31 mithro: how is Stockholm? Aug 02 13:56:13 03mickeyl 07org.oe.dev * r00837c91... 10/packages/ (14 files in 8 dirs): (log message trimmed) Aug 02 13:56:13 xfce changes: Aug 02 13:56:13 - remove settings out of xfce.bbclass that affect only one package Aug 02 13:56:13 (bbclasses are destined to express similarities in packages, not for easy hacking) Aug 02 13:56:13 - remove 4.2.0 versions, since 4.2.2 versions build fine Aug 02 13:56:14 - unmark libxfce4util as BROKEN Aug 02 13:56:16 - install libxfce* header files into STAGING_INCDIR/, no need to extragroup Aug 02 13:56:40 mickeyl: *hust* Aug 02 13:57:00 pb_: is receiva coming to the IFA? Aug 02 13:57:22 * Pigi needs to sleep Aug 02 13:57:24 nite all Aug 02 13:57:32 pb_: digital radio is a big topic there Aug 02 13:57:46 IFA? Aug 02 13:57:50 koen@dominion:/data/build/oe/monotone/org.openembedded.dev$ grep BROKEN packages/*/*.bb -rn | wc -l Aug 02 13:57:53 38 Aug 02 13:58:10 Crofton: "Internationale Funk Ausstellung" Aug 02 13:58:20 international radio exhibition Aug 02 13:58:21 CosmicPenguin: internation 'radio' fair Aug 02 13:58:22 that's almost 4% of the packages Aug 02 13:58:29 too bad Aug 02 13:58:39 koen|tv: we should utilize monotone to flag broken packages Aug 02 13:59:25 okay good nite Aug 02 13:59:31 g'night zecke Aug 02 13:59:39 'night zecke_ Aug 02 14:01:14 still getting errors from the webserver ... Aug 02 14:01:45 same here Aug 02 14:01:52 ah, but openomap is there now Aug 02 14:03:19 koen|ewi, errors from the tinderbox web server? Aug 02 14:08:22 Crofton: yep Aug 02 14:08:37 I ok, I got it to start building but I get all sorts of Internal Server Errors and then an exception in bb Aug 02 14:08:39 wel then I will just hope it gets better :) Aug 02 14:17:28 Error when trying to bitbake for a tinderbox set up to e-mail: http://pastebin.com/327883 Aug 02 14:17:59 mickeyl: any reason why our hostap-modules does not allow firmware download? Aug 02 14:17:59 re Aug 02 14:18:30 s/firmware download/firmware upload to non-volatile/ Aug 02 14:18:37 yes, because it's a) dangerous Aug 02 14:18:41 b) not even well tested on x86 Aug 02 14:18:59 aha Aug 02 14:19:07 i wouldn't risk doing it on an arm Aug 02 14:19:15 people who want to do that should patch the config Aug 02 14:19:32 so they're really self-responsible for that one :) Aug 02 14:19:53 ;) Aug 02 14:19:56 is the tinderbox server broken? I'm getting all sorts of internal server errors. Aug 02 14:20:17 My local machine is also throwing a python exception after the first attempted package Aug 02 14:22:29 error 500 -> server error Aug 02 14:22:39 python -> something at your end is broken Aug 02 14:23:17 well I got a bunch of 550s, then a python exception Aug 02 14:23:25 when trying to mail I just get an exception Aug 02 14:23:37 well....seems to be working now I guess..... Aug 02 14:27:25 root@c7x0:/var/tmp# hostap_diag wlan0 Aug 02 14:27:26 Host AP driver diagnostics information for 'wlan0' Aug 02 14:27:26 NICID: id=0x800c v1.0.0 (PRISM II (2.5) PCMCIA (SST parallel flash)) Aug 02 14:27:26 PRIID: id=0x0015 v1.0.7 Aug 02 14:27:26 STAID: id=0x001f v1.3.6 (station firmware) Aug 02 14:27:33 STAID: id=0x001f v1.7.4 (station firmware) Aug 02 14:27:38 updated Aug 02 14:27:49 brave guy Aug 02 14:28:33 well, I'm not seeing any updates on the page yet....although packages *are* building. I suppose I'll let it crunch on and hope the data is actually getting stored somehow (although this seems unlikely due to the 500's...) Aug 02 14:28:52 same here JustinP Aug 02 14:29:11 that card can also have primary firmware updated but that thing is probably more dangerous then secondary... - am I right mickeyl? Aug 02 14:29:40 hrw: that's correct. If something fails while flashing the primary fw you have a nice brick CF Aug 02 14:29:50 ok Aug 02 14:29:51 JTAG! Aug 02 14:29:53 hehe Aug 02 14:29:55 yeah Aug 02 14:29:57 jtag a prism2 Aug 02 14:31:18 mickeyl: koen is ipaq user - they jtag everything ;) Aug 02 14:31:32 hehe Aug 02 14:32:00 the wonders of flashing a bootloader Aug 02 14:32:43 ok - 1.7.4 is latest prism2 firmware for that card Aug 02 14:32:54 so another flashed device... :) Aug 02 14:33:31 few pc mainboards, matrox gfx cards, cd/cdrw/dvd drives, zauruses, one cf wifi card Aug 02 14:35:43 ~lart wrt54gs or dlink dcf-660w for slow 'so called speed' Aug 02 14:35:43 * ibot strangles wrt54gs or dlink dcf-660w with a doohicky mouse cord for slow 'so called speed' Aug 02 14:35:59 aha Aug 02 14:36:17 ~lart debian for shipping wrong cronscripts Aug 02 14:36:17 * ibot whips debian with a wet and grimy noodle just because for shipping wrong cronscripts Aug 02 14:37:39 mickeyl: Do you thing "Embedded power" is too cheesy as a slogan? Aug 02 14:37:51 koen: nah, I like it! Aug 02 14:38:59 time wget http://desktop-over-wifi/10M_file == 178s Aug 02 14:39:07 hrw: which firmware ? Aug 02 14:39:59 zecke_: is that the thing in September? Aug 02 14:40:03 I think we will be there. Aug 02 14:40:27 mickeyl: 1.7.4 Aug 02 14:40:40 same over usbnet - 11s Aug 02 14:40:55 hmm Aug 02 14:40:56 oh, too slow, he went to sleep Aug 02 14:41:23 mickeyl: was 1.3.6 before Aug 02 14:42:45 pb_: september 2nd to 7th according to their website Aug 02 14:44:15 yah, that sounds about right Aug 02 14:44:40 there's also some other radio-fest happening in Hong Kong, but I think that's in late September or October. Aug 02 14:47:13 http://tuxmobil.org/linux_award.html Aug 02 14:47:23 The Winners and The Prizes Aug 02 14:47:23 OpenEmbedded : SHARP SL-6000L Linux PDA Aug 02 14:47:24 OpenZaurus : SHARP SL-C750 Linux PDA Aug 02 14:47:38 woot! Aug 02 14:47:40 is that real? Aug 02 14:47:53 cool Aug 02 14:48:05 atleast Bernardo and I nominated OE to award, I also nominated OZ Aug 02 14:48:11 pretty cool Aug 02 14:48:15 well done!! Aug 02 14:48:16 we deserve it Aug 02 14:48:17 :) Aug 02 14:48:21 thanks Aug 02 14:48:39 someone add info about it to OE homepage? Aug 02 14:48:52 good idea Aug 02 14:48:59 you or me? Aug 02 14:49:08 you - I will add into Oz Aug 02 14:49:14 k Aug 02 14:49:20 ah, very cool. Aug 02 14:49:22 congratulations :-) Aug 02 14:50:24 ~lart mysql server on sf.net Aug 02 14:50:25 * ibot whips mysql server on sf.net with a wet and grimy noodle just because Aug 02 14:50:31 ~lart mysql server on sf.net again Aug 02 14:50:32 * ibot hits mysql server on sf.net again with an anvil and laughs with a contralto voice ... Haha Ha HA Ha Aug 02 14:51:11 I hope that Angstrom wont be hosted on so slow machine as OZ is Aug 02 14:52:27 no worries Aug 02 14:53:45 a few million Euro in VC and we'll have our own compilefarm ;) Aug 02 14:55:13 heh Aug 02 14:55:14 righto Aug 02 14:55:31 howdy people Aug 02 14:55:36 hey mithro Aug 02 14:56:13 what is up Aug 02 14:56:14 * chouimat is away: Dinner Aug 02 14:56:22 * mithro is waiting for his washing to get done Aug 02 14:56:35 * koen watches http://www.internettablettalk.com/video/maemo-doom.mpg Aug 02 14:57:32 http://www.oesf.org/forums/index.php?showtopic=14202 for all which dont read OZ.org Aug 02 14:58:16 3rd place was kdepim/pi... Aug 02 14:58:35 ~emulate zautrix Aug 02 14:58:36 this is exactly what the GPL is about. grab the stuff and run away Aug 02 15:02:56 but who exactly gets the devices :P Aug 02 15:04:44 6000l has wifi or bt? Aug 02 15:05:06 wifi Aug 02 15:05:13 6000w had to have bt Aug 02 15:06:14 but not both? Aug 02 15:06:24 yep Aug 02 15:06:35 Dirk added BT to his 6000L Aug 02 15:07:19 I was looking at his work for my c700 Aug 02 15:07:41 but E60 for a chip is just too much Aug 02 15:09:20 PRIID: id=0x0015 v1.1.1 Aug 02 15:09:33 * chouimat is back. Aug 02 15:11:59 heh Aug 02 15:12:09 you really want to brick something tonite, eh? Aug 02 15:12:10 :) Aug 02 15:12:51 mickeyl: that cf card flashing is finished Aug 02 15:13:25 NICID: id=0x800c v1.0.0 (PRISM II (2.5) PCMCIA (SST parallel flash)) Aug 02 15:13:25 PRIID: id=0x0015 v1.1.1 Aug 02 15:13:25 STAID: id=0x001f v1.8.4 (station firmware) Aug 02 15:14:04 mickeyl: cannot flash newer versions so I stop on that ones Aug 02 15:15:14 hmm.. need to 'unstick' some posts on oz.org to get tuxmobil awards one on top Aug 02 15:15:42 yeah Aug 02 15:15:52 * mickeyl just finished the celebration mail to oe@hh.org Aug 02 15:15:59 ~kill sf.net Aug 02 15:16:00 * ibot shoots a hyper-charged pseudotachyon gun at sf.net Aug 02 15:16:12 good news for the next OOO newsletter Aug 02 15:16:15 :) Aug 02 15:16:23 mickeyl: give info when you send it - I will look into archive then to read it Aug 02 15:16:32 mickeyl: now someone has to device a scheme to get the devices in to the right hands Aug 02 15:16:42 koen: yeah Aug 02 15:17:42 If I were to decide I would hand the devices to people commited to doing kernel work Aug 02 15:17:48 one could sacrifice the 6k for kernel work Aug 02 15:17:52 :) Aug 02 15:17:53 because that's the area that matters most Aug 02 15:17:54 *nod* Aug 02 15:19:21 yep Aug 02 15:19:41 too bad that I'm not kernel hacker guy Aug 02 15:19:55 http://openzaurus.org/wordpress/2005/08/02/tuxmobil-gnulinux-award-2005/ for OZ.org readers Aug 02 15:20:03 * koen wonders what the scratchbox-loving maemo people will think about it (http://planet.maemo.org/) Aug 02 15:21:27 ;) Aug 02 15:23:31 hi all Aug 02 15:23:50 hi florian Aug 02 15:24:02 hmm.. secondary firmware 1.7.0+ is needed for WPA - good to know Aug 02 15:24:05 hi florian Aug 02 15:25:03 Congratulations on winning the award! Aug 02 15:26:11 florian: thx Aug 02 15:27:24 03nail 07org.oe.nslu2-linux * r8910413f... 10/packages/ (4 files in 4 dirs): Update to working update-rc.d stuff Aug 02 15:27:38 NAiL: what was updated? Aug 02 15:28:01 hrw: to actually inherit update-rc.d Aug 02 15:28:13 samba, openntpd, thttpd and cron Aug 02 15:28:22 NAiL: could you add package names into commit message? Aug 02 15:28:35 hrw: yes Aug 02 15:28:42 thx Aug 02 15:28:45 np Aug 02 15:29:47 yeah, ": ..." as commit messages makes reviewing a lot easier Aug 02 15:30:00 If I'm not mistaken Werner 'tuxmobil' Heuser is the one with the red shirt: http://dominion.kabel.utwente.nl/Gallery/albums/bordeaux-stad2004/IMG_5892.sized.jpg Aug 02 15:30:54 I have to finally get someone with pda/wifi here to trace essid when router and zaurus hide it Aug 02 15:34:19 * NAiL tries to color the whole tree pinkish-orange: http://monotone.vanille.de/viewmtn/revision.psp?id=d8c660afac38811f3c2042ab1dd35d5e148e1785&ancestry_limit=0 Aug 02 15:34:22 ;) Aug 02 15:34:37 hehe Aug 02 15:34:39 go go go Aug 02 15:38:22 mickeyl: one question about signal/noise in wifi.. Aug 02 15:38:26 Link Quality:100 Signal level:0 Noise level:156 Aug 02 15:38:39 those 100, 156 are what? Aug 02 15:38:48 hehe Aug 02 15:38:59 welcome to the wonderful word of bogus levels Aug 02 15:39:05 156dBm I guess Aug 02 15:39:10 in theory, it should be db Aug 02 15:39:24 in practice it's just arbitrary values from the firmware Aug 02 15:39:32 drivers _should_ fill in a limit filed Aug 02 15:39:33 156dB and 100dBm Aug 02 15:39:38 and then report 50/90 Aug 02 15:39:42 etc. Aug 02 15:39:46 but most don't do that Aug 02 15:40:16 i talked to mr. wireless extensions about that and he told me "forget those values for anything serious" Aug 02 15:40:57 ;) Aug 02 15:42:07 opie-networksettings gives signal 64-66%, quality 91-92% - but from Z to AP is <1m and nearly nothing between Aug 02 15:43:11 anyway wifi husky<>my ap suxx Aug 02 15:46:38 Does anybody have any idea how OE could be configured so that it would use prebuild gcc 3.3.2 and glibc instead of starting to build gross-gcc 3.4.4, linux-headers, glibc, etc... In my previous build I attempted following without success: ASSUME_PROVIDED = "virtual/arm-linux-gcc virtual/libc " Would following work better and prevent gross-gcc to be build: ASSUME_PROVIDED = "virtual/arm-linux-gcc virtual/arm-linux-gcc-3.3.2 virtual/glibc virtual/libc v Aug 02 15:46:41 you should try another card once Aug 02 15:48:27 lamikr: take a peek into conf/distro/sharprom-compatible to see what you need to assume provided Aug 02 15:48:27 n8 Aug 02 15:48:31 n8 Aug 02 15:48:33 good idea, btw. Aug 02 15:48:37 g'night folks Aug 02 15:48:39 mickeyl: I have to talk with friend with ipaq and pcmcia wifi Aug 02 15:48:42 night mickeyl Aug 02 15:49:16 'night mickeyl Aug 02 15:54:26 'night mickeyl Aug 02 15:58:41 mickeyl: Thanks, that list looked good... one of the item sharprom has in it's list is: virtual/arm-linux-gcc-2.95, is that distro really still using the 2.9.5 version of gcc? Or is that somekind of initial gcc that oe normally builds for building the final version of cross-gcc? Aug 02 15:58:58 lamikr: sharp still use 2.95 Aug 02 15:59:01 lamikr: sharp really uses 2.95.3 Aug 02 16:02:42 'night all Aug 02 16:03:00 'night all Aug 02 16:04:05 hrw and koen: Thanks. In my case, do I in addition need to specify somewhere that the gcc version I want to use is 3.3.2 instead of 3.4.4 when building gpe-image, or how does the sharp build for example endup for searching gcc-2.9.5? Aug 02 16:04:50 lamikr: decide - you want to build for sharprom or not for it? Aug 02 16:08:32 I want to build familiar-0.8.3 application set but just by using different compiler/glibc than familiar-0.8.3 does. I want to try this because we have coredump problems when we have build the gpe-image for ipaq h6340 by using gcc-3.4.4 that got's build during the image build. Aug 02 16:09:46 hrw: We are really not sure are the problems in our kernel, lcdc driver, or where but the X for example coredumps after we have logged in with the gpe-login. Aug 02 16:10:55 lamikr: set PREFERRED_VERSION_gcc-cross = "3.3.2" in conf/local.conf Aug 02 16:11:13 and unset it in familiar-0.8.3.conf (if is set) Aug 02 16:12:40 hrw: Ok and then take the ASSUME_PROVIDED list from the sharprom.conf and just change the arm-linux-gcc-2.9.5 with arm-linux-gcc-3.3.2 and keep rest of the ASSUME_PROVIDED list exactly same? Aug 02 16:13:00 no Aug 02 16:13:07 lamikr: you want to build familiar so forget about sharprom Aug 02 16:13:17 you don't want to do anything with ASSUME_PROVIDED. Aug 02 16:13:41 if you put gcc-3.3.2 in there, oe will... assume it is provided, and not try to compile it. Aug 02 16:14:14 n8 all Aug 02 16:14:21 pb_: this is what he wants Aug 02 16:14:26 pb_: he has his own toolchain Aug 02 16:14:44 pb: Put that would be good, because I would like to test by using the gcc-3.3.2 that is available in linux-omap-site. Aug 02 16:14:45 oh, I see. in that case, there is no need to do anything with PREFERRED_VERSION. Aug 02 16:14:59 PREFERRED_VERSION has no effect at all if oe is not building that package. Aug 02 16:16:05 So would following work: 1) I have /usr/local/arm/3.3.2/bin as a first entry in my path Aug 02 16:17:03 2) I add following to local-conf: Aug 02 16:17:05 ASSUME_PROVIDED += "virtual/arm-linux-gcc-3.3.2 binutils-cross gcc-cross-initial gcc-cross" Aug 02 16:17:10 ASSUME_PROVIDED += "virtual/arm-linux-libc-for-gcc virtual/arm-linux-binutils virtual/arm-linux-gcc" Aug 02 16:17:14 ASSUME_PROVIDED += "virtual/arm-linux-gcc-initial virtual/arm-linux-gcc-initial Aug 02 16:17:18 ASSUME_PROVIDED += "virtual/libc linux-libc-headers" Aug 02 16:18:08 yeah, something like that. Aug 02 16:18:27 external toolchain builds are not at all well tested, and I think you will run into some small packaging issues around libgcc, but other than that it should be ok. Aug 02 16:19:46 Ok, I will try that... How about those 5 PREFERRED_PROVIDERS lines which are in local.conf by default. Should I leave them all, or take away lines like: PREFERRED_PROVIDERS += " virtual/${TARGET_PREFIX}gcc-initial:gcc-cross-initial" Aug 02 16:20:52 it doesn't matter; you're telling it that both of those things are already provided in any case. Aug 02 16:25:38 Ok. Last thing: Do I still need to find out where the familiar-0.8.3 distro defines the usage of GCC 3.4.4 and change that to 3.3.2? (It was not in familiar-0.8.3.conf because line PREFERRED_VERSION_gcc-cross = "3.4.4" was remmed there) Aug 02 16:27:07 no, that shouldn't be necessary. Aug 02 16:27:26 familiar-0.8.3 doesn't explicitly call in gcc-3.4.4; it just happens that 3.4.4 (or 3.4.4+csl) is the default compiler version. Aug 02 16:27:30 Or can I just take the gcc version number away from the ASSUME_PROVIDED list so that oe will think that arm-linux-gcc is always provided no matter what version is required. (Ie. just putting ASSUME_PROVIDED += "virtual/arm-linux-gcc binutils-cross gcc-cross-initial gcc-cross") Aug 02 16:27:49 I see... Aug 02 16:35:59 Is the sharprom.conf btw, having error because it does not leave "any space after last line": Aug 02 16:36:01 ASSUME_PROVIDED += "virtual/a" Aug 02 16:36:06 ASSUME_PROVIDED += "virtual/b" Aug 02 16:36:10 --> "virtual/avirtual/b" instead of "virtual/a virtual/b" that should be correct. Aug 02 16:36:42 no, the += operator adds a space automatically Aug 02 16:37:34 ok, nice hidden trick. Aug 02 16:40:01 I started now the build attempt: I will see the result in the morning. Thanks again very much and good night pb_! Aug 02 16:43:22 lamikr: good night Aug 02 16:45:51 good night Aug 02 17:21:09 wonderful, looks like someone killed the opie image again: ERROR: Nothing provides libipkg Aug 02 17:41:59 Can anyone help me with monotone-viz (this is on gentoo) - I have mt 0.21 and mt-viz 0.10 and I have a build failure: File "unidiff.ml", line 181, characters 16-46: Aug 02 17:41:59 Unbound value Glib.Utf8.to_unichar_validated Aug 02 17:42:56 (ocaml is 3.08.1) - I know some people have got this thing to compile ;-) Aug 02 18:06:48 jbowler-away: yes, I know what that is Aug 02 18:07:05 it's the library it depends on, gentoo has a tool-old version Aug 02 18:07:06 03nail 07org.oe.nslu2-linux * rf17d263d... 10/packages/openssh/openssh_4.0p1.bb: Update openssh to use update-alternatives Aug 02 18:08:40 jbowler-away: http://pear.reversefold.com/oe/lablgtk2.20050701.tgz dev-ml Aug 02 18:08:43 oops ;-) Aug 02 18:09:32 jbowler-away: pear.reversefold.com/oe/lablgtk2.20050701.tgz unzip this into /usr/local/portage and add /usr/local/portage as your PORTDIR_OVERLAY (if you haven't already) Aug 02 18:13:53 03nail 07org.oe.nslu2-linux * r6ed3e28f... 10/packages/dropbear/dropbear_0.46.bb: Make dropbear use update-alternatives Aug 02 18:20:54 03nail 07org.oe.nslu2-linux * ra7be664b... 10/packages/openssh/openssh_4.0p1.bb: Make update-rc.d sneak sshd before dropbear sshd Aug 02 18:46:02 JustinP: thanks... time for another overlay... Aug 02 18:51:08 jbowler-away: np. You may want to remove your current lablgtk as well Aug 02 18:52:09 Why - if I do the overlay won't emerge automatically take it away? Aug 02 18:52:58 jbowler-away: ....ummm...I think mine's named differently :-| Aug 02 18:53:23 yeah, lablgtk2 vs lablgtk Aug 02 18:53:54 I'm not a gentoo hacker so I just did enough to get it to work Aug 02 18:54:28 Ah - when I did this to something else I copied the ebuild fila and gave it a name sufficiently 'bigger' to cause gentoo to take it in preference. Aug 02 18:54:58 jbowler-away: I tried those things but....argh...this "version" of lablgtk has a date Aug 02 18:55:16 not only that, the tarball has one date and the directory it expands to has another Aug 02 18:55:24 all sorts of annoyances.... Aug 02 18:58:19 Curious... Aug 02 18:59:52 but monotone-viz requires it Aug 02 18:59:58 the author told me ;-) Aug 02 19:00:40 Well, since nothing else I have uses lablgtk I may as well just get rid of it. Aug 02 19:01:18 I think there is some way of saying 'incompatible with' in the ebuild, so that's probably a good move (I'm assuming both ebuilds install the same files). Aug 02 19:06:48 Yes, just add !dev-ml/lablgtk to the end of the DEPENDS list Aug 02 19:21:38 Anyone know how I can run configure in a subdirectory of the work dir? Ie. in ./cfg? Aug 02 19:23:47 You can change do_configure[dirs] or you can change WORK_DIR or you can cd into it inside do_configure Aug 02 19:26:08 I just tried the last solution Aug 02 19:26:15 but for some reason that didn't work either Aug 02 19:26:31 cd correct_dir Aug 02 19:26:37 oe_runconf Aug 02 19:27:35 can I do something like do_configure(${WORKDIR}/cfg) ? Aug 02 19:28:12 The first one does that - search through the bbclass files for the syntax. Aug 02 19:28:43 Is this an inherit autotools? Aug 02 19:28:50 yes Aug 02 19:29:24 It's possible that it uses WORKDIR explicitly Aug 02 19:29:47 Any chance someone here could do a bit of small-scale email backup and/or www mirroring for a few days? Aug 02 19:30:03 NAiL: It tests for ${S}/configure Aug 02 19:30:06 seems to be using ${S} Aug 02 19:30:09 ah, yes Aug 02 19:30:35 but if I set ${S}, it screws up the patching Aug 02 19:30:54 Right. Aug 02 19:31:22 I encountered the same problem in another package... Aug 02 19:31:24 hmm.. guess I can rewrite the patch to use a path of ../../somedir/file ;-) Aug 02 19:32:39 but it's not a solution that I like.. Aug 02 19:33:06 I found another way round it. Aug 02 19:34:37 Nope, it's in db/db4, but I was able to set S because there is no patch. Aug 02 19:35:26 Oh, hang on - take a look at packages/db/db_4.3.27.bb (the do_configure task) Aug 02 19:36:16 aha Aug 02 19:36:25 NAiL: i.e. construct your own ${S}/configure which just cd's to the right place and does it. Aug 02 19:37:19 It's a bit like the Cheshire cat, it disappears while you look at it :) Aug 02 19:37:57 looks like a workable solution ;) Aug 02 19:41:17 well, got further Aug 02 19:41:17 checking whether the C compiler works... configure: error: cannot run C compiled programs. Aug 02 19:41:20 If you meant to cross compile, use `--host'. Aug 02 19:41:38 I thought that part was kinda.. handled automagically? :) Aug 02 19:47:35 ugly Aug 02 19:47:49 * NAiL gives up getting iperf to compile and goes to bed instead Aug 02 20:01:19 03nail 07org.oe.nslu2-linux * r26c7aa9a... 10/packages/procps/ (procps-3.2.5/install.patch procps_3.2.5.bb): Fix lib64 issue (SlugBug #219) Aug 02 20:30:41 well, my tinderbox stuff seems to have been submitted but it's showing all sorts of normal things as errors Aug 02 21:48:52 crap, OZ doesn't use "update-alternatives" for the hacked init Aug 02 22:59:37 hi all Aug 02 23:00:42 CoreDump|home: you still around? Aug 02 23:01:07 yep Aug 02 23:01:15 kinda busy though Aug 02 23:01:34 ok, just saw your msg about openobex Aug 02 23:02:03 have copied the stuff to my nfs root anyway, so prb. kinda resolved at the moment Aug 02 23:02:30 :) Aug 02 23:03:10 strange, but it seems like autotool-based bbs won't generate packages for me - will do a full rebuild Aug 02 23:17:20 monotone troubles - how can I cope with a failed 3-way merge? Aug 02 23:47:50 got it -- EDITOR wasn't set to emacs Aug 02 23:54:44 if only that was the answer to all build problems. "EDITOR wasn't set to emacs." Aug 02 23:58:46 that sounds like a very stupid reason to fail **** ENDING LOGGING AT Tue Aug 02 23:59:56 2005