**** BEGIN LOGGING AT Thu Oct 06 02:59:56 2005 Oct 06 03:00:09 hrw|movie: thx Oct 06 03:39:32 03koen 07org.oe.dev * re456a231... 10/: Oct 06 03:39:32 packages/meta/: image tweaks Oct 06 03:39:32 * make gpe-image use ROOT_FLASH_SIZE to determine what to install Oct 06 03:39:32 * introduce gpe-task-connectivity Oct 06 03:39:32 * ship gpe version of industrial Oct 06 03:46:05 03koen 07org.oe.oz354fam083 * rc352b67e... 10/: gpe-image tweak from .dev Oct 06 03:48:16 koen: 16M contra rest of machines? :) Oct 06 03:50:01 koen: what's the default when ROOT_FLASH_SIZE isn't given? Oct 06 03:53:27 Arjan: empty Oct 06 03:53:59 hrw|movie: yeah, I copied it from opie-image :) Oct 06 03:55:58 koen: so when it isn't given, it has the same behaviour as before? just install the standard stuff? Oct 06 04:25:38 cu Oct 06 04:32:14 Arjan: it will add the big package for anything that doesn't set ROOT_FLASH_SIZE to 16 Oct 06 04:46:50 * Geo_KM is back Oct 06 04:55:25 03koen 07org.oe.dev * r6d6af9eb... 10/: packages/matchbox-panel/matchbox-panel_0.9.2.bb: use small icons for h2200 Oct 06 05:01:22 <_law_> http://pastebin.com/384860 gpe-image doesnt build Oct 06 05:02:39 03koen 07org.oe.dev * r7e26dca9... 10/: packages/meta/meta-gpe.bb: add RDEPENDS for gpe-task-connectivity Oct 06 05:21:46 * MP0 sighs Oct 06 05:22:13 18 hrs and monotone still hasn't verified a pull... needed to let that out Oct 06 05:23:00 use the snapshot Oct 06 05:24:12 I was just about to pull after a week or so, perhaps I'll do the snapshot thing too - where are they? Oct 06 05:24:39 see GettingStarted Oct 06 05:24:47 I just thought of that... Oct 06 05:25:33 the location of the snapshots is mentioned twice there Oct 06 05:26:23 koen: yep, found it. Perhaps I hang out on oesf too much - forgotten how to search ;) Oct 06 05:27:23 I'm used to forums where posts like that get thrashed by the mods with 'Search harder' Oct 06 05:28:43 is the snapshot on GettingStarted up to date? Oct 06 05:29:36 would it be mentioned there if it wasn't? Oct 06 05:29:44 yes Oct 06 05:29:49 you can also check the timestamp at http://ewi546.ewi.utwente.nl/OE/ Oct 06 05:31:39 hey hey ! Oct 06 05:36:37 what are the plans for cleaning out .bb that clearly not going to be used ever again, e.g. gpe-login_0.71->0.81 ? do they ever go or are we keeping them for fun ? Oct 06 05:37:56 once in a while we kill off old versions Oct 06 05:38:34 I think we usually keep 2 or 3 versions Oct 06 05:50:03 Ciao all Oct 06 05:51:35 hey Pigi Oct 06 05:52:40 03koen 07org.oe.dev * red3a73f5... 10/: packages/gpe-bootsplash/gpe-bootsplash_1.14.bb: add patch to make it compile against the cairo we are using in gpe-2.7 Oct 06 05:53:32 hi pigi Oct 06 05:53:57 hi koen pb_ ! Oct 06 06:07:42 shit, gpe-image fails because gst-plugins fails: http://pastebin.com/384921 anybody has an idea? Oct 06 06:10:28 03koen 07org.oe.oz354fam083 * rb8ac1e99... 10/: packages/meta/meta-gpe.bb: add RDEPENDS for gpe-task-connectivity Oct 06 06:13:24 hmm Oct 06 06:13:52 koen: the udev --move fix... Didn't I commit that? Or am I confused now? ;) Oct 06 06:14:59 afternoon Oct 06 06:16:59 Arjan: it builds fine over here Oct 06 06:17:31 koen: are you using the oz354fam083 tree too? Oct 06 06:18:23 yes Oct 06 06:18:39 hmm, strange... *tries again* Oct 06 06:25:06 oh.. different branch Oct 06 06:34:23 03koen 07org.oe.dev * r43606e02... 10/: packages/gpe-bootsplash/files/splash-*.svg: update svgs to gpe 2.7 Oct 06 06:34:28 03koen 07org.oe.dev * r88df4b08... 10/: packages/meta/meta-gpe.bb: add gpe-bootsplash back to gpe-base Oct 06 06:48:22 <_law_> koen: http://pastebin.com/384957 Oct 06 06:54:18 Pigi: a packes always provides itself, right? Oct 06 06:56:59 koen, at least it should ;) Oct 06 06:59:27 I have a meta-package with the following control-file: http://pastebin.com/384966 Oct 06 06:59:53 which is built and all Oct 06 06:59:56 but I get http://pastebin.com/384957 Oct 06 07:02:16 koen, gimme 5 minutes and I'll look at it Oct 06 07:09:11 koen, I suspect that the "Packages" file is correctly up-to-date, isn't it ? Oct 06 07:09:28 gstreamer still breaks on the fact that it needs cdaudio.h :( Oct 06 07:09:56 what's happen if you do "ipkg list | grep gpe-task-connectivity" Oct 06 07:10:35 Pigi: it's during build Oct 06 07:10:43 http://pastebin.com/384977 Oct 06 07:12:19 I know, and I see that is during the build, but ( AFAIR ) ipkg needs a Packages file to work ( or a single ipk ) Oct 06 07:12:47 I know why Oct 06 07:12:56 I put it in DEPENDS and INSTALL Oct 06 07:13:10 so it's an OE error I get Oct 06 07:13:22 meta-packages should only go into INSTALL Oct 06 07:13:41 pfiuuu..... I tought you had find another bug :) Oct 06 07:14:07 And I can't look at it until my PC comes back from repair Oct 06 07:31:04 files//foo.svg will get picked over files/foo.svg, right? Oct 06 07:31:24 03koen 07org.oe.dev * re49417c7... 10/: packages/meta/gpe-image.bb: sort out DEPENDS vs. INSTALL business with tasks Oct 06 07:34:55 later all Oct 06 07:35:25 monotone: misuse: branch org.openembedded.dev has multiple heads Oct 06 07:36:14 monotone: misuse: branch 'org.openembedded.dev' is merged Oct 06 07:36:24 pull again I'd say Oct 06 07:37:06 ok Oct 06 07:41:05 <_law_> koen, http://pastebin.com/385002 Oct 06 07:41:45 hmm, I should have fixed that with that commit ten minutes ago Oct 06 07:42:28 <_law_> monotone: already up to date at e49417c7a343d2f9b0ca072431ef95a3fc918a22 Oct 06 07:42:55 <_law_> try it later again, moving home now Oct 06 07:46:32 found it! Oct 06 07:46:38 ~stab local changes Oct 06 07:46:40 * ibot_ runs at local changes with an origami Swiss Army knife, and inflicts a nasty paper cut. Oct 06 07:47:06 * koen cleans some packages, monotone reverts and builds again Oct 06 07:55:43 what should that be with swiss army knife?? Oct 06 07:56:19 koen: is it normal that it takes more then 48 hours to pull the fam0.8.3 from server with monotone? Oct 06 08:12:51 monotone: connecting to monotone.vanille.de Oct 06 08:12:51 monotone: network error: failed to connect: Connection refused Oct 06 08:13:14 anyone able to fix that? Oct 06 08:14:46 reenoo_: fixed Oct 06 08:15:04 thanks Oct 06 08:16:06 03koen 07org.oe.dev * r982dea7f... 10/: packages/meta/gpe-image.bb: it now builds all deps Oct 06 08:24:29 03koen 07org.oe.oz354fam083 * r77426890... 10/: packages/meta/gpe-image.bb: unbreak gpe-image with various patches from .dev Oct 06 08:29:21 03rw 07org.oe.dev * rc1171d13... 10/: kernel.bbclass: don't run 'make dep' for 2.6 kernels Oct 06 08:29:33 heh Oct 06 08:29:41 surprised that fix has made it this long without being made Oct 06 08:30:20 well. last time I checked it didn't hurt to run make dep Oct 06 08:31:25 apart from a "no need for this, dumbass" message or something Oct 06 08:39:50 :) Oct 06 08:48:34 <_law_> new gpe-image boots, but gpe-splashscreen is black on my Sl-760 Oct 06 08:58:02 <_law_> but suspend works again :-) Oct 06 09:01:57 _law_: How's backlight control? Oct 06 09:02:48 _law_: Try building and installing packages/bl/bl_0.0.bb Oct 06 09:07:42 <_law_> lardman|work: how does it work? Oct 06 09:08:57 _law_: It may already be present in your build I'm not sure - it's the /usr/bin/bl binary which controls the backlight Oct 06 09:09:36 _law_: Hacked to work with 2.6.x kernels (and other machines I hope) - I was just happy to see it's been added to OE Oct 06 09:10:25 <_law_> if i try to call it from command line i get: Unable to open /dev/touchscreen/0 Oct 06 09:10:39 <_law_> istn available Oct 06 09:11:01 _law_: Can you look at ipkg status and see which version of bl you've got installed? Oct 06 09:11:47 <_law_> 0.0cvs20051006-r0 Oct 06 09:12:10 _law_: Yeah, you need to remove the cvs version & install the plain 0.0 version Oct 06 09:12:53 <_law_> i only have bl_cvs.bb Oct 06 09:14:29 _law_: Download it from: http://people.bath.ac.uk/enpsgp/Zaurus/bl/bl_0.0-r0_armv5te.ipk Oct 06 09:15:07 _law_: it is in the OE data I just pulled about an hour ago Oct 06 09:17:38 <_law_> up and down keys doesnt work :-( Oct 06 09:17:57 <_law_> lardman|work: bl says on 168 Oct 06 09:17:58 03koen 07org.oe.dev * r42c5e248... 10/: packages/gpe-bootsplash/gpe-bootsplash_1.14.bb: ship working svg files and bump PR Oct 06 09:18:23 hmm. viewmtn on vanille.de is somehow broken Oct 06 09:18:32 no revision details Oct 06 09:18:35 it doesn't show diffs anymore Oct 06 09:18:37 right Oct 06 09:18:44 _law_: that's working then - that's told you that you're on 168/255 * 100% of backlight power Oct 06 09:19:01 reenoo_: and http://ewi546.ewi.utwente.nl/tmp/viewmtn/ only works when the database isn't busy Oct 06 09:19:11 _law_: now stuff like auto-dimming, etc should work Oct 06 09:19:44 koen: any particular reason for that breakage? Oct 06 09:21:42 reenoo_: when monotone push/pull locks the db the web frontend breaks Oct 06 09:22:21 <_law_> lardman|work: light-applet doesnt work Oct 06 09:22:42 is OZ still shipping /dev/sharp_fl? Oct 06 09:22:43 koen how long takes pull from fam0.8.3 becuase I have it running for about 3 days Oct 06 09:23:05 slow computer? Oct 06 09:23:07 _law_: what's light-applet? Oct 06 09:23:17 koen: ehm no Oct 06 09:23:33 koen: how many revs are there? Oct 06 09:23:46 I can pull the whole of .dev in 8 hours which is 2500+ revs Oct 06 09:23:57 the branch has far less revs Oct 06 09:24:32 <_law_> lardman|work: Frontlight control (gpe) Oct 06 09:24:42 _law_: minilite? Oct 06 09:24:54 _law_: How doesn't it work? Oct 06 09:26:14 <_law_> lardman|work: yes minilite, i can move the scrollbar but it doesnt anything Oct 06 09:26:16 3 Oct 06 09:26:17  Oct 06 09:26:46 _law_: Oh; did it work before? Oct 06 09:27:09 _law_: My GPE image is a couple of weeks old and both my bl and minilite work fine Oct 06 09:27:12 _law_: does ls /dev/sharp* return anything? Oct 06 09:27:42 koen: Surely he's running a 2.6.x kernel... Oct 06 09:27:49 _law_: You are aren't you? Oct 06 09:28:27 <_law_> lardman|work: /dev/sharp istn available Oct 06 09:30:41 lardman|work: yes but OZ used to ship /dev/sharp* regardless of kernel-version Oct 06 09:31:41 koen: Ah, right. Well minilite 0.49 built on 27/09/05 works for me on a C750 Oct 06 09:41:15 <_law_> koen: bootsplash doesnt work for me :-( Oct 06 09:42:16 03koen 07org.oe.oz354fam083 * r736c81c2... 10/: Update bootsplash and include it in the image Oct 06 09:42:50 _law_: where did you build from? .dev or .oz354fam083? Oct 06 09:42:59 <_law_> dev Oct 06 09:43:21 I see Oct 06 09:43:32 try building from .oz354fam083 Oct 06 09:44:54 koen: why? Oct 06 09:44:54 Anyone know anything about Ambicom WL1100C not working on Zaurus 5600 with 3.5.3 (as I've run out of ideas on #openzaurus)? Oct 06 09:45:17 lardman|work: see topic Oct 06 09:45:34 ~topic Oct 06 09:46:59 koen: there should *never* be any changes in .oz354fam083 that aren't in .dev. otherwise we will end up with an unmergeable fork. Oct 06 09:47:22 reenoo_: but the reverse will happen Oct 06 09:47:40 especially when RP returns and starts breaking stuff Oct 06 09:47:52 <_law_> ok upload screen also doesnt work :-( Oct 06 09:48:06 if RP wants to break stuff *he* should create a new branch Oct 06 09:48:18 i plan to do some ugly hacks to the openzaurus 2.6 kernel which never will land in .dev Oct 06 09:48:29 e.g. Oct 06 09:48:32 anyways. later all Oct 06 09:48:44 increasing keyboard timeout and SD/MMC detection timeout Oct 06 09:48:59 <_law_> add account is ugly on gaim :-( Oct 06 09:49:07 mickeyl: why do you want to remove 2.4 support for clamshells? Oct 06 09:49:25 koen: do I? that's news to me Oct 06 09:49:39 * koen prepares to repeat the OE != OZ speech Oct 06 09:49:59 mickeyl: you said a few days ago you wanted to remove the zaurus 2.4 machines Oct 06 09:50:20 oh. these clamshells Oct 06 09:50:35 yes. I want to remove that from HEAD because it's truly obsolete for c7x0. Oct 06 09:50:41 no other models though Oct 06 09:50:44 maybe for OZ Oct 06 09:51:14 GPE works perfectly w/ 2.6 nowadays on c7x0 doesn't it? Oct 06 09:51:21 if that's not the case we will have to reconsider it Oct 06 09:51:21 that's not the point Oct 06 09:51:28 OE != OZ Oct 06 09:51:32 so what? Oct 06 09:51:59 things like sharprom-compatible might rely on 2.4 Oct 06 09:52:08 they don't Oct 06 09:52:22 sharprom-compatible doesn't build kernels nor images Oct 06 09:52:38 if someone adds these, we can bring 'em back but I doubt that will happen Oct 06 09:52:40 mickeyl: yes it does imo Oct 06 09:52:59 mickeyl: the working perfectly bit that is Oct 06 09:53:07 ah Oct 06 09:53:12 yeah, that was my impression as well Oct 06 09:53:29 I can't see why they *have* to be deleted Oct 06 09:53:43 nothing in the world *has* to be deleted Oct 06 09:53:49 i see reasons for it though Oct 06 09:54:24 The only valid reason I see is that you can't build it with OE Oct 06 09:54:33 (2.95.3 toolchain required) Oct 06 09:55:30 I rather delete unmaintained stuff from the repository Oct 06 09:55:44 yeah Oct 06 09:56:12 I also think that maintaining OE patches outside OE is bad Oct 06 09:56:26 case in point: oz kernels Oct 06 09:58:21 i don't have a strong opinion on that. we can ask RP if he wants to maintain his patches in OE as well, but as long as his upstream is reliable, i don't see a special necessity. Oct 06 09:58:33 it isn't reliable Oct 06 09:58:47 that's news to me. never had any problem with rpsys.net Oct 06 09:58:53 <_law_> scap suspends my Z :-) Oct 06 09:58:55 we've had missing patches and the whole 'we have no stable kernel' sistuation Oct 06 09:59:15 what missing patches? Oct 06 09:59:21 no stable kernel ? Oct 06 09:59:26 must have missed that Oct 06 09:59:47 the whole kernel fuzz a month ago Oct 06 10:00:02 '2.6.foo broke stuff, we can't go back' Oct 06 10:00:07 i think 2.6.11 can be resurrected any time Oct 06 10:00:18 i agree that he deleted it a bit too early Oct 06 10:00:25 too much faith in upstream :D Oct 06 10:01:16 when I asked 'why don't we go back to an earlier version?' the answer was 'patches are gone' Oct 06 10:01:39 sounds like a misunderstanding. to my knowledge, he archives all his patches Oct 06 10:02:03 the second problem with that is with the GPL Oct 06 10:02:13 http://www.rpsys.net/openzaurus/patches/archive/ looks pretty comprehensive to me Oct 06 10:02:20 complete, even Oct 06 10:02:23 koen: Can you commit oe bug 373? Oct 06 10:02:35 we always presented sources + OE metadata as 'contains all' Oct 06 10:03:02 sure, that's the case Oct 06 10:03:08 patches are included in sources Oct 06 10:03:13 aren't theyß Oct 06 10:03:13 ? Oct 06 10:04:04 are they? Oct 06 10:04:10 i hope so Oct 06 10:04:11 yes they are Oct 06 10:04:29 excellent. otherwise the FSF would kill us again Oct 06 10:05:07 mickeyl: RP indeed lost a couple of his patches by overwriting them... Oct 06 10:05:34 that's sad, but I guess he will be more careful now. Oct 06 10:05:44 like i said, we can kindly ask him to add them to OE Oct 06 10:06:12 but if he doesn't want to do that, I rather leave him working on the kernel than working with OE Oct 06 10:06:41 besides, it's of course a question of space. Oct 06 10:06:49 do we really want all patches in OE ? Oct 06 10:06:52 even large ones? Oct 06 10:07:00 even those who are shortlived? Oct 06 10:07:02 *shrug* Oct 06 10:07:14 (not speaking about kernel here, think big) Oct 06 10:07:14 yes Oct 06 10:07:20 that's the point of a SCM Oct 06 10:07:34 uhm. by that logic we would kill all our fetches and would include all sources and binaries Oct 06 10:07:52 no Oct 06 10:07:57 well. why not? Oct 06 10:08:01 hah Oct 06 10:08:11 other projects do that Oct 06 10:08:13 people complaining when OE uses 4G of scratch Oct 06 10:08:22 people complaining when OE db is 100MB Oct 06 10:08:26 and you want to make it GigaBytes Oct 06 10:08:28 :D Oct 06 10:08:41 building takes GBs Oct 06 10:08:42 I wouldn't use monotone for that Oct 06 10:08:57 and I'm not saying all sources have to be distributed Oct 06 10:09:20 for sources and patches which are not included in OE we decided to use oesources.org. I don't see why we are reconsidering this now Oct 06 10:10:39 there's no correlation between the source I as a maintainer am working with and what oesources.org offers for download Oct 06 10:11:29 checking in sources and .bb into a sepository of some sort at the same time *does* make sense Oct 06 10:12:33 oesources.org was supposed to have the sources from a world build Oct 06 10:12:42 i don't see what you mean with lacking correlation Oct 06 10:13:24 perhaps a distributed populating of oesources.org is what you're after? Oct 06 10:13:36 so that every source one of us fetches is automatically uploaded there Oct 06 10:14:08 I think reenoo_ means by hand Oct 06 10:14:13 add a .bb, upload source Oct 06 10:14:53 i don't think it'd be a big problem to add a upload-to-oesources.org bbclass that does excactly that - uploading every new source Oct 06 10:15:10 security might be a problem Oct 06 10:15:15 if you want to upload by hand, sounds good to me as well - pester treke to give logins Oct 06 10:16:25 03frederic 07org.oe.dev * re5ed779b... 10/: Oct 06 10:16:26 libsdl: disable building static lib, don't install libtool .la file Oct 06 10:16:26 libsdl-qpe: fix sdl-config flags for compatibility with libsdl-x11 Oct 06 10:17:03 If people are complaining about the size of oe.db, why not make openembedded a server-side thing? Have some server run bitbake as a daemon (no reload of data... keep it all loaded) and perhaps some disk space to keep state information on the users. Then the server can tell the users only what actions they need to perform. Oct 06 10:17:39 that has evil disaster written all over it Oct 06 10:18:19 The only small problem I can see is a mismatch of state between clients and servers. Oct 06 10:18:31 i think the only ones complaining are just a few people who are not really interested in OE anyway. they want to grab as few stuff as possible to get the latest software for their machines, but they are not interested in the development of OE at all. I can very well ignore most of those. Oct 06 10:19:09 must .... resist..... Oct 06 10:19:12 So OE is only for developers of OE? Oct 06 10:19:14 *cough* gentoo *cough* Oct 06 10:19:24 damn, I failed Oct 06 10:19:30 MP0: who said only? Oct 06 10:19:48 hi Oct 06 10:19:53 Well, the implication that you can ignore people who aren't intending to improve OE is that OE isn't for them. Oct 06 10:19:56 but yes, the primary goal for OE is for people who are interested in maintaing stuff Oct 06 10:20:02 k Oct 06 10:20:11 OE is minimally at least for developers of OE, I imagine Oct 06 10:20:20 Its the only goal - If you don't want to get dirty building a image, then you're not going to use OE Oct 06 10:20:33 What are you going to use? Oct 06 10:20:40 of course, you can just grab it and build software and never give back. but that's not exactly the open source idea, you know Oct 06 10:20:43 You're going to grab a binary from somebody Oct 06 10:20:47 if OE weren't at least minimally sufficient, we'd need a new build system... something like OE Oct 06 10:21:11 hmm my RSI is getting worse Oct 06 10:21:14 i need to quit typing Oct 06 10:21:18 :/ Oct 06 10:21:19 However one of the goals of bitbake, and probably OE, is flexibility of the image. A prebuilt binary can't give anyone that. Oct 06 10:21:22 later Oct 06 10:21:44 Then you pay the price Oct 06 10:21:48 mickeyl: I've read ESR's rantings, and talked to various OSS folks, but I'm still not sure I understand what the "Open Source Idea" is. Oct 06 10:21:55 "Free Software" now, I get :) Oct 06 10:21:58 Bulding linux images is a nasty, difficult, confusing process Oct 06 10:22:16 no kidding! Oct 06 10:22:30 oh, sorry. bye mickeyl! Oct 06 10:22:30 OE mitigates some but not all of that - the point of OE is to facilitate the pain, not eliminate it Oct 06 10:22:41 hmm. maybe qpe-gaim is stabler... Oct 06 10:23:01 cosmicpenguin: its a developers' tool. pain-management without the sugar-coating. Oct 06 10:23:33 what did i miss? Oct 06 10:23:34 No Mary Poppins :/ Oct 06 10:24:02 hrw: you missed Mary Poppins, but so did we. Oct 06 10:24:19 raduga )) Oct 06 10:25:31 hrw: i'm wondering if its safe to try GPE on akita. Oct 06 10:25:43 im bored... being sick is boring... and i dont even have books to read ( Oct 06 10:25:46 borked very bad on my husky, to the point of uselessness. Oct 06 10:25:56 raduga - try... Oct 06 10:26:05 hrw: need etexts? Oct 06 10:26:26 raduga you dont saying about 353gpe i think? Oct 06 10:26:34 hrw: i'm... almost happy with Opie/akita now. Oct 06 10:27:01 reading etexts from c7x0 is to hardcore for me. not that eyes... Oct 06 10:27:04 hrw: I *believe* it was 3.5.3 gpe. Might have been 3.5.2, but I'm not certain. Oct 06 10:27:16 raduga: It's much better now Oct 06 10:27:17 353 is history Oct 06 10:27:30 we NEED new release Oct 06 10:27:30 would have been circa december 2004. Oct 06 10:27:56 Old phantom wisdom: Try to configure and build everything you need for gpe from cratch and then you know what OE is... Oct 06 10:28:02 hrw|husky: yep Oct 06 10:28:09 I think something's gotta be fundamentally wrong if people dedicated to building embedded linux kernels say that its nasty, tricky business. Does it really have to be? Oct 06 10:28:18 lamikr: it took some convincing ;) Oct 06 10:28:20 lardman: 3.5.3 release is better now? or snapshot of current dev is better now? Oct 06 10:28:29 raduga: snapshot current Oct 06 10:28:30 snapshot Oct 06 10:28:40 i'm using 3.5.3/hentges and its mostly very nice. Oct 06 10:28:49 koen: I tried myself couple of times... Oct 06 10:29:09 mickey|resting: took me a while to find the link, sorry: http://www.specifixinc.com/technology/Reprint-Wilson-OLS2004.pdf Oct 06 10:29:09 mp0 - building toolchain for only kernel is easy.. but rest of apps... Oct 06 10:30:13 raduga- current snapshot has many hires fixes Oct 06 10:30:23 you're right... I had a kernel compiled in only a few days. Oct 06 10:30:45 And thats only because somebody gave you a toolchain Oct 06 10:31:02 * raduga still totally newbish Oct 06 10:31:06 mp0- days? for m68k on x86 it took me few hours.. are those embedded are so fscked? Oct 06 10:31:11 in regards to OE at least Oct 06 10:31:35 raduga- config, run bitbake opie-image then flash Oct 06 10:31:40 depends what you build on too... Oct 06 10:31:55 hrw, I made the mistake of trying to use git to d/l linux repository Oct 06 10:32:04 hrw: right. Oct 06 10:32:19 I've *used* bitbake a few times. built a kernel, and a few apps. Oct 06 10:32:22 hrw, eventually, I came to my senses and just RSYNCed it. there were still some .config issues and patches I had to apply to get a working kernel though. Oct 06 10:32:25 reenoo_: that link doesn't work for me Oct 06 10:32:27 mp0 ah... i tried that in 2.2.10 times... when potato was stable.. Oct 06 10:32:43 how do you update the current oe.db? Oct 06 10:32:52 mp0 so why not tarballs? need to on the edge? Oct 06 10:32:55 from wiki: Oct 06 10:32:59 monotone --db=/stuff/oe.db pull monotone.vanille.de "org.openembedded.{dev,dreambox}" Oct 06 10:33:00 right? Oct 06 10:33:05 raduga- mt pull,merge,update Oct 06 10:33:07 mickey|resting, koen: sorry. try this one: http://www.linuxsymposium.org/proceedings/reprints/Reprint-Wilson-OLS2004.pdf Oct 06 10:33:09 montevista doesn't provide snapshots Oct 06 10:33:12 ok Oct 06 10:33:29 hail zecke Oct 06 10:33:31 hi zecke Oct 06 10:33:37 mp0 - auc.. Oct 06 10:33:41 hey Oct 06 10:33:51 gosh BenQ is sooo awesome Oct 06 10:33:53 what's the invocation for merge? Oct 06 10:34:01 zecke- how 1701 patches? Oct 06 10:34:20 ~benq Oct 06 10:34:53 * CosmicPenguin wishes -c compile worked.... :( Oct 06 10:35:23 raduga- cd working copy and do mt pull ; mt merge; mt update Oct 06 10:35:31 hrw|husky: Siemens handed its mobile department to BenQ Oct 06 10:35:50 cosmic- wfm from shell Oct 06 10:36:02 hrw|husky: ah ar upgraded mantis Oct 06 10:36:12 zecke- a... forgot that. what device you had? Oct 06 10:36:29 hrw|husky: yeah, but it gets hungup on quilt-native for some reason Oct 06 10:36:55 cosmic. i use -c Oct 06 10:36:56 with -b only Oct 06 10:37:01 yeah, yeah Oct 06 10:37:05 I know Oct 06 10:37:11 :) Oct 06 10:37:16 monotone.vanille.de needs a kick Oct 06 10:37:28 nevermind Oct 06 10:37:31 * CosmicPenguin stops whining Oct 06 10:37:34 monotone's just slow Oct 06 10:37:57 /nick hrw|sick Oct 06 10:37:57 * MP0 apologizes profusely Oct 06 10:38:09 hm. gaim does not ssspport it Oct 06 10:39:54 what's the dreambox branch? Oct 06 10:40:07 ~google dreambox Oct 06 10:40:43 ibot_: are you there? Oct 06 10:41:06 ibot: hello Oct 06 10:41:07 privet Oct 06 10:41:19 ibot_: hello Oct 06 10:41:23 ~dreambox Oct 06 10:41:32 ibot: dreambox Oct 06 10:41:44 ibot: wakeup Oct 06 10:42:08 heh Oct 06 10:42:12 ibot_: ping Oct 06 10:42:20 ~botsnack Oct 06 10:42:23 koen: thanks Oct 06 10:42:26 there he is Oct 06 10:42:38 hrw|husky : well I can not download the patch Oct 06 10:43:25 me too - dont want to run konq/e Oct 06 10:44:05 damn safari did download the file without telling me Oct 06 10:46:04 monotone: misuse: branch org.openembedded.dev has multiple heads Oct 06 10:46:17 I'm beginning to really dislike monotone Oct 06 10:47:15 MP0: well make someone merge it? Oct 06 10:48:04 I'm not sure what's causing it... I d/led a snapshot as someone suggested and then did a pull to update the snapshot Oct 06 10:48:24 hrw/koen: may I ask *you* "what is dreambox?" Oct 06 10:48:36 and I keep getting that... I have no existing org.openembedded.dev directory so I'm not sure what's going on. Oct 06 10:48:37 raduga- dvb settopbox Oct 06 10:48:39 raduga: a PVR Oct 06 10:48:56 quite popular iirc Oct 06 10:48:57 thanks :) Oct 06 10:49:03 should I re-download the snapshot and _not_ try to update it? Oct 06 10:49:08 according to http://monotone.vanille.de/viewmtn/index.psp everything is merged Oct 06 10:50:01 hello everyone - is there an example .bb I could look at to reference how to build & install kernel modules in the correct places of my image? Oct 06 10:50:24 dw- loook into machine configs Oct 06 10:50:28 I'm porting over a Makefile that uses 'uname -r' and such Oct 06 10:50:54 dw- loook into hostap-modules? Oct 06 10:51:59 hrw - thanks! I'll look over those now Oct 06 10:52:17 np Oct 06 10:58:12 bye Oct 06 11:02:33 hi! Oct 06 11:04:26 anyone has tried to bitbake a full gpe or opie image for the sl5500?? I always get stuck compiling libsoundgen or opie-tinykate :S Oct 06 11:05:14 you are the one with symlink problem? Oct 06 11:05:20 yes Oct 06 11:05:42 works for me is all i can tell Oct 06 11:06:42 hrw|husky: okay... I've now and trying now... Oct 06 11:06:58 *I've now updated* (was the missing art) Oct 06 11:07:01 *part Oct 06 11:08:11 ko Oct 06 11:08:24 hrw|gone: did you compile libsoundgen? I got an error 2 days ago about ld not finding libesd Oct 06 11:08:31 ~lart qpe-gaim Oct 06 11:08:32 * ibot_ holds qpe-gaim to the floor and spanks him with a cat-o-nine-tails Oct 06 11:08:46 kraptor- dont rememberl Oct 06 11:09:09 k Oct 06 11:15:21 koen: Still here? Oct 06 11:15:31 yes Oct 06 11:16:15 kraptor: Oops! y still get symlink errors packing opie-tinykate... maybe have oe misconfigured? Oct 06 11:16:15 about this 373, do you have time to commit it? Oct 06 11:16:33 if I was able to reach bugs.oe.org, yes Oct 06 11:16:40 ~lart kraptor Oct 06 11:16:40 * ibot_ drops a humongous exploding nuke on kraptor Oct 06 11:17:05 hrw|husky: last msg was for you :/ Oct 06 11:17:20 I am planning to put new image to building on today so I can test it. Oct 06 11:19:40 kraptor- im too sick to follow Oct 06 11:20:13 hrw|husky: ok Oct 06 11:25:17 cu Oct 06 11:40:40 ok... now I'm annoyed Oct 06 11:40:48 I downloaded a snapshot Oct 06 11:40:52 like it says on the tutorial Oct 06 11:41:04 then when I tried to check it out Oct 06 11:41:12 pingemi@robingoodfellow:/opt/oe$ monotone --db=/opt/oe/oe.db checkout --branch=org.openembedded.devmonotone: misuse: branch org.openembedded.dev has multiple heads Oct 06 11:41:56 cu all! Oct 06 11:42:09 why is that happening? Oct 06 11:44:33 sounds like the db had multiple heads when the snapshot was taken Oct 06 11:46:44 fooey. how long's a mt pull supposed to take? Oct 06 11:47:14 how do I get a good snapshot? Oct 06 11:48:19 there are snaps on http://ewi546.ewi.utwente.nl/OE/ Oct 06 11:48:53 what do you mean by "good snapshot"? there's nothing wrong with the snapshot. either merge the two heads yourself or pull from vanille.de or ewi546 Oct 06 11:49:21 I tried pulling from vanille.de... froze in the middle of verifying Oct 06 11:49:33 as for merging them myself... how do I do that? Oct 06 11:49:39 renoo: is it reasonable for a pull to take more than an hour? Oct 06 11:50:14 raduga: depending on how many revisions are transferred, yes Oct 06 11:50:21 reenoo: thanks. Oct 06 11:50:33 I can wait. I was only unsure if it was reasonable. Oct 06 11:51:15 I'm guessing this is how to do it: monotone --db=/opt/oe/oe.db merge --branch=org.openembedded.dev Oct 06 11:51:29 MP0: the verifying takes a while.. Oct 06 11:51:41 good evening folks Oct 06 11:52:03 MP0: might work, yeah. I've never merged without a working copy Oct 06 11:52:05 reenoo_, I'm aware, however it seemed to be stuck on a particular revision Oct 06 11:52:35 hmm, no, it didn't work. I think I need my own private key and it would sign the new merge with that. Oct 06 11:52:50 and then I'd be permanently out of sync with OE Oct 06 11:53:29 oh well, I'm going to do a pull, hope it works overnight. Oct 06 11:53:41 or I'm screwed.' Oct 06 12:21:28 hmm Oct 06 12:21:30 pulling from monotone.nslu2-linux.org Oct 06 12:21:44 seems a good deal faster than from vanille.de Oct 06 12:22:42 * raduga hopes its complete Oct 06 12:25:51 should be more or less the same Oct 06 12:26:11 yeah Oct 06 12:44:16 I just checked out the familiar 0.8.3 branch, and now bitbake is complaining about a missinig conf/bitbak.conf Oct 06 12:48:20 [21:48:45 lucas@wservices <0> build]% bitbake nano Oct 06 12:48:20 ERROR: Unable to open conf/bitbake.conf Oct 06 12:48:37 but where should that conf/bitbake.conf be? Oct 06 12:57:07 lucasvo: in org.openembedded./conf/bitbake.conf (or whatever your oe directory is called). make sure your BBPATH envvar is up to date. Oct 06 13:01:59 koen|tv: ping Oct 06 13:02:43 pH5: this is strange Oct 06 13:03:34 reenoo_ pong Oct 06 13:04:03 koen|tv: could you restart the monotone server on vanille.de once more? Oct 06 13:04:35 reenoo_: done Oct 06 13:04:49 thanks Oct 06 13:06:08 koen: can you think of a good reason why a pull from vanille.de would take >2hr while one from monotone.nslu2-linux.org would take < 5m? Oct 06 13:06:15 assuming nothing is broken on either side Oct 06 13:06:26 ERROR: [Errno 2] No such file or directory: '/home/lucas/pda/monotone.vanille.de/*/*.bb' while parsing /home/lucas/pda/monotone.vanille.de/*/*.bb Oct 06 13:06:27 no idea Oct 06 13:07:33 thanks :) Oct 06 13:09:37 ah Oct 06 13:33:09 lucasvo: your BBPATH is probably wrong Oct 06 13:33:24 JustinP: YES NOW IT WORKS Oct 06 13:33:46 ups sorry I have lag, I don't even see what I am writing :D Oct 06 13:35:32 ERROR: ${N2K_SRCURI} while parsing /mnt/disk/rw/bakery/org.openembedded.dev/packages/linux/openslug-kernel_2.6.12.2.bb Oct 06 13:35:48 could someone fix that please? Oct 06 13:44:08 03jbowler 07org.oe.dev * r1b00e951... 10/: Oct 06 13:44:09 ucslugc: le/glibc version: fix ucslugc-native for gcc and remove patcher Oct 06 13:44:09 Patcher is no longer required, so is not in ucslugc-packages.conf Oct 06 13:47:22 lucasvo: I just wanted to make sure you'd resolve dyour issue. Glad you got it working. Oct 06 13:48:22 JustinP: yes thank you Oct 06 13:48:32 it was the bbpath I have forgotton Oct 06 13:48:51 JustinP: do you know how long gpe compiling takes on a 1ghz? Oct 06 13:51:36 lucasvo: just qpe or everything from the toolchain on up? Oct 06 13:52:46 hi all Oct 06 13:53:03 JustinP: what do I have to bitbake to get a running ipaq? :D Oct 06 13:55:49 lucasvo: everything Oct 06 13:56:05 lucasvo: just "bitbake opie-image" or "bitbake gpe-image" depending on what you want Oct 06 13:56:18 * florian votes for gpe-image ;-) Oct 06 13:56:21 lucasvo: it will compile everything it needs (assuming you installed the prereqs) Oct 06 13:56:29 JustinP: I want gpe and do you know how long it takes? Oct 06 13:56:36 lucasvo: a long time Oct 06 13:56:38 is it reasonable for bitbaking a target like "opie-image" to generate many hundreds of locale ipks? Oct 06 13:56:43 yes Oct 06 13:56:49 glibc always creates those Oct 06 13:56:56 JustinP: I am missusing my friends webserver... :D Oct 06 13:57:04 is there a simply configurable setting to encourage it not to do that? Oct 06 13:57:14 justinp: I can live with it, if not. Oct 06 13:57:37 lucasvo: it will take hours. The good news is that once you finish making it once as long as you don't delete the tmpdir further compiles should be much faster Oct 06 13:57:43 raduga: no idea Oct 06 13:58:03 lucasvo: make sure you have the space needed as well Oct 06 13:58:14 lucasvo: I would estimate- half a day or longer Oct 06 13:58:28 perhaps more than a day. probably less than a week. Oct 06 13:58:34 raduga: well it took me 3d for pulling with monotone Oct 06 13:58:45 having lots of memory helps, I would imagine Oct 06 13:59:19 well if I don't do it in 3 days I will go to holiday Oct 06 13:59:30 arrgh Oct 06 13:59:32 ERROR: /mnt/disk/rw/bakery/org.openembedded.dev/packages/ipkg-utils/ipkg-utils-native_1.6cvs.bb depends on itself (eventually) Oct 06 13:59:35 building enough of the chain to make a kernel, and a few applications took me several hours Oct 06 13:59:49 I'm guessing bitbaking GPE will take awhile... Oct 06 13:59:53 (what I'm building now) Oct 06 13:59:58 reenoo_: 0.8.2 snap? Oct 06 14:00:11 koen|tv: no. current .dev Oct 06 14:00:23 weird Oct 06 14:01:14 also rebuilding a win2k box at work :[ Oct 06 14:01:47 Date: 2005-06-30T08:19:37 Oct 06 14:01:47 Branch: org.openembedded.dev Oct 06 14:01:47 Tag: clean bk import at cset 1.3670 Oct 06 14:01:59 that's what monotone log says for that .bb Oct 06 14:02:10 but some other .bb could have been changed of course Oct 06 14:02:26 the .bb isn't the problem Oct 06 14:02:46 * raduga wonders how long it would take to rebuild windows 2000 from source Oct 06 14:16:05 hi Oct 06 14:49:08 hi all Oct 06 14:49:20 hi hanjo-home Oct 06 14:49:47 Is anyone here working on a .bb for postgresql? Oct 06 14:51:25 I can't seem to find one in the monotone repository, as far as I can see, only mysql is there from the major databases Oct 06 15:20:35 bitbake gpe-image failing when building openzaurus-pxa27x-2.4.20-rmk2-embedix-r15 Oct 06 15:21:07 /local/home/silver/oe/build/tmp/work/openzaurus-pxa27x-2.4.20-rmk2-embedix-r15/temp/run.do_compile.26423: line 193: 2.4: command not found Oct 06 15:21:09 wacky Oct 06 15:24:56 raduga: do you have the zaurus 2.95 toolchain installed? Oct 06 15:25:59 should be. let me verify Oct 06 15:27:05 the error is happening at the end of kernel compilation Oct 06 15:28:28 right when it tries to build zImage Oct 06 15:28:51 and yes, gcc-2.95 is there Oct 06 15:31:56 sorry. it happens at the start of compile :/ Oct 06 15:32:16 misread stamps Oct 06 15:33:00 and you're sure you installed that toolchain right? Oct 06 15:33:06 correct path and links? Oct 06 15:35:01 justinP: yes. it was working earlier, when I explicitly built the kernel, from an older snapshot (mid september) Oct 06 15:35:18 something in the run.do_compile script is getting borked, looks like Oct 06 15:36:03 I see Oct 06 15:36:27 sorry, I'm distracted with work-things, else I'd give you a better description Oct 06 15:39:02 JustinP: what generates the run.do_compile scripts? Oct 06 15:39:49 don't ask me.... Oct 06 15:39:59 sorry to bother :) Oct 06 15:40:05 I just don't know ;-) Oct 06 15:40:11 sorry to bother you! Oct 06 15:40:47 wonder if the problem is in the .bb file for openzaurus-pxa27x Oct 06 15:47:33 it *looks* like the literal "2.4" is being inserted into the run script, instead of a variable VERSION or something. Oct 06 15:49:08 this test is failing: Oct 06 15:49:08 if 2.4 != "2.6"; then Oct 06 15:49:09 oe_runmake dep CC="ccache arm-linux-gcc-2.95" LD="arm-linux-ld- Oct 06 15:49:09 fi Oct 06 15:52:33 looks like kernel_do_compile is defined in kernel.bbclass Oct 06 15:53:09 interestingly, the bogus test isn't showing up in there. Oct 06 15:54:30 oh, yes it does! Oct 06 15:54:39 JustinP: think I've found the problem. Oct 06 15:54:45 ok Oct 06 15:55:01 JustinP: if you would prefer I stop talking to you, ok :) Oct 06 15:55:29 the problem is: Oct 06 15:55:30 if ${KERNEL_MAJOR_VERSION} != "2.6"; then Oct 06 15:55:50 is getting expanded to the literal, instead of variable, when the run script is generated Oct 06 15:55:56 not sure why Oct 06 15:56:46 who's responsible for bitbake? Oct 06 15:56:55 er. sorry. Oct 06 15:57:05 I blame George Bush Oct 06 15:57:11 * raduga digs into documentation to find appropriate place to report the "anomaly" Oct 06 15:57:22 this isn't bitbake Oct 06 15:57:25 it's a bb file Oct 06 15:57:28 so it's an OE bug Oct 06 15:57:33 bugs.openembedded.org Oct 06 15:57:33 justinP: yes. Oct 06 15:57:36 thanks. Oct 06 15:57:50 np Oct 06 15:59:41 maybe adding [ ] would help? Oct 06 16:19:15 justinP: grr. Oct 06 16:19:30 tried escaping it a few different ways. no luck. Oct 06 16:19:38 commenting out the bloody test makes it work. Oct 06 16:20:21 I'll report it to the bugs page Oct 06 16:24:34 raduga: if [ ${KERNEL_MAJOR_VERSION} != "2.6" ]; then Oct 06 16:25:57 hi all Oct 06 16:26:18 er. possibly I placed the [ wrong Oct 06 16:26:21 * raduga sighs Oct 06 16:26:48 JustinP: again thanks Oct 06 16:27:03 the oz distribution doesn't come with the arp command Oct 06 16:30:18 raduga: so it worked? cool :-) Oct 06 16:47:23 the arp command seems missing in the oz default kernel, I tried rebuilt the kernel with arp-tables but doesn't work still don't give me the "arp" command. am I missing something? Oct 06 16:50:28 abm_y4k: kernel has nothing to do with binaries or "commands" Oct 06 17:11:20 JustinP: what did I miss? how to get this binary command to bundle with the oz? Oct 06 17:11:35 when I bitbake opie-image, its not there Oct 06 17:12:40 abm_y4k: figure out which package it's in and install that package Oct 06 17:20:41 JustinP: all I find a package that close to its name is arpwatch, but thats not it.... :( Oct 06 17:27:27 well I don't know which package it is... Oct 06 17:30:18 anyone might know here? Oct 06 17:30:48 ipkgfind.handhelds.org or packages.debian.org should know Oct 06 17:34:32 thankx Oct 06 17:40:04 'night all Oct 06 18:18:59 How often is that OE.db mirror in the wiki updated? Oct 06 18:20:31 if the server is up every hour i think Oct 06 18:20:40 ah Oct 06 18:20:46 or maybe its once a day Oct 06 18:20:52 its one or the other Oct 06 18:21:25 Hrm. Oct 06 18:21:56 libqte2_2.3.10-r27_poodle.ipk, what i just built, locks up on my zaurus Oct 06 18:22:31 i wonder if a newer version is out Oct 06 18:36:08 monotone update monotone: multiple update candidates: Oct 06 18:36:10 whats up with this Oct 06 18:36:36 df00z: multiple heads Oct 06 18:36:39 df00z: pull Oct 06 18:38:09 pull what **** BEGIN LOGGING AT Thu Oct 06 18:45:03 2005 Oct 06 19:04:02 hey raster pumpkin Oct 06 19:04:30 hey emte, what actually provides the qpe binary? Oct 06 19:04:40 what package? Oct 06 19:04:45 its not libopie or qte Oct 06 19:04:48 qpe perhaps? Oct 06 19:05:13 libqpe? Oct 06 19:05:27 sounds plausible Oct 06 19:05:48 hmm Oct 06 19:06:58 libqpe-opie-1.2.1+cvs-20051007-r0 Oct 06 19:07:04 i dont see a qpe binary in the work directory Oct 06 19:07:39 no..just a library it looks ;\ Oct 06 19:12:09 ah-ha Oct 06 19:12:11 opie-taskbar Oct 06 19:12:17 ipkg search '*qpe*' Oct 06 22:32:11 kergoth, you maintain the sysvinit package right? Oct 06 22:34:12 I'm trying to create an image that will boot as an initrd so i'm wondering where linuxrc should point to. Oct 06 22:35:49 Our current, hacked ourself, build system uses busybox rather than init Oct 06 22:44:16 http://www.linux-boot.net/InitRD/ Oct 06 22:44:20 this might help Oct 06 22:44:29 i think kergoth is probably asleep Oct 06 22:44:31 :D Oct 06 22:45:15 its a decent outline/example anyway Oct 06 23:34:26 hello ! Oct 07 00:45:39 wow... gpe-image, opie-image and e-image are all broken in the current oz354fam083 branch... :( Oct 07 00:53:18 03koen 07org.oe.oz354fam083 * rfc70e15a... 10/: packages/gpe-bootsplash/files/cairofix.patch: add missing patch Oct 07 00:53:22 03koen 07org.oe.dev * r063f5fcd... 10/: packages/blueprobe/blueprobe_0.15.bb: add patch for h6300 support from #373 Oct 07 00:53:26 03jbowler 07org.oe.dev * r8cb9fe03... 10/: nslu2-kernel: integrate 2.6.14-rc3 patchset 4 Oct 07 01:03:43 morning Oct 07 01:33:33 morning all Oct 07 01:34:42 morning all Oct 07 01:34:44 too Oct 07 01:35:14 hi RP Oct 07 01:35:30 hi ade|desk Oct 07 01:35:45 anyone know who rday is ? Oct 07 01:40:30 ade|desk: robert day from the mailinglist Oct 07 01:40:34 ade|desk: why do you ask? Oct 07 01:43:46 ade|desk: I must say his inability to copy/paste commands from the wiki is astounding Oct 07 01:45:48 koen: I've just been reading scrollback - You misunderstand what I meant about being able to go back to easlier kernels Oct 07 01:46:25 koen: We can go back to any of the kernels I've ever written if we restore the approproate .bb and changes the urls to look in my archives directory Oct 07 01:46:57 so all the fuss about 'we have no working kernel!!!' was a bubble of hot air? Oct 07 01:47:25 No, it was becuase I couldn't find a kernel I was happy releasing for the latest release Oct 07 01:48:27 2.6.11 had loads of bugs which were fixed in later versions. Those later kernels had other bugs Oct 07 01:49:06 I found a version which fixed them all and called it the new stable series. Problem solved Oct 07 01:49:41 the one that is default in OE now has been rock solid for me, a lot more stable than 2.6.11 Oct 07 01:50:29 so can we put all the patches for that one in OE and mark it as stable? Oct 07 01:50:51 koen: It marked as stable and I'd much prefer the patches stay where they are. Oct 07 01:51:06 koen: i just thought that looking at the ml he was taking the piss with not reading what people have said and his inability to use google and the wiki Oct 07 01:51:12 The reason being I can upgrade both trees from the same patch sources Oct 07 01:51:31 and break the branch with the same source Oct 07 01:51:42 koen: I'd only apply bug fixes to both trees Oct 07 01:54:50 ade|desk: and copy pasting from console to mail also seems to be problematic Oct 07 01:55:23 RP: what is the problem with putting all the patches for the stable kernel into the branch? Oct 07 01:55:39 RP: the idea of that branch is that only mickeyl and I touch it Oct 07 01:56:22 koen: I fail to see the need. It just increases the size of the monotone repo and makes things more difficult for me (or you) to change Oct 07 01:56:27 koen: i can understand that, my oe, alas, has to be in vmware, c&p is not really possible Oct 07 01:56:52 ade|desk: you not seen colinux? Oct 07 01:57:19 XorA: corp policy :( Oct 07 01:57:43 ade|desk: sucks, if it wasnt for the syscall driver colinux wouldnt need admin at all Oct 07 01:57:48 If you want to do it on "your" branch, that's up to you and mickeyl. I will not be happy doing that in .dev though as it would badly affect my system of working. Oct 07 01:58:37 If you do it on your branch, patching any bugs will become your problem. Oct 07 01:58:46 it already is Oct 07 01:59:39 at the moment a vital part of OZ isn't managed in an SCM Oct 07 02:00:30 Its managed by a slightly different system that still maintains history Oct 07 02:01:24 Anyhow, my position is quite clear and its been shown to work. I can't see the point in dicussing this further Oct 07 02:02:42 and it has shown to fail at points Oct 07 02:03:11 When did it fail? Oct 07 02:03:25 when patches got overwritten Oct 07 02:04:31 I have once screwed up a couple of patches. It was a really minor problem and if it hadn't been, I could have restored backups Oct 07 02:05:05 The main issue was people had corrupt files in their sources direcotry rather than any data loss Oct 07 02:05:27 Which was easily fixed by a version bump Oct 07 02:17:15 morning Oct 07 02:18:31 hi reenoo_ Oct 07 02:22:48 hey RP Oct 07 02:28:15 * XorA needs a faster machine for nessus Oct 07 02:29:29 03koen 07org.oe.dev * r3c33d888... 10/: packages/libgpewidget/libgpewidget_cvs.bb: add (disabled) cvs version of libgpewidget Oct 07 02:38:13 http://pastebin.com/386013 Oct 07 02:38:17 what is wrong? Oct 07 02:40:08 lucasvo: it seems the build for your jornada failed Oct 07 02:44:23 hmm, familiar's giving me EOL while scanning single-quoted string syntax errors. Oct 07 02:44:48 hopefully that won't be an issue since I'm not building familiar Oct 07 02:46:01 * Arjan still can't build gst-plugins :( Oct 07 02:46:04 koen: jornada? Oct 07 02:46:52 lucasvo: yes Oct 07 02:47:06 linux-jlime-sh3-2.6.11 Oct 07 02:47:18 koen: but I didn't set it up for jornada Oct 07 02:49:38 koen: what do I have to setup in conf/local.conf? ipaq-px250 and arm isn't that correct for hx4700? Oct 07 02:50:06 you didn't paste the build summarry you get at the start Oct 07 02:50:20 and ipaq-px250 is wrong for two reasons: Oct 07 02:50:24 1) it doesn't exist Oct 07 02:50:41 2) ipaq-pxa270 is the one for hx4700 Oct 07 02:50:42 koen: and what is for hx4700 Oct 07 02:50:44 aha Oct 07 02:50:51 you can see that in conf/machine/*.conf **** ENDING LOGGING AT Fri Oct 07 02:59:56 2005