**** BEGIN LOGGING AT Fri Mar 27 02:59:57 2009 Mar 27 07:42:11 i am unable to compile gimp, i get the error: http://pastebin.ca/1373787 Mar 27 07:42:45 it is looking for some definitions missing in libavcodec.h Mar 27 07:42:53 avcodec.h Mar 27 07:43:06 has any of you faced this issue before Mar 27 08:37:06 good morning Mar 27 08:39:29 hi Mar 27 09:37:32 03Steffen Sledz  07org.openembedded.dev * rc133a97e0c 10openembedded.git/recipes/linux/linux-2.6.24/oxnas/oxnas.diff: linux-2.6.24: tiny jffs2 compile bug for OXNAS machine fixed Mar 27 09:41:18 gremlin[it]: hi, what about sheevaplug? Mar 27 09:46:04 mckoan: hi! Mar 27 09:51:25 mckoan, no news basically ... i think i'll start order two at time this weekend Mar 27 09:52:17 no chance to get shipment by the italian distributor then? Mar 27 09:53:30 valhalla, no, cause basically isn't sell from Marvell but from globalscale Mar 27 09:55:31 on elinux.com there is link to other similar device, one of them is from a Israel company ... no european at now Mar 27 09:55:38 yes indeed, and globalscale told me that buying >10 items the price won't change Mar 27 09:55:40 * ant_work read on Il Sole XXIV Ore 23.03.2009 a tech-report about Marvell and plug-computing Mar 27 09:55:46 mckoan: have you had a chance to test Qt 4.5.0 I commited to OE? Mar 27 09:56:16 jeremy_laine: I scheduled such test today Mar 27 09:56:25 then I guess that I'll order it directly, since there are already 2 of us in the local lug wanting one Mar 27 09:56:50 mckoan: ok cool. What I have tried: building apps against qt4-x11-free: OK Mar 27 09:57:05 mckoan: building and running apps against qt4-embedded: OK Mar 27 09:57:19 valhalla: unfortunately also if I would import many of them I must apply VAT and shipment expenses, so price doesn't change for the final user Mar 27 09:57:57 jeremy_laine: thats'a a cool news! Mar 27 09:58:32 jeremy_laine: I try qt4-x11-free now as first Mar 27 09:59:05 mckoan: for now, I have set DEFAULT_PREFERENCE = "-1" for all qt 4.5.0 packages (qt4-tools-native, qt4-x11-free, qt4-embedded) so you'll need to override that in your local.conf Mar 27 10:00:01 jeremy_laine: so if I want qt4-x11-free what do I have to do? Mar 27 10:00:40 mckoan: add to your local.conf: Mar 27 10:00:48 PREFERRED_VERSION_qt4-tools-native = "4.5.0" Mar 27 10:00:50 PREFERRED_VERSION_qt4-x11-free = "4.5.0" Mar 27 10:00:52 PREFERRED_VERSION_qt4-embedded = "4.5.0" Mar 27 10:01:17 mckoan: also, do you use "oestats" ? Mar 27 10:02:06 mckoan: it's a class that can be used to send build reports to http://tinderbox.openembedded.net/ Mar 27 10:02:07 jeremy_laine: never used Mar 27 10:02:08 valhalla, good ... we will try then to build an italian community for sheevaplug ;) Mar 27 10:02:41 mckoan: see http://wiki.openembedded.net/index.php/How_do_I_send_automatic_success_and_failure_reports Mar 27 10:02:45 gremlin[it]: I'll buy a couple of them for personal use (no pusher) Mar 27 10:03:16 mckoan, hahah ok, good ! Mar 27 10:03:19 jeremy_laine: sounds interesting, thx for the info Mar 27 10:03:54 mckoan: I'd be interested if you use it, it allows me to track what versions of Qt have been built / for what machine Mar 27 10:04:10 mckoan: for instance http://tinderbox.openembedded.net/packages/qt4-embedded/ Mar 27 10:06:27 does anybody know where xmms-tremor could be available? the URL in the recipe doesn't work anymore, and google doesn't help Mar 27 10:14:56 gremlin[it]: do you know how much is shipping cost to ITA for 1 sheevaplug ? Mar 27 10:15:26 mckoan, only one is about 31$ Mar 27 10:15:54 mckoan, i suggest to buy 2 ... 32$ for both Mar 27 10:16:11 gremlin[it]: US $106.83 for 10 units Mar 27 10:17:03 gremlin[it]: that is good, but you want to bet that customs won't apply VAT to small shipment ;-) Mar 27 10:17:04 mckoan, i know ... you know how much should be import tax + IVA tax ? Mar 27 10:17:20 VAT = 20% Mar 27 10:18:01 and if i remember correct should be 12% or 17% as import tax ... right ? Mar 27 10:18:13 AFAIK no Mar 27 10:18:34 just IVA ? Mar 27 10:19:05 yes Mar 27 10:19:56 mhh ok ... Mar 27 10:20:15 * mckoan is reading a customs document about last import from Taiwan Mar 27 10:22:11 gremlin[it]: should be 100 euro including VAT, but I have to ship them to you so add 15 euro + VAT Mar 27 10:23:22 gremlin[it]: I think you can get it at the same cost directly from US Mar 27 10:23:51 mckoan, yes ... Mar 27 11:15:57 Hello everyone. Can someone here please give me some information on the maemo recipes in the OE tree? as in, their status: is the maemo integration on the N770 device complete or are they skewed bb files as of now ? Mar 27 11:19:53 which package will give me a file manager Mar 27 11:53:20 morning Mar 27 11:55:56 hi hrw Mar 27 11:56:46 rkirti: Oe do not contain whole maemo Mar 27 11:57:05 rkirti: and 770 (not n770) maemo support was iirc dropped long time ago Mar 27 11:59:05 hrw, true, maemo isnt wholly contained. Florian had placed OE integration as one of the GSoC ideas and I am interested in that. I was asking this to know if I could use the existing bb files and build on top of that work or if support for n770 was complete so that I could replicate it for another device. Mar 27 11:59:23 rkirti: gsoc for maemo? Mar 27 11:59:48 hrw, yes. for maemo Mar 27 11:59:52 link? Mar 27 12:00:12 hrw, http://wiki.maemo.org/GSoC_2009/Project_ideas Mar 27 12:01:04 I see Mar 27 12:01:22 rkirti: maemo 2008 'diablo' recipes are available in experimental Poky branch Mar 27 12:01:29 hrw, I had spoken about this to Florian but thought it might be a better idea to aks if anyone on the channel knew about the status of the work Koen and Florian did in 2005 rather than pester those busy devels with my naive queries :( Mar 27 12:01:44 hrw, Ok, thanks, will take a look at those. Mar 27 12:01:45 rkirti: http://git.pokylinux.org/cgit.cgi/experimental/meta-maemo/ - with it you can boot to basic hildon-desktop Mar 27 12:01:59 rkirti: I worked on those recipes some months ago :) Mar 27 12:02:13 rkirti: if you will get them merged into OE I would be happy :) Mar 27 12:02:29 but no support from me for them rather as I am not interested in maemo anymore Mar 27 12:03:00 kdepimpi idea should get killed. that application was ugly fork Mar 27 12:03:54 hrw, :) My vested interests are on the OE side as well :P The only reason I want to take up this GSoC project is that I have been working with OE for the Atmel eval boards and want to move from OE user to Oe contributor. Mar 27 12:04:19 ;) Mar 27 12:11:36 * keesj is tempted to pick up this maemo task pretty soon Mar 27 12:23:41 does COMPATIBLE_MACHINE work in u-boot recipes? Mar 27 12:27:26 are there any docs which say what packages does what? Mar 27 12:27:47 i know i need some feature, but not sure which package i need to use Mar 27 12:27:53 Crofton|work: should do, if it doesnt its a bug Mar 27 12:28:28 I'm going to push a recipe for u-boot-2009.3 Mar 27 12:28:48 and mark it as COMPATIBEL_MACHINE for boards I have tested Mar 27 12:28:54 and let other s add their boards Mar 27 12:29:10 good plan Mar 27 12:29:14 or, should I mark it DEF_PREF="-1_ and use prefered version Mar 27 12:29:33 the problem is, I would need to mark the git redicpe DEF_PREF=-1 Mar 27 12:29:37 I like the COMPATIBLE_MACHINE idea, it tells people what is actually tested Mar 27 12:29:44 me to Mar 27 12:31:08 maybe put an explanation in the commit message Mar 27 12:31:14 some people sometime read them :-) Mar 27 12:31:55 Crofton|work: compat machine seems nice and elegent. No messing about and clear, we know it works on on your own head messages ;-). Mar 27 12:33:20 I've updated the openembedded dev branche just yet, and noticed qt-embedded 4.5 now has a recipe. However, it wants to 'bitbake' both 4.4.3 as well as 4.5 Mar 27 12:33:26 is this normal? Mar 27 12:33:39 I will break people depending on the git recipe sorting the highest Mar 27 12:34:12 <_cpo_> i have a SIL3811 SATA<->PATA converter connected to a sata port of one of my machines Mar 27 12:34:34 <_cpo_> anybody does know how to get this converter to work under linux? because my CF card that is connected to this chip is now recognized by the kernel Mar 27 12:34:43 <_cpo_> (i know this is a bit offtopic here) Mar 27 12:35:31 I'll think how to push with minimum breakage Mar 27 12:35:35 _cpo_: such converters iirc are transparent Mar 27 12:35:39 then work on the git recipe being the default :) Mar 27 12:43:27 Crofton|work: what's wrong with SRCREVs? Mar 27 12:43:45 in the u-boot git recipe? Mar 27 12:43:53 yep Mar 27 12:44:01 why add another .bb? Mar 27 12:44:11 The original comment suggested using the _git recipe to build 2009.3 Mar 27 12:44:27 so you tet for that SRC_REV Mar 27 12:44:48 then in a few months, someone else changes the rev to test some new 733t u-boot feature Mar 27 12:45:00 and messes it up :) Mar 27 12:45:33 there are lots of boards still using old u-boot's Mar 27 12:45:46 but I have some boards that were only recently supported Mar 27 12:45:51 well, without patches Mar 27 12:46:04 the point is to provide stable builds Mar 27 12:46:36 the git recipes, y design, do not provide a stable user experience :) Mar 27 12:46:44 <_cpo_> hrw: i thought so, too but during boot it gets i/o timeouts Mar 27 12:46:47 I'll see if u-boot for Zaurus (very old SRCREV) is still ok (_git) Mar 27 12:46:51 later on Mar 27 12:47:17 no hurry if you aren't having trouble :) Mar 27 12:47:37 btw: there were a patch floating in the ML for auto-extraction of OFFSET and ADDR Mar 27 12:47:51 abybody tested it? Mar 27 12:48:12 s/a patch/patches/ Mar 27 12:48:41 dunno Mar 27 12:48:52 scary stuff :) Mar 27 12:49:17 ok, I volunteer then Mar 27 12:49:19 ;-) Mar 27 12:50:23 ah, and possibly the mkimage-openmoko crap is away... Mar 27 12:50:52 really, t's long I don't play u-boot ... Mar 27 12:51:04 are there any docs which specify which bb file does what? Mar 27 12:53:35 Crofton|work: no, in kernel.bbclass is still depends = "%s u-boot-mkimage-openmoko-native" % depends Mar 27 12:54:01 that is to make mkimage Mar 27 12:54:01 I thought hrw sanitized this Mar 27 12:54:18 I made a u-boot-tools recipe at some point Mar 27 12:55:07 see, we have u-boot-mkimage-native_1.3.2.bb Mar 27 12:57:03 why should I download tons of OM crap for my machine? Mar 27 12:57:26 I think the reicpe is poorly names Mar 27 12:57:39 I do not think it leads to tons of om stuff Mar 27 12:57:46 ;-) Mar 27 12:57:52 just SRC_URI Mar 27 12:58:12 but you got the point Mar 27 12:58:19 yeah Mar 27 13:04:30 om one is overpatched Mar 27 13:04:40 1,3,2 one is sane Mar 27 13:04:56 the om one was just the first one, thats all Mar 27 13:05:07 its fine to downgrade it and use later better ones Mar 27 13:07:32 btw why aren't the older version tagged and merged in uboot_git? Mar 27 13:07:46 no one got around to it Mar 27 13:07:49 thre are 7 + _git recipes! Mar 27 13:08:08 so? Mar 27 13:08:53 he he...anyway is not as bad as /linux dir :-) Mar 27 13:09:00 no Mar 27 13:09:01 (overpopulated too) Mar 27 13:09:24 I should talk with some people abobut the omap recipes Mar 27 13:11:34 the problem with stuff like linux/ and u-boot/ is someone pops up makes it work with their board, then dissapears Mar 27 13:11:59 btw do I understand right, the 'optimal' way to add recipes/patches is creating a specific subdir and not xy-override_machine? Mar 27 13:12:59 <_cpo_> hrw: ok its really transparent ... the failure was a misconfigured bios Mar 27 13:14:41 XorA: now that you're the first tosa-2.6.29 on the world, would you help me to finally close the packaging issues with zaurus-updater? Mar 27 13:15:11 don't tell me you had any during build ;-) Mar 27 13:16:37 I for me would build it as separate package and not as EXTRA_IMAGEDEPENDS Mar 27 13:17:46 thesing proposed the same (throw away the installkits) but koen was against Mar 27 13:18:05 you see..no maintainers..anarchy Mar 27 13:18:36 good morning Mar 27 13:21:25 kgilmer, gm Mar 27 13:21:37 how are things in bugworld Mar 27 13:22:27 pretty good Crofton|work :) I've somehow figured out how to get some time to write documentation. Mar 27 13:22:49 masochist Mar 27 13:22:53 so, i'm starting over with our wiki :) Mar 27 13:23:00 most of us spend our time figuring out how not to write docs :) Mar 27 13:23:04 haha Mar 27 13:23:36 I am particularly slow at writing Mar 27 13:23:36 well, i've found that being able to say "for info on that go here..." is faster than having a long conversation about the same things over and over. Mar 27 13:24:09 and most of our wiki was written by... QA people Mar 27 13:24:16 which is why i'm taking a chainsaw to it. Mar 27 13:24:38 * kgilmer hopes none of those qa people are listening Mar 27 13:24:49 heh Mar 27 13:25:07 how are things on your end Crofton|work Mar 27 13:25:08 ? Mar 27 13:25:35 good Mar 27 13:25:51 plenty to keep me busy and some is even paying a little Mar 27 13:25:56 <_cpo_> anybody ever tried building xserver 1.6 with OE? Mar 27 13:26:28 <_cpo_> I'm getting a libtool No such file or... error, but libtool/libtool-cross/libtool-native is build and staged Mar 27 13:26:44 good to hear Crofton|work Mar 27 13:32:14 tsjsieb: sent you an email -- let me know if it does not make sense Mar 27 13:34:56 cbrake: Thank you very much, I think I can get further with this :) Mar 27 13:43:50 For my project I use directfb-1.0.0 (not 1.2.7). That's why I have PREFERRED_VERSION_directfb = "1.0.0" in my machine config file. I'd like to apply project-specific patch for directfb. For this, I copy directfb_1.0.0.bb, directfb.inc and files/ directory into my overlay directory, modify bitbake for patch applying, update PR variable. But somehow OE build system doesn't compile mine "1.0.0-r7" package but generic "1.0.0-r6" one. WTF? If I build directfb wi Mar 27 13:43:55 th 'bitbake directfb-1.0.0-r7' then OE compiles the proper recipe version but not for generic building through deps Mar 27 13:44:14 and directfb-1.0.0-r7 somehow depends on directfb-1.0.0-r6 :-\ Mar 27 13:49:03 hi guys Mar 27 13:49:28 I am trying to compile x264 with OE. I put in .bb in EXTRA_OECONF = "--extra-cflags=\"-O3 -Wall\"" .. but when I try to build the package using bitbake ./configure takes only the first argument .. Here it will take only -O3 and say Unknown option -Wall", ignored (in log.configure) Mar 27 13:49:32 If I ./configure it from the command line (without bitbake).. --extra-cflags can take lots of arguments... Mar 27 14:02:17 ~curse n810 Mar 27 14:02:18 May you be reincarnated as a Windows XP administrator, n810 ! Mar 27 14:03:56 hrw, maybe you can help with my directfb problem? Seems that you are one of the people who understands the inners of OE... Mar 27 14:15:33 well u-boot 2009.3 does not work on the OSK Mar 27 14:17:03 Sinky: how would you type it from the command line. Can you use pastebin.ca to show me? Mar 27 14:18:44 ;) Mar 27 14:19:16 booxter: read about bitbake collections. you can set priorities for overlay Mar 27 14:38:33 my gdk.pc fails a sanity check, which library generates gdk.pc Mar 27 14:46:15 I hate people who ask a question and quit 2 minutes later Mar 27 14:47:19 XorA, I am not defending him , but he could be facing a bad net connection. Mar 27 14:47:38 rkirti: then you get a different message Mar 27 14:48:38 XorA, ah I see. I am sorry then, my fault. [/me hasnt used web-based irc clients toom much] Mar 27 14:50:22 rkirti: get it a lot from non web clients as well, people seem to think we are awake every millisecond :-) Mar 27 14:53:11 is it false ? :) Mar 27 14:53:36 starox: very false :-) we need to drink beer occasionally Mar 27 14:53:42 XorA, I ld agree, and I feel very sorry for the same. [However, an irrelevant note here - in institutions like ours, our LUGs have instilled a *deep* respect in us for devels, so we do try and ensure people dont do such things, atleast the Linux users we know.] Mar 27 14:54:27 rkirti: its more annoying that we cant help them because they left :-) Mar 27 15:06:50 midori and epiphany are displaying both messages like this: http://pastebin.com/d2a462bc . any idea? Mar 27 15:07:28 in beagleboard using angstrom Mar 27 15:07:44 no idea, its all foreign to me :-) Mar 27 15:07:54 morning Mar 27 15:08:04 morning Mar 27 15:12:44 recalcati_oe: missin java stuff ? Mar 27 15:17:12 g'day kergoth Mar 27 15:17:29 java stuff ? Mar 27 15:17:36 really ? I need it ? Mar 27 15:17:52 phoneme or similar ? Mar 27 15:18:25 recalcati_oe: try opening a simpler page as first Mar 27 15:18:35 I remember that with oe commit 768fe337b086a509a1381f9b1000faf1e99689cd midori worked Mar 27 15:18:40 isn't this about JavaScript? Mar 27 15:19:04 mckoan: now I have rebuild all again, it take be some minutes to try Mar 27 15:19:25 that isn't java, it's a mixture of javascript and css Mar 27 15:20:38 you don't, luckily, need java to render the google page :-) Mar 27 15:21:10 (actually I don't think you actually need javascript either, but you do need to be able to at least parse the html correctly and that seems to be where your browsers are falling down.) Mar 27 15:21:11 pb_: yes, but that messages are not nice to see inside the web page Mar 27 15:21:56 right, but that's nothing to do with a missing anything. Mar 27 15:22:16 ./usr/lib/libwebkit-1.0.so.2.1.0 Mar 27 15:22:43 you can try turning java and javascript off in firefox (or whatever you use on your pc) and visiting google; it should render just fine without those things. Mar 27 15:23:20 pb_: as soon as possible I try Mar 27 15:23:36 oh, you're using the webkit-based epiphany? Mar 27 15:24:16 I thoght this, but I haven't really verified Mar 27 15:24:34 it'd be worth checking Mar 27 15:24:52 if you are indeed using epiphany-webkit, then that would indicate fairly strongly that webkit is at fault Mar 27 15:24:57 openembedded/recipes/gnome/epiphany_2.24.2.bb Mar 27 15:25:11 if you're using epiphany-gecko, then that would be more of a mystery; it would be unlikely for webkit and gecko to both be broken in the same way. Mar 27 15:25:12 http://www.linuxdevices.com/news/NS7648514863.html Mar 27 15:25:20 yes --with-engine=webkit Mar 27 15:25:23 We need some "branding" guidelines Mar 27 15:26:18 recalcati_oe: try with gecko, see if that works any better Mar 27 15:26:40 http://www.linuxdevices.com/cgi-bin/survey/survey.cgi?view=archive&id=1130200716721 Mar 27 15:27:14 compare 7 and 8 Mar 27 15:31:53 find openembedded | grep gecko Mar 27 15:31:57 not present Mar 27 15:34:16 03Robert Schuster  07org.openembedded.dev * r1965abdae3 10openembedded.git/conf/checksums.ini: checksums.ini: Added toppler 1.1.3 entries. Mar 27 15:34:27 03Robert Schuster  07org.openembedded.dev * rf3fdd55a8f 10openembedded.git/recipes/toppler/toppler_1.1.3.bb: toppler 1.1.3: New recipe. Mar 27 15:34:56 google work on webkit epiphany here Mar 27 15:35:10 wonder if its a proxy or localised google version problem Mar 27 15:35:30 oe gti hash c133a97e0c262fb707eecbc79a41cececb391e5a Mar 27 15:36:15 from my pc I see correctly Mar 27 15:36:19 from firefox Mar 27 15:37:01 recalcati_oe: gecko isn't a separate package. I meant to build epiphany with the gecko engine. Mar 27 15:37:07 also from ubuntu 8.10 and epiphany-webkit www.google.it works Mar 27 15:37:21 pb_: sorry I didn't know Mar 27 15:40:11 "OpenEmbedded Angstrom Linux" :) Mar 27 15:40:47 odd, www.google.it works on epiphany webkit here, the standard Angstrom version Mar 27 15:40:53 Crofton|work: branding guidelines sounds good Mar 27 15:40:58 ok, epiphany, 2.24.2, can see very simple html pages Mar 27 15:43:03 instead I've opened a google document, and it appears all with textual strange messages Mar 27 15:44:28 mickeyl: good morning! Mar 27 15:44:44 * mwester wonders what the weekly "break OE" commit will be for this weekend... and wonders if he should think about doing it if nobody else has. Mar 27 15:44:45 good morning pb_ Mar 27 15:44:47 hi folks Mar 27 15:45:06 mwester: go ahead, everyone is entitled to do a share ;) Mar 27 15:45:17 cbrake: nice work w/ the news Mar 27 15:45:32 But I live in fear of breaking OE! :-D Mar 27 15:45:33 cbrake: do we have to take care that it's not getting to big or will old ones be autopurged? Mar 27 15:45:40 mickeyl: thanks Mar 27 15:45:57 mwester: just put on your flameproof suit and do it :D Mar 27 15:45:59 mickeyl: its just a wiki page, so when we add new ones, we simply remove all ones for now Mar 27 15:46:04 cbrake: ok Mar 27 15:46:08 * mickeyl just added one Mar 27 15:46:09 mickeyl: perhaps we should have an news archive page Mar 27 15:46:18 cbrake: yes, that sounds good Mar 27 15:46:53 god i love git-bisect Mar 27 15:46:56 ♥ Mar 27 15:47:29 hehe Mar 27 15:48:02 I guess it's a matter of perspective -- rather like the "jaws of life", I rather dislike the use of them -- but admit that the tool is great when you might need it. Mar 27 15:49:26 03Ihar Hrachyshka  07org.openembedded.dev * r413872c423 10openembedded.git/ (conf/checksums.ini recipes/libzip/libzip_0.9.bb): Mar 27 15:49:27 libzip: new recipe Mar 27 15:49:27 This is a C library for reading, creating, and modifying zip archives. Mar 27 15:49:27 Package is created for libzip-0.9 Mar 27 15:49:32 mickeyl, cbrake: what about making 'recent news' a box like we have with users/develoeprs etc? Mar 27 15:49:38 bisect has saved me time on a number of occasions. definately special purpose, but lovely Mar 27 15:50:08 mickeyl, cbrake: and maybe we can make this list point to longer versions on our site too Mar 27 15:52:12 firefox, also with java and javascipt disabled, see www.google.it nicely, instead epiphany not. Also epiphany seems to have problem with css in www.mozilla.org Mar 27 15:52:56 khem, you got anything for doing gcc 4.4 stuff w/ OE? Mar 27 15:53:07 opkg list_installed | grep webkit Mar 27 15:53:07 libwebkit-1.0-1 - 0.1+svnr40814-r8.1 - Mar 27 15:53:12 instead epiphany on my pc see correctly www.google.it and www.mozilla.org Mar 27 15:53:15 is what I have here are seems to be working fine Mar 27 15:53:37 libqtwebkit4 - 4.4.3-r10.1 - libwebkit-1.0-2 - 0.1+svnr41827-r9.1 - Mar 27 15:54:08 why I have libqtwebkit4 , strange Mar 27 15:54:54 mickeyl: perhaps you remember...I see in 'minimal' DEPLOY_DIR is weak-assigned. So if the user does not set this var in local.conf, the value from bitbake.conf is taken (DEPLOY_DIR ?= "${TMPDIR}/deploy") without *libc suffix. All fine with this? Mar 27 15:56:20 I have addded qt4-x11-free Mar 27 15:56:38 ant_work: well, i'd appreciate if we could handle this better Mar 27 15:57:01 ant_work: like, both having the default to the libc, but allowing users to override Mar 27 15:57:04 XorA: thx, I go back to libwebkit-1.0-1 - 0.1+svnr40814-r8.1 - Mar 27 15:57:43 ant_work: perhaps we need to intruduce something like USER_DEPLOY_DIR Mar 27 15:57:48 similar to USER_DISTRO Mar 27 15:57:57 we're discussing this on ML Mar 27 15:58:06 seems sane move Mar 27 15:58:45 there is however the isssue with base.bbclass ... Mar 27 15:59:40 (was Was: package-stagefile-shell() breaks if DEPLOY_DIR is outside of TMPDIR) Mar 27 16:00:12 yeah, i have removed this in my kernel.bb Mar 27 16:00:16 I don't know if this is a corner-case, however I stumble *every time* in it Mar 27 16:00:19 did not have time to investigate Mar 27 16:01:03 possibly is on kergot's to-do (bottom of the list ;-) Mar 27 16:01:04 i carry this two-line patch locally now Mar 27 16:01:10 until someone fixes it for goodf Mar 27 16:16:11 mickeyl: may I refresh your memory about making a new standalone package zope-interface ? Mar 27 16:16:47 ah, right Mar 27 16:16:54 let me start a build and then i'll take a loko at that Mar 27 16:18:13 s/loko/look Mar 27 16:18:56 mickeyl: thanks alot! Mar 27 16:21:55 03Michael 'Mickey' Lauer  07org.openembedded.dev * r8b70868d1f 10openembedded.git/recipes/linux/ (linux-msm7xxxx/htcraphael/defconfig linux-msm7xxxx_git.bb): linux-msm7xxx: enable framebuffer console rotation for htcraphael Mar 27 16:28:16 Is it possible to get an anonymous UK PAYG Internet enabled Linux phone? Mar 27 16:28:26 I am trying to sort out a PAYG Skype mobile (UK). I notice 3 does one, but it has cobbled Skype so that you cannot Skype-call any UK mobile number (unless they are using Skype also). Which kind of defeats the purpose - all of my contacts are in my Skype address book. Does anyone know a way to get full Skype functionality? Mar 27 16:28:35 The part I am having trouble getting my head around is: how to get the Internet i.e. hook up with the provider? would I need to get something like an O2 SIM? I notice that 3 network are packaging everything together... you MUST use their phone if you want their service. Mar 27 16:28:57 mibbyte: ? Mar 27 16:29:27 mibbyte: uh, you're way off topic Mar 27 16:29:59 Sorry, someone directed me from #handhelds Mar 27 16:30:07 mibbyte: Ill answer you private Mar 27 16:30:21 Fantastic, thanks Mar 27 16:30:22 there are no open linux phones but the Openmoko ones Mar 27 16:30:44 although if you're a kernel guy, you might want to help us with htclinux and openezx Mar 27 16:30:58 only the a910 has wifi though Mar 27 16:31:46 mickeyl: htcdiamond/htcraphael lack working wifi under linux? Mar 27 16:32:41 hrw: I dont think any of the qualcomm series have wifi yet Mar 27 16:33:33 XorA: obviously findind the recipe and the version of epiphany/midori good for libwebkit-1.0-1 - 0.1+svnr40814-r8.1 is quite difficult. Mar 27 16:39:22 hrw: no, the a910 is the only MOTO one with wifi Mar 27 16:39:26 ok Mar 27 16:39:29 most of the qualcomm series have wifi Mar 27 16:39:34 although it doesn't work on linux yet Mar 27 16:39:36 :) Mar 27 16:49:17 ok, I sent summary of #oe-stable discussion to mailing list Mar 27 16:49:24 please read and reply Mar 27 16:50:10 thanks Mar 27 16:50:25 eFfeM: should i build zope-interface out of the same sourcefile or is there a lighter one that just contains the interface? Mar 27 16:50:32 162 lines so long read ;D Mar 27 16:50:42 hrw when? which thread? I don't see it Mar 27 16:51:05 mckoan: just sent, as reply to meeting mail Mar 27 16:51:33 hrw: maybe mailserver require some time Mar 27 16:51:58 hrw: under "IRC discussion about stable branch" ? Mar 27 16:52:53 yes Mar 27 16:53:09 * mwester has it Mar 27 16:53:11 :) Mar 27 16:53:31 Bah, I hate packaging Mar 27 16:53:47 I dunno why, but I've got an ubuntu chroot where debs don't work, rpm doesn't work Mar 27 16:53:58 lets hope ipk works so I can get back to what I wanted to do here, make an mx31pdk conf file Mar 27 16:53:59 * mckoan is reading an article on italian magazine talking about OE :-D page #3 http://www.ilb2b.it/eo_news/pdf_rivista/EONEWS_510.pdf Mar 27 16:55:03 can you use a shell command in a conf file (ie to obtain data for a variable assignment) and if so what is the syntax? Mar 27 16:56:11 you can use inline python, not inline shell Mar 27 16:56:49 what is the syntax for that? ${@script...}? Mar 27 16:57:54 mickeyl there is a lighter one Mar 27 16:57:58 * eFfeM was afk for dinneer Mar 27 16:58:00 WII Mar 27 16:58:32 eFfeM: got it Mar 27 16:58:46 kergoth, ah... I see, same syntax as in bb files - thx Mar 27 16:59:00 http://pypi.python.org/pypi/zope.interface Mar 27 16:59:03 tharvey: bitbake.conf have a number of examples :P Mar 27 17:01:57 ya, looking at those thx that opens up a whole new world :) Mar 27 17:02:07 hehe Mar 27 17:02:08 np Mar 27 17:02:27 mckoan: I read it, nice Mar 27 17:03:39 see you guys. have a nice weekend Mar 27 17:04:01 rschuster1, hi I'd like to try to make the new wesnoth recipe but: 1)which boost lib I should use 2)which build system should I use?(automake or cmake or scons,gentoo uses cmake but on wesnoth they told me long time ago that they wouldn't change the build system but it was long time ago...) Mar 27 17:04:40 recalcati_oe: thanks ;-) Mar 27 17:06:39 Gnutoo: cool Mar 27 17:06:52 Gnutoo: you can link against latest boost Mar 27 17:07:03 rschuster1, wow so boost-36 thanks Mar 27 17:07:14 Gnutoo: however you need to get boost-iostreams-bzip2 and boost-iostream-gzip working Mar 27 17:07:22 I know Mar 27 17:07:30 this is excluded in the current recipe Mar 27 17:07:43 yes but can I change boost-36's recipe...they work Mar 27 17:07:49 I got it to compile in 1.34 once upon a time Mar 27 17:07:53 Gnutoo: excellent Mar 27 17:07:59 please push :) Mar 27 17:08:11 ok wow...with or without review? Mar 27 17:08:11 for wesnoth 1.6 I would say go for cmake Mar 27 17:08:38 Gnutoo: can you pastebin the change? Mar 27 17:09:00 I have the boost-36 fixes in bugzilla...but as I was told that boost needed to make the 3 boost recipes installed at the same time I didn't commit Mar 27 17:09:06 ok Mar 27 17:09:15 let me find it Mar 27 17:09:41 yeah, it would be great to be able to install different boost versions at the same time Mar 27 17:10:00 but this larger work should not hinder smaller fixes Mar 27 17:11:05 the bug is at http://bugs.openembedded.net/show_bug.cgi?id=4977 I'll diff it Mar 27 17:13:08 bye Mar 27 17:15:44 03Steffen Sledz  07org.openembedded.dev * rbb9b5dd8f3 10openembedded.git/ (10 files in 5 dirs): hipox: new machine added (derived from oxnas) Mar 27 17:16:43 here are the recipes diff: http://rafb.net/p/fnrcEO96.html i'll pastebin the patches Mar 27 17:17:11 of course it need to be polished before commit Mar 27 17:17:49 03Michael 'Mickey' Lauer  07org.openembedded.dev * r75b82c6cc3 10openembedded.git/recipes/ (5 files in 2 dirs): Mar 27 17:17:49 zope-interfaces: new recipe; standalone Zope interfaces for extension products Mar 27 17:17:49 adjust main zope recipe accordingly Mar 27 17:17:50 eFfeM: this should be it. Mar 27 17:20:27 here are the patches: http://rafb.net/p/ONDS0S22.html Mar 27 17:20:57 rschuster1, ^^^ Mar 27 17:21:39 Gnutoo: I'll look at them later. Busy here :) Mar 27 17:21:44 thanks alot mickey|sports Mar 27 17:22:30 ok thanks a lot Mar 27 17:49:08 some problems in webkit-gtk passing from 768fe337b086a509a1381f9b1000faf1e99689cd oe hash to c133a97e0c262fb707eecbc79a41cececb391e5a oe hash Mar 27 18:47:33 hello is 1.6.0a.bb correct or should I use 1.6.0+1.6.0a ? Mar 27 19:21:18 Bah Mar 27 19:21:28 Anyone recall what python sqlite stuff hrw added for rpm? Mar 27 19:21:33 I thought he put in a package, but I don't see it Mar 27 20:30:30 khem, http://www.pastebin.ca/1374287 - kernel-headers patch - this time suitable for building kernel modules against Mar 27 20:33:26 khem, I should post this to the oe list for RFC eh? Mar 27 21:00:24 tharvey: yeah post it to ml you wil have more eyeballs to look at it Mar 27 21:01:03 will do Mar 27 21:01:15 patch looks ok Mar 27 21:27:33 any opinion on my creating an autotools_base class, which basically just updates the config.sub,config.guess, and runs oe_runconf.. basically autotools but not regenerating w/ autoreconf. quite a few packages do that manually Mar 27 21:32:01 kergoth: updates how? Mar 27 21:32:52 right now autoreconf calls gnu-configize. it can be called manually Mar 27 21:33:07 alternatively, the gnu-configize perl script can be replaced with about 10 lines of shell. Mar 27 21:49:15 kergoth: sounds ok. Regenerating should only be done on need bases Mar 27 21:49:46 well, we sort of have to regenerate for most things, to get our libtool modifications, for example Mar 27 21:50:52 he khem Mar 27 21:51:03 * kergoth is trying to bring our autotools bits a bit closer to upstream, less deviation would be good.. and to simplify what we're going a bit Mar 27 22:01:31 kergoth: yeah for libtool case I feel we need to keep regeration part handy Mar 27 22:01:34 woglinde: hi Mar 27 22:01:42 it would be nice if we could avoid it.. Mar 27 22:01:47 * kergoth will have to think about that Mar 27 22:02:11 regeration ? Mar 27 22:02:30 one thing that might be beneficial, is to at least avoid doing --force on the autoreconf Mar 27 22:02:49 every package uses a different version of autotools libtool etc. Mar 27 22:02:49 but itd be nice to avoid regeneration completely.. can't do it if we have any changes to the .m4 files though Mar 27 22:03:18 we normally patch projects to work with 2.5x autoconf and 1.7+ automake or so Mar 27 22:03:33 To avoid regeneration we have to patch Mar 27 22:03:52 that's true, we could patch generated files.. but thats pretty gross :) Mar 27 22:03:59 still, if itd speed things up at all, it might be worth considering Mar 27 22:05:34 course, I'm sure its a statistically insignficant bit of processing Mar 27 22:15:48 hey XorA Mar 27 22:16:13 hey Mar 27 22:16:36 hope you were not referring to me before (fire question, disappear :) Mar 27 22:16:51 big boss was calling... Mar 27 22:16:57 heh no Mar 27 22:17:22 I miss my Lafer's utilities Mar 27 22:17:50 dunno if you can remember these ... Mar 27 22:20:32 name doesnt ring a bell Mar 27 22:20:35 http://www.halfbakery.com/idea/Boss_20pedal -> Mar 27 22:20:36 hmm Mar 27 22:20:47 Leisure Suit Larry had the best boss key.. (pie chart of 3rd quarter condom sales or something) Mar 27 22:20:47 — Mr Burns, Oct 15 2002 Mar 27 22:20:47 Mar 27 22:20:53 haha Mar 27 22:20:57 hehe Mar 27 22:21:06 i miss the old sierra adventure games :( Mar 27 22:21:10 interent is evil... Mar 27 22:21:17 Roger Wilco, space guy.. Mar 27 22:21:22 words never disappear Mar 27 22:21:46 ZackMcCracken (sp) Mar 27 22:22:12 btw I had today to 'content-filter' my office...somebody is playing with Facebook to long...;-) Mar 27 22:22:18 hehe Mar 27 22:22:46 an old compnay i was at inadvertantly blocked wikipedia for like a week... engineers were not happy.. Mar 27 22:23:01 innocent employees have no idea of firewall syslogs... Mar 27 22:23:52 http://www.allowe.com/Larry/BOB.htm .. wtf Mar 27 22:24:23 about firewalls and routers...it looks like Broadcom 63xx is almost supported by OpenWrt... Mar 27 22:24:46 any interest to speed things a bit with the help of our resident hackers? Mar 27 22:25:16 * ant__ has two crappy routers sitting on a desk Mar 27 22:25:55 * ant__ already bricked a first one...was just too early Mar 27 22:26:45 ugh, so many recipes duplicate stuff because they don't use .inc's well Mar 27 22:26:51 this is why i advocate proper maintainership. Mar 27 22:29:15 kergoth: would you put S = "${WORKDIR}/xy-${PV}" in the .inc as rule of thumb? Mar 27 22:29:31 I mean S= blah Mar 27 22:29:39 not necessary Mar 27 22:29:57 the default S in bitbake.conf uses ${BP}, which uses the base package name (-native/-cross removed) Mar 27 22:30:04 hm..I see some cleaning-space in our very own kexecboot recipe Mar 27 22:30:06 don't need to set it in the recipe at all unless its something unusual Mar 27 22:30:16 most recipes could use some cleanup for current best practices Mar 27 22:30:18 the_ git uses different S = Mar 27 22:30:29 but they really need indidividual attention imo, not mass changes Mar 27 22:30:33 individual, even Mar 27 22:30:40 I see Mar 27 22:31:20 it's just my opinion, but i think when people make a change across a ton of recipes, they're likely to screw something up accidentally :) Mar 27 22:31:29 kergoth: I'll try without defining S...I sense it could just work Mar 27 22:31:30 hehe.. Mar 27 22:31:36 hello JaMa Mar 27 22:31:41 hello Jay7 Mar 27 22:31:46 both :-) Mar 27 22:31:56 * Jay7 some time ago has closed russian facebook's clone for one organization :) Mar 27 22:32:03 lol Mar 27 22:32:16 (by boss request) :) Mar 27 22:32:35 * Jay7 is BOFH some time :) Mar 27 22:32:47 hey ant__ :) Mar 27 22:33:05 hi jay7 Mar 27 22:33:18 I'm trying to continue my work on kexecboot.. Mar 27 22:33:23 hey woglinde :) Mar 27 22:34:02 woglinde: servus Mar 27 22:35:02 Jay7: have you found time to follow the zaurus-updater odyssey:- Mar 27 22:35:07 ) Mar 27 22:35:11 ? Mar 27 22:35:24 ant__: no :( Mar 27 22:35:43 it make all my image-builds fail Mar 27 22:35:55 I have to bitbake it before others Mar 27 22:36:09 then, it breaks on packaged-staging Mar 27 22:36:17 bringing kexecboot to buildable and working state is now my primary task (after commercial some) :) Mar 27 22:36:18 if yoy move deploy Mar 27 22:36:58 Jay7: here I've done like thesing suggested...zap Mar 27 22:37:17 don't move deploy :) symlink it :) Mar 27 22:37:59 he..workdir is in ram (tmpfs...) Mar 27 22:38:35 I need on disk-deploy Mar 27 22:38:43 create proper symlink when ajusting environment for bitbake :) Mar 27 22:39:07 could be a workaround, thx Mar 27 22:41:09 I propose to create a new image for installkits, where one can have zImages or uImages Mar 27 22:41:32 and get rid of updater.sh everywhere Mar 27 22:41:58 let see who will be the maintainer Mar 27 22:42:05 updater.sh is needed only to install kexecboot now Mar 27 22:42:13 that is Mar 27 22:43:39 for ordinary images would be great to have some installer image Mar 27 22:44:24 but it is needed only when installing into NAND Mar 27 22:47:00 well.. here is too late Mar 27 22:47:03 * Jay7 -> sleep Mar 27 22:47:30 nite jay7 Mar 27 22:53:16 ant__: made a task that checks if S can be removed from a recipe, and warns about it :P Mar 27 22:53:36 kergoth: thx. BTW how would you unify recipes when a git-tag is available? all under xy_git.bb? Mar 27 22:53:47 hmmm, not sure offhand Mar 27 22:53:58 let see kexecboot Mar 27 22:54:15 I see 0.4, 0.5, git Mar 27 22:54:39 the git recipe (done by me) has the older tags Mar 27 22:55:21 oh..in english... Mar 27 22:55:31 contains the tags of previous versions Mar 27 22:55:41 better now :-) Mar 27 23:00:40 * kergoth thinks about what other best practices for recipes can be easily tested for in a task Mar 27 23:01:07 please don't disturb kergoth for a while... Mar 27 23:01:12 :-) Mar 27 23:01:19 :) Mar 27 23:02:35 kergoth: gentoo has lately a message like ' package has poor programming...) Mar 27 23:02:46 cool Mar 27 23:03:24 : * QA Notice: Package has poor programming practices which may ... Mar 27 23:04:38 kergoth: *a lot* of ebuilds are emitting these QA :/ Mar 27 23:04:46 hehe Mar 27 23:04:54 very common ones Mar 27 23:04:56 guess they have similar problems to OpenEmbedded Mar 27 23:06:59 All distributions do, pretty much. Mar 27 23:07:16 * kergoth nods Mar 27 23:07:32 * kergoth still thinks more individual reponsibility would help matters Mar 27 23:11:00 kergoth: look at an example here: http://rafb.net/p/xNnqQ348.html Mar 27 23:11:08 they hint: report upstream Mar 27 23:12:43 not $DISTRO Bugzilla! Mar 27 23:14:04 broonie: any other distro requires you to compile it Mar 27 23:14:24 (antwell, others too...) Mar 27 23:14:52 I see the Gentoo-DNA in OE... Mar 27 23:15:33 then evolution did the rest :p Mar 27 23:37:09 ant__: In a lot of cases distros prefer to have reports go via the distro in case it's a local issue (eg, borked dependencies, a distro local patch). Mar 27 23:37:43 Though obviously that case is fairly clear about what to do :) Mar 27 23:38:53 right Mar 27 23:41:54 kergoth: what I find worth to copy is the idea of saving the screen logs (QA in a separate dir if you want) Mar 27 23:43:44 ok,very late here...good night Mar 28 00:05:41 good nite Mar 28 00:34:17 * kergoth chuckles, as far as he can tell, we can remove the override of the S variable from 5,172 .bb files, out of a total of 6,617 .bb files Mar 28 00:35:23 of course, a lot of those came in via .inc's, but still Mar 28 00:35:49 * kergoth double checks the sanity of his script Mar 28 00:37:44 ah, inverted logic. can remove it from 1445 .bb files Mar 28 00:50:21 heh **** ENDING LOGGING AT Sat Mar 28 02:59:57 2009