**** BEGIN LOGGING AT Fri Sep 02 02:59:56 2005 Sep 02 03:01:48 One bug is assigned to list so list get info messages about changes/delays is that correct? Sep 02 03:02:07 Bugzilla Bug 1386 Sep 02 03:02:18 alarms not working while suspended on a zaurus 6000l Sep 02 03:30:21 argh.. Sep 02 03:30:27 34K tmp/rootfs/ Sep 02 03:30:47 gints: is a RTC actually running? Sep 02 03:31:54 hail zecke_ Sep 02 03:34:17 Master of Interactive TV ;) Sep 02 03:34:33 ~seen ruediger Sep 02 03:34:37 ruediger was last seen on IRC in channel #oe, 20h 41m 22s ago, saying: 'Hi zecke'. Sep 02 03:37:11 ~lart ipkg Sep 02 03:37:11 * ibot takes out ipkg with the trash Sep 02 03:37:16 pb_: whats up? Sep 02 03:55:19 03xora 07org.oe.dev * r0a900d63... 10/packages/gnumeric/ (gnumeric-1.5.5/remove-docs.patch gnumeric_1.5.5.bb): gnumeric: update to latest dev build 1.5.5 Sep 02 04:08:33 03koen 07org.oe.dev * r517a459e... 10/packages/libmimedir/libmimedir_cvs.bb: libmimedir_cvs: fix staging Sep 02 04:09:56 ok, who broke ipkg again .... Sep 02 04:10:19 * koen guesses pigi Sep 02 04:10:44 no, I mean who removed the patch which nslu2-linux has had to put in everytime someone updates ipkg to a new version ,... Sep 02 04:10:57 dyoung: do you have the libdir.patch somewhere handy ? Sep 02 04:11:07 Yes. I can put it someplace. Sep 02 04:11:30 packages/ipkg/files is probably a good place :-) and if we get it into ipkg.inc this time it might stick ... Sep 02 04:13:04 rwhitby: thats probably me - sorry Sep 02 04:13:04 03koen 07org.oe.dev * r21480fa4... 10/packages/ipkg/ipkg_0.99.153.bb: set default_preference=-1 since it has problems Sep 02 04:13:24 hrw|work: no problem - we can spot it immediately by now :-) Sep 02 04:13:26 I'm stuffing it into packages/ipkg/files . Sep 02 04:13:35 shouyld be pushed any moment now Sep 02 04:13:41 currently I dont know any good ipkg version btw Sep 02 04:13:43 did anyone file a bug for ipkg already? Sep 02 04:14:15 I should have done that yesterday.. Sep 02 04:14:15 Sep 02 04:14:20 03koen 07org.oe.dev * r4bd9a63f... 10/conf/distro/preferred-gpe-versions-2.7.inc: conf/distro/preferred-gpe-versions-2.7.inc: update gpe-bluetooth to 0.44 Sep 02 04:14:33 ipkg *153 doesn't listen to depends Sep 02 04:14:44 NAiL: and 152 to provides ;( Sep 02 04:15:02 yeah :( Sep 02 04:20:56 ah, dyoung: looks like it's not the old libdir bug after all - it's just the depends bug rearing it's head in the do_rootfs (cause it doesn't pull in libipkg into the rootfs and something later expects it to be there). Sep 02 04:21:23 hrw|work: wasn't you at all - apologies for the rash assumption Sep 02 04:21:33 ok Sep 02 04:21:48 my liers will contact you Sep 02 04:22:18 is that the new spelling for lawyers ? Very appropriate :-) Sep 02 04:22:29 10 "Hail OE" 's ought to do it eh? Sep 02 04:22:44 goto 10 :P Sep 02 04:23:32 Hmm - all the old versions of ipkg have been removed, so we can't go back to one that works ... Sep 02 04:23:50 I can blame someone for that instead :-) Sep 02 04:24:21 yeah Sep 02 04:24:32 Anyone remember which version we were on before? Sep 02 04:24:54 152 Sep 02 04:28:34 dyoung: I'm pinning nslu2-linux to 0,99,152 Sep 02 04:29:56 funky Sep 02 04:37:04 schuris goes amock... http://openzaurus.org/wordpress/newsletters/ooo-newsletter-200508/ 3.rd Sep 02 04:38:13 amock like amock.. monotone is really slow Sep 02 04:38:36 complaining doesn't make it faster Sep 02 04:38:42 koen: I know Sep 02 04:39:01 koen: I use it cause I have to use it Sep 02 04:39:29 we had a long period of time where people could propose SCMs Sep 02 04:40:03 so anyone who didn't speak up then should really keep his mouth shut Sep 02 04:40:27 yeah, I find it interesting that svk wasn't even mentioned during that time (as far as I could see), but OE developers had been using it since February. Yet it starts being talked about after the decision ... Sep 02 04:41:07 changing SCM *again* isn't an option anymore Sep 02 04:41:32 03dyoung 07org.oe.dev * rb144329b... 10/packages/ipkg/files/libdir.patch: ipkg: Add the libdir.patch that keeps getting lost Sep 02 04:42:18 I'm always looking for better stuff... Sep 02 04:45:12 koen...you can push changes into the repository, right? Sep 02 04:46:48 moin Sep 02 04:46:48 among 30 other people, yes Sep 02 04:46:54 lol Sep 02 04:46:58 apt-get moo Sep 02 04:47:01 aptitude moo Sep 02 04:47:04 aptitude -v moo ... Sep 02 04:47:13 Hi zecke Sep 02 04:47:15 :) Sep 02 04:47:19 hey Sep 02 04:47:29 koen has a special flair for typing mt push though .... Sep 02 04:48:27 well... Sep 02 04:48:32 rwhitby: scripts :) Sep 02 04:48:35 i didn't mean to offend anyone... Sep 02 04:48:37 :-D Sep 02 04:48:45 i have just found a problem in a .bb Sep 02 04:48:45 03rwhitby 07org.oe.dev * rcb9a7aa4... 10/conf/machine/nslu2.conf: Pin ipkg to 0.99.152 until it's fixed for depends Sep 02 04:48:58 rwhitby: ehm Sep 02 04:48:59 and figured if I talked to someone specifically...it had a better chance of getting fixed Sep 02 04:49:03 rwhitby: pin where? Sep 02 04:49:07 rwhitby: MACHINE.confs aren't for that Sep 02 04:49:38 koen: since it's temporary, it was easier than making three temporary changes to dists confs Sep 02 04:49:46 it is the wrong place Sep 02 04:49:58 and ipkg .153 doesn't get built by default Sep 02 04:50:01 rwhitby: then create common part for them Sep 02 04:50:05 agreed, I'll fix it after ipkg is fixed :-) Sep 02 04:50:23 rwhitby: I disabled .153 a few csets ago Sep 02 04:50:33 rwhitby: report a bug in hh.org bugzilla Sep 02 04:53:09 did anyone notice we still have ZERO replies to the mail stating people should test other SCMs? Sep 02 04:53:14 yep Sep 02 04:53:35 which means ZERO people have the right to complain about monotone Sep 02 04:54:11 I tested making copies of files marked with the date as a suffix, but found it to be unmanageable compared to monotone ... Sep 02 04:54:37 I've tested (co)git(o) and monotone Sep 02 04:54:43 monotone clearly won Sep 02 04:55:19 I'd suggest ClearCase, but it's even more expensive than BitKeeper ... Sep 02 04:56:37 apart from being tied to bkbits.net bk was pretty good Sep 02 04:57:07 no argument here ... Sep 02 04:57:18 if I understand svk correctly you have to run a svn server to publish your stuff Sep 02 04:58:04 running a monotone server is trivial Sep 02 04:58:25 I only see 2 problems with monotone: the merging algo (and I want empty merges too) and the speed Sep 02 04:59:02 * koen *really* wants empty merges in monotone Sep 02 04:59:07 and it needs a friendlier front-end ... Sep 02 04:59:21 but we've scripted that away for the moment ... Sep 02 05:00:23 montone 0.23 should be a lot faster in the common operations (add, commit) Sep 02 05:03:10 somebody updated libmimedir_cvs.bb to include mimedir-recurrence.h, which was causing a problem for me Sep 02 05:03:22 but they didn't add the 2 other headers that also aren't in the .bb Sep 02 05:03:30 is there a reason for this that someone knows? Sep 02 05:03:39 or should I submit a bug report? Sep 02 05:03:39 laziness Sep 02 05:03:53 ok...'cause the lack of the other headers are hurting another build... Sep 02 05:03:59 the gpe-bluetooth build... Sep 02 05:04:06 so I'll submit a bug... Sep 02 05:04:08 :-D Sep 02 05:04:14 ehm, no Sep 02 05:04:45 ah crap Sep 02 05:05:42 LoDown: gpe-bluetooth 0.43 still uses the old libmimedir api Sep 02 05:09:34 hm Sep 02 05:09:57 so...libmimedir_cvs.bb will not supply bluetooth with what it needs? Sep 02 05:10:19 well...crap...here I thought I was being all smart...oh well...the bb needs to be changed anyway Sep 02 05:11:55 haha Sep 02 05:12:10 i just saw you were the one that fixed that bb, koen Sep 02 05:13:42 koen, http://bugs.openembedded.org/show_bug.cgi?id=285 Sep 02 05:14:50 could you make a patch? Sep 02 05:15:17 i don't know how to make a patch...so I just edited the .bb Sep 02 05:15:19 :( Sep 02 05:15:28 and uploaded my changed bb Sep 02 05:15:40 i guess I should go learn how to make patches, eh? Sep 02 05:15:45 diff -Nurd oldfile newfile Sep 02 05:15:54 monotone diff file Sep 02 05:18:21 thanks...from now on it will be in the form of a patch Sep 02 05:18:39 it easier to spot the differences and easier to apply Sep 02 05:19:01 yep Sep 02 05:19:12 i understand that... Sep 02 05:20:06 so, what do I do about gpe-bluetooth? Sep 02 05:20:17 go and talk to them in #gpe? Sep 02 05:20:19 wait till 0.44 gets released Sep 02 05:20:24 haha Sep 02 05:21:24 no, seriously, it's going to be released today Sep 02 05:21:28 ok Sep 02 05:21:33 14:20 < CIA-3> florian * gpe/base/gpe-bluetooth/ (configure.ac main.c): Release Version 0.44 Sep 02 05:21:43 i was like...hmmm...i saw you were asking about it in #gpe... Sep 02 05:22:00 good : Sep 02 05:42:38 mornign Sep 02 05:42:52 morning mickeyl Sep 02 05:44:38 ~seen RP Sep 02 05:44:45 rp was last seen on IRC in channel #oe, 4h 37m 37s ago, saying: 'morning ade|desk '. Sep 02 05:46:03 03mickeyl 07org.oe.dev * rc6803271... 10/packages/ (36 files in 12 dirs): Sep 02 05:46:04 bluez: update and put related .bb files into one directory Sep 02 05:46:04 - bluez-libs, bluez-utils, and bluez-hcidump are pretty tied, so I think it makes sense to put them into the same directory Sep 02 05:46:04 - update bluez-libs to 2.20 Sep 02 05:46:04 - update bluez-hcidump to 1.25 Sep 02 05:46:05 - enable default-manfid patch Sep 02 05:48:03 mickeyl: I postulated that ZERO responses to our 'test another SCM and report back' means ZERO people have the right to complain about monotone Sep 02 05:51:18 NOTE: package bootstrap-image-1.0-r0: task do_build: completed Sep 02 05:51:20 yes ;) Sep 02 05:51:28 ipkg 0.99.153 traced Sep 02 05:51:45 atleast it looks like Sep 02 05:52:30 koen: *nod* Sep 02 05:53:38 mickeyl: http://openzaurus.org/wordpress/newsletters/ooo-newsletter-200508/ in the comments Sep 02 05:54:20 ibot, seen zecke_ Sep 02 05:54:23 zecke_ is currently on #gpe (21h 54m 25s) #opie (21h 54m 25s) #handhelds.org (21h 54m 25s) #oe (21h 54m 25s). Has said a total of 11 messages. Is idling for 1h 7m 3s Sep 02 05:56:30 can someone apply that to ipkg 0.99.153 and try to rebuild image? http://pastebin.ca/21934 Sep 02 05:59:17 dan2003_: yes? Sep 02 06:00:01 koen: but if we would switch the SCM, tons of developer would start helping developing ;) Sep 02 06:00:13 heh Sep 02 06:00:16 funny funny Sep 02 06:00:51 yeah Sep 02 06:01:08 and all companies (inerted by management) will dump OE Sep 02 06:01:33 we just need a cvs (or svn) gateway Sep 02 06:02:19 koen: One need to differentiate, renoos complains are valid but not constructive Sep 02 06:02:30 mickeyl: bitbakeshell lack "poke provider-version var value" :( Sep 02 06:02:31 so you all are talking about changing what SCM is used? Sep 02 06:02:34 koen: schurigs random bitch about monotone twice a day... Sep 02 06:02:37 zecke_, i was hoping you might recall some info on the simpad display issues with the hh 2.6 kernel? i have 2 simpad sl4, one i flashed with an image i built just a few days ago and was greeted witha corupt screen, this had previosuly had a 0.9.0 verion flashed to it along with the hh bootldr, however the other unit, first boot showed the display corectly,this had literlly just been flashed with hh bootldr, however after r Sep 02 06:02:37 eboot the screen was completly blank, and contiues to be so Sep 02 06:02:57 zecke_: schurig ramdonly bitches twice a day Sep 02 06:03:21 dan2003_: the hh.org 2.6.11 kernel update, broke the display light activation Sep 02 06:03:47 dan2003_: I simply lacked the time to look into it further Sep 02 06:04:03 dan2003_: also there are three different LCDs used Sep 02 06:04:13 zecke_, i have the exact same image on both units, the light works on the one, but displayu is garbled mess, on the other this could be the case Sep 02 06:04:16 ahha! Sep 02 06:04:26 iw as about to get onto that! Sep 02 06:05:08 i have looked at the hardware and it would seem that i have too different displays, ill look furtehr at this tonite if i get time Sep 02 06:06:01 dan2003_: actually it is two different revisions of the graphic controler as well Sep 02 06:06:26 dan2003_: but currently I can simply say, good luck in debugging ;) Sep 02 06:06:39 my mq200 chips, although both have the same part number stamped on them, were clearly manufactured in two different places, one even has germany stamoped on it whilst the other has not Sep 02 06:06:49 dan2003_: I'm extremely busy earning money to afford shelter and food in the semester break Sep 02 06:07:11 and the packages is slighlty different not to mentuon thr logo and fnts being entirly diffreent, both have suffix 'P' Sep 02 06:07:13 dan2003_: first of all look at the boot messages on serial console Sep 02 06:07:26 ~lart ljp for breaking mediaplayer1 after 20050827 Sep 02 06:07:26 * ibot grabs a large, mis-shapened log, with squirrels, and beats ljp until only the nuts remain ... which the squirrels run off with for breaking mediaplayer1 after 20050827 Sep 02 06:07:28 zecke_: if you were a cat I now an address in the UK for shelter ;) Sep 02 06:07:49 dan2003_: i print the revision and manufacturer numbers at boot time Sep 02 06:07:49 yup, have done some comparisons, but then i ran out of time, i will continue tonight if i get time Sep 02 06:07:50 ah, are our TT buddies sabotiging opie again? Sep 02 06:07:58 koen: I bet he would even have food to offer?! Sep 02 06:08:06 17973 and 512 i belive Sep 02 06:08:17 same on both anyhow Sep 02 06:09:15 hrw|work: yeah, that might be helpful. => bugzilla please. Sep 02 06:09:40 ipkg bugzilla that is? Sep 02 06:09:54 * koen really wants malone to be opensource Sep 02 06:10:36 mickeyl: will do Sep 02 06:11:01 first have to dig my access to bugzilla Sep 02 06:11:16 mickeyl: should we point complainers to monotone.cvssync and tell them to test it? Sep 02 06:11:17 (i was talking about poke providerversion) Sep 02 06:11:32 mickeyl: ah.. Sep 02 06:11:33 koen: sure Sep 02 06:12:57 * koen ponders about revoking monotone access for complainers Sep 02 06:13:03 let them use tarballs Sep 02 06:13:23 netsync can't blacklist IPs right? Sep 02 06:13:30 03dyoung 07org.oe.dev * rdca105f1... 10/ (3 files in 3 dirs): zd1211: patch makefile to make compatible with 2.4 kernels; add to unslung-packages Sep 02 06:15:44 one thing about policy I have... Sep 02 06:16:23 If 'foo' RDEPEND on 'bar' then 'foo' should also DEPEND on 'bar'? Sep 02 06:19:24 i don't we ever came to an agreement about that Sep 02 06:19:30 there are two oe developer camps here Sep 02 06:19:44 Hello Sep 02 06:19:47 one tries to enforce the strict "DEPENDS = minimal build time dependencies" Sep 02 06:20:06 I'm in that one which say "should" anyway so opie-console will be pushed with fix Sep 02 06:20:09 the other one is "DEPENDS = everything to make the package work in an image installation" Sep 02 06:20:23 go ahead then Sep 02 06:20:38 opie-console is the only one app which want opie-keytabs iirc Sep 02 06:22:27 hrw|work: and embedded konsole? Sep 02 06:23:36 03koen 07org.oe.dev * r445191ac... 10/packages/gpe-bluetooth/gpe-bluetooth_0.44.bb: gpe-bluetooth-0.44: add it to the repo Sep 02 06:23:45 zecke_: it does not have it in rdepends Sep 02 06:25:31 http://www.handhelds.org/~bugzilla/show_bug.cgi?id=1393 Sep 02 06:26:27 whoo hoo Sep 02 06:26:29 :-D Sep 02 06:26:44 zecke_: but should - yes Sep 02 06:26:58 wow 1/5th of GNOME developers are paid for working on it Sep 02 06:27:40 zecke_, dang, you all should so get paid (and I should too, if I start contributing ;) ) Sep 02 06:28:25 No I find it funny that GNOME got 100 paid developers where KDE has ~5 Sep 02 06:28:29 koen, can you push the change to libmimedir_cvs.bb as well? Sep 02 06:28:41 ha Sep 02 06:28:56 zecke_: I guess the KDE developers must be 20 times as clever. Sep 02 06:29:06 zecke_: and see how companies dumbed down gnome Sep 02 06:29:31 pb_ heh Sep 02 06:29:47 zecke_: or, perhaps they are just paid 20 times as much Sep 02 06:29:54 * Twiun laughs Sep 02 06:30:41 mayhaps they are payed 20 times as much because they are 20 times as clever... Sep 02 06:30:59 LoDown|school: did you attach the patch to bugzilla? Sep 02 06:31:08 oh...sh*t Sep 02 06:31:12 here I am pressuring you... Sep 02 06:31:13 tsk tsk Sep 02 06:35:03 koen: http://bugs.openembedded.org/show_bug.cgi?id=285 Sep 02 06:35:05 its up now Sep 02 06:35:33 if I screwed it up...privmsg me so that I will know better next time...must go to class now Sep 02 06:41:40 03koen 07org.oe.dev * rbc003912... 10/packages/libmimedir/libmimedir_cvs.bb: libmimedir_cvs: stage some more headers, patch courtesy Logan Owen, closes #285 Sep 02 06:45:19 yes :) finally opie-image is back to 20M Sep 02 06:45:43 HMM Sep 02 06:45:45 hmm Sep 02 06:46:13 mickeyl: c7x0 Sep 02 06:46:58 * kergoth yawns Sep 02 06:47:06 aah Sep 02 06:47:09 morning kergoth Sep 02 06:47:14 good morning kergoth Sep 02 06:48:21 mickeyl: http://www.hrw.one.pl/tmp/ has current versions of opie-image.bb and meta-opie.bb Sep 02 06:48:53 someone can take a look at busybox 1.01 which I submitted into bugzilla? Sep 02 06:49:31 http://bugs.openembedded.org/show_bug.cgi?id=247 Sep 02 06:57:44 morning all Sep 02 06:57:59 Hopefully my connection will last a bit longer than last time... Sep 02 06:57:59 morning Sep 02 07:00:09 OE has killed two of my servers disks now... Sep 02 07:00:29 overheating? Sep 02 07:01:12 dan2003: Not that I've noticed. Just worn them out it would seem :/ Sep 02 07:01:37 http://tree.celinuxforum.org/CelfPubWiki/OpenEmbedded Sep 02 07:02:26 didn't I paste that link 2 days ago? Sep 02 07:02:27 :) Sep 02 07:03:09 * CosmicPenguin rolls his eyes Sep 02 07:03:47 koen: I worked 16hours two days ago Sep 02 07:04:06 sponsor a conference/retreat of the core OE developers ? for who? Sep 02 07:04:36 i am sure lots of people would liek to visit you tho Sep 02 07:05:57 will soon have access to a dulacore opteron server, it will be idle at night (GMT) and i may be able to host nightly builds on it Sep 02 07:06:13 but is it a dual dual core opteron? :) Sep 02 07:06:16 after rotuine backups have run anyay Sep 02 07:06:35 its duak, but we only have 1 dual core cpu at present Sep 02 07:06:42 my add another lter Sep 02 07:06:44 so you are only half as l33t Sep 02 07:06:48 lol Sep 02 07:07:07 i bet CosmicPenguin is itching to say that for a cluster as well ... Sep 02 07:07:18 sorry, I'm torqued because I have a dual dual-core opteron on order Sep 02 07:07:24 lol Sep 02 07:07:29 must..not...brag..about...hardware Sep 02 07:07:44 ah Sep 02 07:07:54 it will be on a 1gig fibre link in telecom house, if taht ads to any l33tness ? Sep 02 07:08:08 * koen remembers the "*Cosmicpengiun sees the 8 way opteron rip through the glibc build" quotes Sep 02 07:08:29 Heh - I wish Sep 02 07:08:48 but limited to 2megbyte/sec Sep 02 07:08:54 :( Sep 02 07:09:02 * zecke_ needs to start his Tinderbox replacement soon Sep 02 07:13:42 anybodyelse quite excited about getign one of these nokia 770 things? or am i just being sad Sep 02 07:14:13 * koen is still waiting for nokia to mail the codes Sep 02 07:14:31 hehe Sep 02 07:14:48 seems prety much like amoderniased simpad with sliglty less display! Sep 02 07:18:43 dan2003_: yep, I reckon it looks pretty cool - and it seems to be rather checp too from what people have been saying Sep 02 07:19:12 99 euro :) Sep 02 07:19:17 ? Sep 02 07:19:28 Special discount? Sep 02 07:19:28 * koen got the developer discount Sep 02 07:19:29 lardman|work: developer price Sep 02 07:19:41 Ah, cool Sep 02 07:20:12 i saw $350 but i dunno if that was us or what Sep 02 07:20:23 I can find no mention of the internal hw - like what processor (speed) & GPU, etc. Sep 02 07:20:32 its a 220 Sep 02 07:20:33 arm Sep 02 07:20:38 opmap 1710 Sep 02 07:20:46 dan2003_: yes I saw that too $350/350 euros Sep 02 07:20:47 think it saud 220 anyway, seesm werid now Sep 02 07:20:56 ill try find it again Sep 02 07:20:59 lardman|work: http://maemo.org/maemowiki/Nokia_770_Hardware_Specification?highlight=%28hardware%29 Sep 02 07:21:25 and http://focus.ti.com/general/docs/wtbu/wtbuproductcontent.tsp?templateId=6123&navigationId=11991&path=templatedata/cm/product/data/omap_1710 Sep 02 07:21:26 koen: thanks Sep 02 07:21:58 zecke_: freenx is winning the fight Sep 02 07:22:12 http://linuxdevices.com/news/NS5409534614.html Sep 02 07:22:16 220Mhz! Sep 02 07:22:19 zecke_: it just is too hard to get it into OE Sep 02 07:22:23 koen: where?, what fight? Sep 02 07:22:30 koen: well, we have it here Sep 02 07:22:37 One of the reviews said that it had a strange BT config where is wouldn't be able to pair with keyboards for some reason - anything behind this? Sep 02 07:22:42 zecke_: in OE? Sep 02 07:22:57 lardman|work: bullshit like normal Sep 02 07:23:24 koen: that's what I though - someone reading too much into Nokia's statement about how its BT impl would be safe Sep 02 07:23:25 lardman|work: people read 'non discoverable by default' and make stupid assumptions Sep 02 07:23:39 vi /etc/bluetooth/bluez.conf Sep 02 07:23:53 "Flash player version 6" that'll please people on oesf :D Sep 02 07:23:58 RP: ping Sep 02 07:24:02 koen: at least half of the libs Sep 02 07:24:03 hrw|work: pong Sep 02 07:24:28 RP: http://pastebin.com/352806 when building linux-oz 2.6.13-mm1 Sep 02 07:24:41 zecke_: any chance of getting them into the repo soon? Sep 02 07:25:06 koen: need to ask Sep 02 07:25:24 hrw|work: This is fixed locally. I mustn't have pushed it... Sep 02 07:25:55 Is monotone.vanille.de down again? Sep 02 07:26:13 It seems it went down while I was pushing. Sep 02 07:26:21 WTF there are two 770' advertised on ebay at 99 GBP Sep 02 07:26:46 I got a lot of 'discarding revision data packet *** with unmet dependencies' Sep 02 07:27:31 XorA: looks dodgy though Sep 02 07:27:50 XorA: No paypal, comms outside of ebay - sounds a bit like a scam Sep 02 07:28:07 lardman|work: yeah, I gathered that, one of them has been pulled by ebay Sep 02 07:28:31 lardman|work: and technically that auction is for 20 of them for 99 quid Sep 02 07:28:36 shiping cost not mentioned! Sep 02 07:29:17 Anyone know how fast the TI processor is in comparison to a PXA? Sep 02 07:30:12 RP: ok Sep 02 07:30:23 RP: so I'll wait Sep 02 07:30:29 hrw|work: update to sharpsl_pm-r5 Sep 02 07:30:47 hrw|work: monotone is thinking about it Sep 02 07:31:22 ok Sep 02 07:31:37 ~lart belkin for flooding market with non linux BT cards Sep 02 07:31:37 * ibot flings poo at belkin for flooding market with non linux BT cards Sep 02 07:32:15 RP: PR bumped into r1 Sep 02 07:46:10 RP: NOTE: package linux-openzaurus-2.6.13-mm1-r1: task do_build: completed Sep 02 07:47:12 hrw|work: It was tested on spitz so hopefully, c7x0 will be ok as well Sep 02 07:49:00 hrw|work: change pushed Sep 02 07:50:52 RP: will do test tomorrow probably Sep 02 08:13:11 re Sep 02 08:15:48 cu Sep 02 09:01:27 is monotone.vanille.de down? Sep 02 09:04:19 LoDown|school: use the mirrors Sep 02 09:05:47 i am... Sep 02 09:06:01 i just wanted to make sure I didn't do something wrong :) Sep 02 09:09:46 * CosmicPenguin hunkers down and waits for the flames Sep 02 09:10:14 flames? Sep 02 09:10:19 where do I sign up? Sep 02 09:15:07 CosmicPenguin: so it's essiantially a xml-ized .bb Sep 02 09:17:10 yeah Sep 02 09:17:22 what is the status of gpe-bluetooth-0.44? Sep 02 09:18:05 LoDown: it builds over here Sep 02 09:18:23 but concentrating more on the components and versions - I would expect that the individual image .bbs would still exist to handle the actual scripting Sep 02 09:18:39 ok... Sep 02 09:19:40 CosmicPenguin: I see one big problem Sep 02 09:20:02 CosmicPenguin: OE has a bad record of being able to build 'old' distro-versions Sep 02 09:20:26 so a 'revision id' might be handy in the BOM Sep 02 09:20:54 Oh, to track changes in .bbclasses, for example? Sep 02 09:21:02 Thats a good idea Sep 02 09:21:24 or general big improvements Sep 02 09:21:36 like FHS'ing busybox as we did a month ago Sep 02 09:23:16 whoohoo, gpe-bluetooth built...onwards and forwards... Sep 02 09:26:48 Hmmm... I wonder if we're starting to talk about making bitbake RCS aware for its own metadata Sep 02 09:26:51 that would be sort of freaky Sep 02 09:27:01 I just wondered that myself Sep 02 09:27:18 selfaware....just remember to shoot it in the head ;) Sep 02 09:27:40 03hrw 07org.oe.dev * r057273e8... 10/packages/opie-console/ (opie-console.inc opie-console_1.2.0.bb opie-console_cvs.bb): opie-console: added 'opie-keytabs' to DEPENDS Sep 02 09:27:44 03hrw 07org.oe.dev * rbc5db216... 10/packages/opie-embeddedkonsole/ (3 files): opie-embeddedkonsole: added 'opie-keytabs' into DEPENDS Sep 02 09:27:48 03rpurdie 07org.oe.dev * r5678fb66... 10/packages/linux/linux-openzaurus_2.6.13-mm1.bb: linux-oz-2.6: Update sharpsl_pm patch due to compile errors. Sep 02 09:27:52 03rpurdie 07org.oe.dev * rca81ad46... 10/packages/maemo/nokia770-init_1.0.bb: nokia770-init: change to hotplug instead of hotplug-dbus and only need a runtime dependency Sep 02 09:28:02 But it would totally help some of the louder complaints on the mailling list if bitbake was capable of downloading and installing a OE tarball on its own Sep 02 09:28:15 heh Sep 02 09:28:20 yeah Sep 02 09:28:38 we need to have OE tarballs on oe.handhelds.org Sep 02 09:29:02 not just on http://ewi546.ewi.utwente.nl/OE/ Sep 02 09:29:18 and I like to get the cvssync stuff working Sep 02 09:30:42 CosmicPenguin: florian and I discussed branching of for released distros (org.openembedded.familiar.083) Sep 02 09:31:03 that helps Sep 02 09:31:59 hi everyone. Sep 02 09:32:29 where can i have a look at the curently .bb files availiable ? Sep 02 09:32:39 monotone.vanille.de Sep 02 09:32:47 thanks. Sep 02 09:36:12 what are the snapshots availiable at http://ewi546.ewi.utwente.nl/OE/ ? Sep 02 09:36:49 is it what monotone tries to download ? Sep 02 09:37:11 i mean the entire /bla/oe directory ? Sep 02 09:37:26 no Sep 02 09:37:35 ha... Sep 02 09:37:35 monotone downloads data from another monotone server Sep 02 09:39:17 CosmicPenguin: another option is to do OE 'releases' Sep 02 09:39:46 koen: yeah - but thats difficult in a company with a validation cycle - you end up with zillions of "releases" for each validation cycle Sep 02 09:40:04 indeed Sep 02 09:40:04 koen: so what are thos snapshots made for ? Is this a way not to update the db from monotone but to download the entire databse ? Sep 02 09:40:22 alan|home: read GettingStarted Sep 02 09:40:32 Ok. thanks. Sep 02 09:41:47 03justinp 07org.oe.dev * r9aa9c46d... 10/packages/ipkg/ipkg_0.99.153.bb: ipkg: fix bb syntax Sep 02 10:10:49 when bitbake is fetching mtd-utils-native from cvs, do we really need to get the source for jffs3? Sep 02 10:11:02 que? Sep 02 10:11:06 jffs3 is dead Sep 02 10:11:06 i don't know how much time that would save, but I'm sure it would save a little to exclude that branch Sep 02 10:11:10 hmm... Sep 02 10:11:20 branch? it checks out mtd-utils. Sep 02 10:11:48 wrong terminology...sorry Sep 02 10:12:15 but it fetches mtd/fs/jffs3 Sep 02 10:12:22 and all its subdirectories...etc Sep 02 10:12:37 i guess there is no way around that? not easily at least? Sep 02 10:13:00 no, the fetcher does a cvs checkout. thats al. Sep 02 10:13:08 cvs checkouts recurse into all subdirs by default Sep 02 10:13:14 ok Sep 02 10:13:29 if I say something stupid...just fishslap me :-D Sep 02 10:14:05 * LoDown|afk is off to lunch... Sep 02 10:15:10 wow - kergoth is here Sep 02 10:15:22 I thought he would be off with non-computer related gf activities.. :) Sep 02 10:16:13 huh.. the same apparently compiler bug on multiple compilers Sep 02 10:16:23 s/ly// Sep 02 10:17:23 CosmicPenguin: heh :) she's reading a book atm Sep 02 10:17:30 we're going to the minnesota state fair a bit later Sep 02 10:18:17 kergoth: you're in MN? I didn't know that Sep 02 10:18:32 I just moved from there last year Sep 02 10:18:33 yeah. from mn originally, moved back after i qui tthe job at ti Sep 02 10:18:39 ah Sep 02 10:18:55 I was living near park & 32nd in mpls Sep 02 10:19:02 ah Sep 02 10:19:05 i'm in brooklyn park Sep 02 10:19:12 much nicer neighborhood :-) Sep 02 10:19:50 I miss the great restaurants, though not the weather Sep 02 10:19:59 * koen thinks http://ewi546.ewi.utwente.nl/maps/OE.php needs some fresh info Sep 02 10:20:28 mreimer: yeah... i'm not looking forward to the upcoming winter.. i've enjoyed texas's winters Sep 02 10:20:33 mreimer: where you at now? Sep 02 10:20:46 shit, i need to buy a beater, i dont want to drive my crossfire in the snow Sep 02 10:20:50 kergoth: western washington, where it rarely hits 90, and if it does, it's dry :-) Sep 02 10:21:29 its dry in western washington? Thats not what I've heard... Sep 02 10:21:36 With that big wet thing right next door and all Sep 02 10:21:44 Just for July/Aug/Sept :-) Sep 02 10:22:11 it does rain quite a bit the rest of the time Sep 02 10:23:06 JustinP: error building e-wm: "/opt/oe/build/tmp/staging/i686-linux/bin/edje_cc: Error. :434 token FRAME not one of: RAW COMP LOSSY USER" Sep 02 10:23:10 JustinP: what do you suggest? Sep 02 10:29:08 blech, buying hw bites Sep 02 10:29:33 hey kergoth has a gf, thats great Sep 02 10:29:38 he needed one Sep 02 10:30:49 hmm, now i forgot why i am using the computer ... Sep 02 10:31:25 matched ram that was it Sep 02 10:34:38 if anyone knows of a place for 1gbx2 matched for under 350 please let me know Sep 02 10:38:35 mreimer: are you using the newest OE? I fixed an error like that a while ago Sep 02 10:38:47 yes, just updated a few minutes ago. Sep 02 10:38:52 rebuild edje? Sep 02 10:39:05 yeah, try that Sep 02 10:44:09 sigh Sep 02 10:44:29 kergoth: I dunno about tinysnmp - I'm excited to try it, but this guy does all sorts of stupid ass things with this makesystem he has Sep 02 10:45:02 ifndef CROSS Sep 02 10:45:02 COLOR = color Sep 02 10:45:02 else # ifndef CROSS Sep 02 10:45:02 CFLAGS += -mapcs-32 -march=armv4 -mtune=strongarm1100 -mshort-load-bytes Sep 02 10:45:02 endif # ifndef CROSS Sep 02 10:45:04 for example Sep 02 10:46:09 ah.. dual 3ghz xeon is a much better build machine than my 2.1ghz celery Sep 02 10:46:44 allthough.. it only has 256M of ram.. weird. Sep 02 10:49:50 zecke, i have swapped the displays in my simpads, the corruptd display has stayed with the computer and not the display Sep 02 10:50:24 the backlight is lit ont he one with the blanks screen, tho there is no image on the display Sep 02 10:53:16 my displays are both toshiba LTM08C343S Sep 02 10:53:52 03justinp 07org.oe.dev * red6f5e4d... 10/packages/ (4 files in 3 dirs): gpe-dm, gpe-session-scripts, entrance: x11-common => xserver-common Sep 02 10:53:56 03justinp 07org.oe.dev * r6b8389ae... 10/packages/musicpd/ (mpd/save-volume-state.patch mpd_0.11.5.bb): mpd: Add a patch to save/restore volume state in the state_file Sep 02 10:54:00 03justinp 07org.oe.dev * r44e79e78... 10/packages/ (4 files in 3 dirs): gpe-dm, gpe-session-scripts, entrance: bump PR Sep 02 10:58:48 good evening Sep 02 11:02:24 JustinP: same edje error ("token FRAME not one of: RAW COMP LOSSY USER") :-( More suggestions? Sep 02 11:06:58 03justinp 07org.oe.dev * r9ae1962d... 10/packages/synergy/synergy_1.2.4.bb: Add synergy Sep 02 11:07:02 03justinp 07org.oe.dev * r43cbbf0c... 10/packages/slutils/ (3 files in 2 dirs): slutils: add slfb, a command-line utility to turn on/off the FrameBuffer for 2.4 Zaurus kernels Sep 02 11:07:31 mreimer: what's the last time your built e17 in OE? Sep 02 11:07:42 mreimer: also, what's the last time you built from scratch? Sep 02 11:07:58 mreimer: and finally, what kind of host system are you on again? Sep 02 11:08:00 the first time I ever tried was about a week ago Sep 02 11:08:06 host debian x86 Sep 02 11:08:41 I did "rebuild edje" but noticed that most of the edje binaries in the staging area didn't change. I've deleted those and am rebuilding in hopes of getting new binaries Sep 02 11:09:10 well...I've never used the shell myself so I don't know about that Sep 02 11:09:14 aug 27 was the timestamp of the old edje binaries Sep 02 11:09:27 except for edje_config which was today Sep 02 11:09:30 I always do a -cclean then rebuild Sep 02 11:09:35 ok Sep 02 11:09:52 * JustinP should really try the shell.... Sep 02 11:10:08 -c clean and build -b is the same as 'rebuild foo' Sep 02 11:10:48 ok Sep 02 11:11:12 argh Sep 02 11:11:18 stupid fetching crap.... Sep 02 11:12:43 how do I force bb to refetch something? I deleted the source file and all stamps/work and it just goes right on trying to unpack the nonexistant file Sep 02 11:12:46 that's just stupid Sep 02 11:12:54 if the file doesn't exist bb should try to refetch... Sep 02 11:13:15 remove the md5 Sep 02 11:14:30 argh Sep 02 11:14:37 yep, of course Sep 02 11:15:59 hi all Sep 02 11:16:12 hey hrw Sep 02 11:17:31 hi hrw Sep 02 11:17:33 bye all Sep 02 11:21:18 pulling... Sep 02 11:24:44 one funny thing about monotone... Sep 02 11:24:55 try "mt log" during pulling Sep 02 11:25:59 does not test does database is not locked by it's other process (or sth like that) Sep 02 11:27:38 file a bug Sep 02 11:29:32 https://savannah.nongnu.org/bugs/?func=detailitem&item_id=8551 Sep 02 11:30:32 https://savannah.nongnu.org/bugs/?func=detailitem&item_id=9006 is our ipkg empty dirs bug Sep 02 12:01:29 How can I properly tell glib-2.0 to use libiconv as opposed to libc's iconv? Sep 02 12:05:22 see familiar-0.8.3.conf Sep 02 12:07:26 in other words - iconv is in glibc and should be used - unless you build uclibc system Sep 02 12:27:46 heh - gmane puts up an amd logo on my e-mails Sep 02 12:27:47 cute Sep 02 12:33:57 where can i find loader_bl.alt i find metions to use it everytwhere but cant find the file itself Sep 02 12:39:21 k found it, it tells me to use loader.alt tho! Sep 02 12:51:51 Would someone mind committing the simple pcmcia patches in http://bugs.openembedded.org/show_bug.cgi?id=287 Sep 02 12:53:17 what provides `module_id` ? Sep 02 12:54:07 mreimer: since it isn't present on my h5550 running 2.4 Sep 02 12:54:34 koen|slug: I realized it was a function defined in some of the other scripts like 12keymap. It should be in the patch, if I didn't botch it Sep 02 12:54:49 ah, ok Sep 02 12:55:10 this patch takes care of that "module_id not found" error Sep 02 12:55:24 ah, cool Sep 02 12:57:48 when I make a bugfix to something like gpe-session-scripts, do I need to update the version number and 'make source'? Or just 'make source'? Or nothing else? Sep 02 13:02:06 that would be a question for #gpe :) Sep 02 13:02:24 oops, thanks Sep 02 13:03:53 ok...i know this won't be a big deal to you guys...but my first home built image just booted Sep 02 13:04:43 but....i don't think the touchscreen is working...haha Sep 02 13:04:47 congrats, that's a big deal Sep 02 13:04:58 :-D Sep 02 13:05:12 it took me long enough to work all the kinks out...thats for sure... Sep 02 13:06:05 phew...reboot and the touchscreen works now... Sep 02 13:06:06 haha Sep 02 13:06:51 ~cheer LoDown built gpe-image Sep 02 13:06:59 or not Sep 02 13:06:59 haha Sep 02 13:13:32 mreimer: of course you need to rebuild edje-native, not edje, to update the tools in i686/bin Sep 02 13:13:39 evening pb_ Sep 02 13:13:52 mickeyl: I realized that, and did that, but it didn't help Sep 02 13:13:55 hi pb Sep 02 13:13:55 ok Sep 02 13:14:01 then i only see the option to rebuild from scratch Sep 02 13:14:06 :-( Sep 02 13:14:16 if that still doesn't work, we need to inspect the theme file on which edje_cc chokes Sep 02 13:14:40 hi mickey Sep 02 13:14:40 fyi, "Opening "/tmp/edje_cc.edc-tmp-nTINt9" for input" Sep 02 13:14:42 hi matt Sep 02 13:35:35 to anyone who cares, I just tested e-image-core again and it seems to be all working. Just make sure you're not using ipkg 153 Sep 02 13:37:28 ipkg 153 has been disable for some time now Sep 02 13:38:46 it got built for me Sep 02 13:38:57 "some time now" is still less than a day I believe Sep 02 13:39:46 can the shell be told to re-parse certain bb files? Sep 02 13:40:16 seems kind of useless for development if it can't.... Sep 02 13:46:36 How to i completely remove all traces of an ipkg build? Sep 02 13:47:12 There seems to be all sorts of cached data. Id like to get rid of it all so that I can re-compile it from scratch Sep 02 13:48:51 rm the TMPDIR Sep 02 13:50:15 that will get rid of my corss compiler etc. can i selectively remove a single package? or deos this require some extended knowledge? Sep 02 13:57:53 03koen 07org.oe.dev * r0aa9fa51... 10/packages/gpe-session-scripts/gpe-session-scripts_0.66.bb: Add gpe-sessionscripts 0.66 Sep 02 13:58:54 JustinTodd: no, it's not completely possible. The staging directories have no uninstall mechanism Sep 02 14:07:13 03koen 07org.oe.dev * rd83dde07... 10/conf/distro/preferred-gpe-versions-2.7.inc: conf/distro/preferred-gpe-versions-2.7.inc: use gpe-sessionscripts 0.66 Sep 02 14:26:14 Has anyone thought about using cpufreq? Sep 02 14:26:59 to make your cpu go faster when compiling? sounds like a good idea. Sep 02 14:27:38 I was thinking more along the lines of making my cpu go slower when I don;t need it so the batery lasts (slightly) longer Sep 02 14:30:59 a better idea would be contribute your spare cycles to the monotone@home distributed computing effort. Sep 02 14:31:16 :) Sep 02 14:31:18 harness your unused processor bandwidth to speed up other peoples' checkouts Sep 02 14:31:23 I'm not talking about my desktop Sep 02 14:31:28 I'm talking about my Zaurus Sep 02 14:31:53 but if you think it might help, then I'm more than willing ;) Sep 02 14:32:41 * koen encourages pb to experiment with the monotone.cvssync branch Sep 02 14:33:09 I may be the only one, but I'm finding monotone pulls rather speedy these days Sep 02 14:33:24 I don't have any problems with them either Sep 02 14:33:26 0.22 speed those up a lot Sep 02 14:33:31 on both my fast *and* slow machines Sep 02 14:34:11 people just need to 1) stop waiting for a week before pulling (pull != update) and 2) stop thinking it's going to be instant Sep 02 14:34:55 i'd settle for 15min Sep 02 14:34:58 :) Sep 02 14:35:15 it's must faster than 15 minutes usually Sep 02 14:35:17 much Sep 02 14:35:20 Oops, opened another can of worms :) Sep 02 14:35:29 depends on how many revs Sep 02 14:36:07 3) have spoken up earlier when we hadn't made a choice yet Sep 02 14:36:41 i remember monotone's speed being brought up at that point koen Sep 02 14:37:30 altho that now seems to be thier main effort to fix Sep 02 14:37:38 * mickeyl hugs /ignore '*monotone*' Sep 02 14:38:00 heh Sep 02 14:38:18 lardman: it's a good idea and we are going to do it. the problem is that convincing implementations on arm/2.4 were kind of lacking. We are confident we'll have cpufreq working in our 2.6 kernels soon Sep 02 14:38:44 * mickeyl looks to RP and smiles Sep 02 14:39:05 i bet he is smiling back Sep 02 14:39:34 mickeyl: I found a patch for 2.6.0 and am trying to apply it to 2.6.13-rc50mm1, but the files have changed slightly Sep 02 14:39:34 with both fingers Sep 02 14:39:35 RP has to slay the RMK dragon first, hasn't he? Sep 02 14:40:00 I mainly wanted to see how much of a difference it might make Sep 02 14:40:05 pssst... his friends call him 'Siegfried' Sep 02 14:40:45 wouldnt the difference depends on chip specs? Sep 02 14:41:19 lardman: it does make a difference. in the first 2.6 kernels for c7x0 RP didn't enable the PXA turbo mode and people were happy that battery time was increasing a lot Sep 02 14:41:20 heh Sep 02 14:41:39 yes, I remember that, videos weren't so good though ;) Sep 02 14:41:55 yeah. it needs to depend on workload Sep 02 14:42:25 I was reading up on it - it's nice that the implementation for the processor and the strategy are clearly separated Sep 02 14:42:55 governors et al Sep 02 14:47:48 * mickeyl ponders Sep 02 14:48:11 I guess no one here can give a recommendation for whether to use STL, boost, or anything else for a new c++ project, right? Sep 02 14:49:02 I used STL for some stuff in college.....I found it pretty useful Sep 02 14:49:04 I can recommend not to use c++ at all Sep 02 14:49:11 yeah, use PHP Sep 02 14:49:15 :_| Sep 02 14:49:18 lol Sep 02 14:49:37 the language is not negotiatable Sep 02 14:49:43 just the libraries Sep 02 14:49:51 * RP smiles ;-) Sep 02 14:49:57 what are you writting mickeyl ? Sep 02 14:50:01 I would recommend trying STL before you settle on it Sep 02 14:50:17 Some love it, and others abhor it Sep 02 14:50:20 I am in the second camp Sep 02 14:50:30 you hear the same about boost as well tho Sep 02 14:50:49 All I know is that boost .33 won't compile on my damn opteron Sep 02 14:50:51 emte: my EFL C++ wrapper is in a state where I need to decide which utility classes I want to use. Sep 02 14:51:10 lardman: cpufreq is on my todo list. I have lots of things on it though :-/ Sep 02 14:51:38 * JustinP doesn't understand why Xinit isn't being run on the e images.... Sep 02 14:51:50 RP: Understood. I just thought I'd have a play around with it if I can get the patches to apply Sep 02 14:52:30 mickeyl, well if you ask raster he might tell you which is faster :P Sep 02 14:52:43 dunno about anything beyond that tho Sep 02 14:52:48 emte: he'd say use C Sep 02 14:52:54 *nod* Sep 02 14:52:56 lardman: Feel free. The gumstick patches were the closest I last saw. I think cedric had something working on poodle Sep 02 14:52:57 nah Sep 02 14:53:07 he knows what mickeyl is writting Sep 02 14:54:19 I've used STL before. it's ok and i can deal with it. the only thing turning me down about STL is its size Sep 02 14:54:43 I wouldn't like to suck in about 800K Sep 02 14:55:01 * mickeyl takes another look at uClibc++ Sep 02 14:55:23 you sure you need to link the whole damn thing? I'm pretty sure it's just what you use... Sep 02 14:55:33 then again I wasn't trying for speed back when I used STL Sep 02 14:55:40 s/speed/size/ Sep 02 14:55:45 heh, either really Sep 02 14:55:54 hmm well right, i could statically link Sep 02 14:56:22 mickeyl: and encrypt the binary Sep 02 14:56:33 of course :D Sep 02 14:56:35 lol Sep 02 14:56:42 all my future work will be encrypted and closed Sep 02 14:56:45 call it e-update.sh Sep 02 14:56:48 hehehe Sep 02 14:57:17 ah yes, true...if you installed STL as its own thing it would be huge Sep 02 14:57:19 but... Sep 02 14:57:32 hmmm, doesn't it not work that way Sep 02 14:57:34 ? Sep 02 14:57:55 what with the templates and all...it gets re-compiled per type you put in it, doesn't it? Sep 02 14:58:07 damnit i forgot what i was doing again ... Sep 02 14:59:52 yeah, templates are just source reuse. you just pay for the type instances you need Sep 02 15:00:16 hmm uclibc++ ~ 100k Sep 02 15:00:22 i guess that means standard STL for me Sep 02 15:00:43 then i can build small images with the uClibc(++) stuff Sep 02 15:01:36 then again, i could use *cough* qt4core Sep 02 15:01:48 * mickeyl puts on the flame suit# Sep 02 15:02:51 was that QT ref earlier towards you there mickeyl ? Sep 02 15:03:15 * JustinP kicks Xinit Sep 02 15:03:22 don't think so. which ref? Sep 02 15:04:30 mmm gone from my buffer Sep 02 15:04:58 nm. Qt attacks are seldomly against me. QtE perhaps, but not Qt ;) Sep 02 15:05:11 we all love qt, we just have a problem with qte Sep 02 15:05:13 heh Sep 02 15:05:31 cant say i am a fan of qt Sep 02 15:05:39 its a beast to comple Sep 02 15:05:41 except the TT sabotage squad Sep 02 15:06:43 'night all Sep 02 15:06:47 night Sep 02 15:07:14 night koen Sep 02 15:07:19 ~tuckin koen|sleep Sep 02 15:07:28 bot needs a new command :P Sep 02 15:07:45 good evening all! Sep 02 15:07:54 still people awake? Sep 02 15:08:06 generally Sep 02 15:10:03 just about Sep 02 15:10:31 hardly Sep 02 15:10:50 I've build an opie-image for familiar and now wanted to build a gpe-image in the same build tree. Should that work fine? Sep 02 15:11:10 in general yes Sep 02 15:11:16 just take a look at things like sdl and such Sep 02 15:11:40 what does that mean? Do I need to force rebuilding sdl? Sep 02 15:12:10 yeah Sep 02 15:12:18 rebuild sdl-x11 (or so) Sep 02 15:12:21 and all deps Sep 02 15:12:29 or they will still be linking against libqpe Sep 02 15:13:44 Aha. Something for step #2. The build failed with a message "ipkg: Cannot create directory `/usr/lib/ipkg': Permission denied" Sep 02 15:13:57 in gpe-image-1.0-r17 Sep 02 15:14:15 Saw some updates in archive, so now retrying. Sep 02 15:14:28 sgm: that's normal (afaik) grep for ERROR and see what it says Sep 02 15:15:00 | ERROR: Cannot satisfy the following dependencies for gpe-task-base: Sep 02 15:15:00 | ERROR: Cannot satisfy the following dependencies for gpe-task-pim: Sep 02 15:15:00 | + '[' -f /home/sgm/work/OE/build/tmp/rootfs/usr/lib/ipkg/info/libgpg-error0.postinst ']' Sep 02 15:15:00 | + sh /home/sgm/work/OE/build/tmp/rootfs/usr/lib/ipkg/info/libgpg-error0.postinst configure Sep 02 15:15:48 sgm: Is that from the output to the screen? Look through the log file and find what the unsatisifed deps are - they are listed in there Sep 02 15:18:17 Found it: Sep 02 15:18:20 ERROR: Cannot satisfy the following dependencies for gpe-task-base: Sep 02 15:18:20 libmimedir (>= 0.0+cvs-20050902) Sep 02 15:18:20 ERROR: Cannot satisfy the following dependencies for gpe-task-pim: Sep 02 15:18:20 libmimedir (>= 0.0+cvs-20050902) libmimedir (>= 0.0+cvs-20050902) Sep 02 15:18:39 I've a fixed CVSDATE = 20050902 in my local.conf Sep 02 15:19:26 mickeyl: how do I force all dependencies of sdl-x11 to be recreated? Sep 02 15:29:14 no way Sep 02 15:29:20 that's something bitbake doesn't do automatically Sep 02 15:31:10 Yeah, it is a difficult problem to solve in general. A solution would be to build a sdl-for-opie and sdl-for-gpe etc. Sep 02 15:31:25 (and all dependancies in the same way) Sep 02 15:42:30 hmmmm...DISPLAY Sep 02 15:42:47 mickeyl: slfb....any thoughts? I checked it in earlier today Sep 02 15:43:02 it's 2.4 embeddix specific, I know that Sep 02 15:46:21 ya, looks ok Sep 02 15:49:35 mickeyl: with either GPE or OZ, shouldn't dev-entries be created dynamically when a serial cs-device is inserted? I don't see that happen with the most recent monotone builds (pre-3.5.4 an gpe-as-it-is-now) Sep 02 15:49:53 this largely depends on the machine Sep 02 15:50:02 Husky Sep 02 15:50:19 udev is currently in unknown state Sep 02 15:50:21 it's probably that Sep 02 15:50:51 we would need someone who really knows that stuff to fix our udev infrastructure Sep 02 15:51:06 whay, cpufreq patched kernel built successfully Sep 02 15:51:37 mickeyl: ah, ok, since things work when creating the dev entries manually (c 4 67) wouldn't it be, for the meantime, be handy to create it at startup? That way stuff works out of the box Sep 02 15:52:13 no, i'm afraid we're rather waiting for someone to fix it properly. if we workaround it now we just forget about it Sep 02 15:53:26 mickeyl: is it a matter of creating the correct udev rules, or is there some deeper black art required, it, I guess, being related to the pxe-serial/serial driver collision. Sep 02 15:53:58 as far as i can see, it's a) updating our initscript to the udev shipped version, b) writing rules to substitute hotplug. Sep 02 15:54:05 nothing more should be required Sep 02 15:54:20 the pxa/serial is already handled by my hack Sep 02 15:59:31 so....anyone? How do I reparse a single provider in bitbake -i? Sep 02 16:00:39 I'm getting rusty - if there's just a Makefile, which explicitly says something like "gcc -O2 -Wall -o powernowd powernowd.c", should I patch it, or is there some magic which can be worked? Sep 02 16:01:16 JustinP: I'm afraid that's not yet possible. => bugtracker Sep 02 16:01:36 lardman: patch it to use $(CC) $(CFLAGS) Sep 02 16:01:43 $(LDFLAGS) Sep 02 16:02:19 mickeyl: bah.....how is the shell better than normal command-line then? It lets you edit a provider's bb file but you have to reparse everything to use it? Sep 02 16:02:22 * JustinP digs Sep 02 16:03:35 uhm Sep 02 16:03:49 look at the shell's version number Sep 02 16:03:54 ? Sep 02 16:03:59 look at the shell's number of authors Sep 02 16:04:13 heh Sep 02 16:04:14 it does exactly what i need atm. Sep 02 16:04:21 ok Sep 02 16:04:22 if you need more... feel free to provide patches Sep 02 16:04:45 I just don't understand how it's better the way it is....I thought the major upside was that it would lower parsing time Sep 02 16:04:49 I'm looking into it now Sep 02 16:05:14 of course it lowers parsing time. think "bitbake foo" Sep 02 16:05:20 wait until it's finished Sep 02 16:05:24 ah darn, i want bar as well Sep 02 16:05:25 bitbake bar Sep 02 16:05:31 bingo, shell reduces your time 50% Sep 02 16:05:36 true Sep 02 16:05:48 you're right Sep 02 16:05:52 then you can peek and poke vars, get list of providers, and much more Sep 02 16:05:59 yes, it is useful Sep 02 16:06:03 it's just not finished Sep 02 16:06:04 I'm working on adding what I need Sep 02 16:06:08 that'd be excellent Sep 02 16:06:16 even though I've never touched python ;-) Sep 02 16:06:22 oh, and I'm trying to fix entrance Sep 02 16:06:30 Xinit isn't working Sep 02 16:06:37 I assume because DISPLAY isn't set.... Sep 02 16:06:46 you may find that for the one or the other thing you have to refactor things in bitbake itself. don't worry about such things, I had to do it a lot and it improved the structure/maintainability/readability of bitbake Sep 02 16:06:56 (just in case) Sep 02 16:07:08 heh Sep 02 16:07:10 ok Sep 02 16:08:48 Is it possible to do bitbake -c clean on a .bb file directly? Sep 02 16:09:00 yep :-) Sep 02 16:09:16 damn it.... Sep 02 16:09:56 why isn't this getting my new bb parsed? Sep 02 16:10:51 hmmm...I had to exit the shell and re-enter for it to find my changes... Sep 02 16:11:16 there are some subtle caching issues involved. there's a bug tracker entry for that as well IIRC. I'm sorry but it's been a while since i last hacked on the shell and my head is full with other things atm. Sep 02 16:11:24 anyway, sleep time Sep 02 16:11:26 g'night Sep 02 16:11:32 night mickeyl Sep 02 16:12:00 night Sep 02 16:27:26 lol. I just added powernowd to OE, only to see that cpufreqd and cpuspeed are already there Sep 02 16:35:20 night all Sep 02 16:44:20 03justinp 07org.oe.dev * r6a06d3e1... 10/packages/e17/ (4 files in 3 dirs): entrance: set DISPLAY before calling Xinit Sep 02 16:44:24 03justinp 07org.oe.dev * rdd4e3c7a... 10/packages/e17/ (3 files in 3 dirs): entrance: run entrance from entrance_wrapper instead of from Xinit to fix login issues Sep 02 16:46:11 03justinp 07org.oe.dev * rfed15f0f... 10/packages/e17/entrance_20050810.bb: entrance: remove unused directory from SRC_URI Sep 02 17:05:09 * JustinP apologizes for all of the incremental fixes Sep 02 17:06:35 03justinp 07org.oe.dev * r6b98c67d... 10/packages/e17/entrance_20050810.bb: entrance: remove more Xinit.d references Sep 02 17:07:37 Time to go to sleep! Sep 02 17:08:24 good night all! Sep 02 17:10:07 night Sep 02 17:21:19 is it possible to get any debug info out of qpe without recompiling with the debug option set? Sep 02 17:36:15 probably not Sep 02 17:36:33 you might be able to get a coredump of some kind.... Sep 02 17:36:36 why, what's wrong? Sep 02 17:46:19 wow...this is really a pitiful diff.... Sep 02 17:46:24 but it works, so.... Sep 02 17:50:45 now....where do I post bitbake patches.... Sep 02 17:53:31 JustinP, it justs seems to die with no explanation as to why Sep 02 17:54:28 if i do qpe after its says going into deaomn mode, nothing happens, ps -A reveals its not running Sep 02 17:57:24 what if you start it not in daemon mode? Sep 02 17:57:36 how do i do that? Sep 02 17:58:38 qpe --help Sep 02 17:58:43 woops, i didi a passwd and it wont except my passwd for login now! think i should have shut it down cleanly! Sep 02 17:59:09 sure i tried that and it just tried to start as usual, will try it on the other one Sep 02 17:59:23 probably --nodaemon Sep 02 17:59:52 --help has no effect Sep 02 18:00:53 maybe alin in the config file would do it, Sep 02 18:02:04 oh well, ill carry on in the morning, need some sleep , cya Sep 02 20:02:00 03rwhitby 07org.oe.dev * r8a85b8d7... 10/conf/distro/ (openslug-packages.conf ucslugc-packages.conf): openslug-packages.conf,ucslugc-packages.conf: Updated location for bluez. Sep 03 01:01:35 morning all Sep 03 01:03:23 morning RP Sep 03 01:56:47 Morning. Sep 03 02:53:59 Morning. **** ENDING LOGGING AT Sat Sep 03 02:59:57 2005