**** BEGIN LOGGING AT Mon Jul 28 02:59:56 2008 Jul 28 03:11:45 RP: was Gitosis considered for OE git access control? http://www.markrichman.com/2008/06/16/installing-git-and-gitosis-on-ubuntu/ Jul 28 03:17:10 RP: http://scie.nti.st/2007/11/14/hosting-git-repositories-the-easy-and-secure-way Jul 28 06:12:17 morning Jul 28 08:01:13 rwhitby: We're using gitosis Jul 28 08:03:07 03tmbinc 07org.oe.dreambox * r79513b92... 10/ (8 files in 4 dirs): libcdio: add libcdio Jul 28 08:03:10 03tmbinc 07org.oe.dreambox * r3bde4eaf... 10/ (6 files in 3 dirs): libcddb: add package Jul 28 08:04:52 bonjour Jul 28 08:08:08 03  07org.openembedded.dreambox * r03ef343c22 10OE.dev/packages/libcdio/ (7 files in 2 dirs): libcdio: add libcdio Jul 28 08:08:09 03  07org.openembedded.dreambox * r565ea242d1 10OE.dev/packages/libcddb/ (5 files): libcddb: add package Jul 28 09:19:43 hello there! :D Jul 28 09:37:22 florian: good morning Jul 28 09:37:28 lrg: good morning Jul 28 09:37:49 hey pb__ Jul 28 09:39:04 hi all Jul 28 09:54:59 morning Jul 28 09:55:16 http://blog.haerwu.biz/2008/07/28/i-am-0x20-now/ Jul 28 09:55:30 hrw: well done, you are catching up :-) Jul 28 09:55:43 o.. we have git cia too? Jul 28 10:15:44 hrw: good morning and congratulations Jul 28 10:16:29 thx Jul 28 10:22:52 ~lart wvstreams Jul 28 10:22:52 * ibot rm -rf's wvstreams Jul 28 11:14:44 hrw: yes, we do Jul 28 11:19:13 Time to start learning git :-} Jul 28 11:20:01 :-} Jul 28 11:20:23 still, look on the bright side, no more monotone :-) Jul 28 11:20:47 heh indeed :-) Jul 28 11:23:35 Who's an admin for the oe-commits list? Jul 28 11:23:53 * RP suspects he needs to talk to mickey|zzZZzz Jul 28 11:24:14 +1 from me ;) Jul 28 11:24:15 mention coffee, that tends to wake him Jul 28 11:33:23 mmm coffee good idea Jul 28 11:35:56 RP: anything I should know about gitosis before deploying it on git.nslu2-linux.org? Jul 28 11:36:16 (it seems to work out of the box) Jul 28 11:36:18 rwhitby: No, seems to work well for the uses I've put it to Jul 28 11:36:31 I'm sure I mentioned it in the instructions I shared with your admin Jul 28 11:36:45 yeah, I forgot to ask him :-) Jul 28 11:40:59 coffee Jul 28 11:41:29 Who will be my personally git assistant? Jul 28 11:41:55 * mwester-trip looks around, bleary-eyed and sleepy, for the coffee that someone mentioned... Jul 28 12:36:25 03thebohemian 07org.oe.dev * r2fb25ef6... 10/ (1 packages/kxml2/kxml2_2.3.0.bb): kxml2 2.3.0: Fixed SRC_URI. Jul 28 12:38:20 03  07org.openembedded.dev * rec09758c34 10OE.dev/: Jul 28 12:38:20 merge of '2fb25ef6fb3cb91ddaf8dd62069a538d6f50df58' Jul 28 12:38:20 and 'c62ede010b1359ffbadc321f747e6f44d9a83dd5' Jul 28 12:38:26 03  07org.openembedded.dev * r4501af159b 10OE.dev/packages/kxml2/kxml2_2.3.0.bb: kxml2 2.3.0: Fixed SRC_URI. Jul 28 12:38:27 03  07master * r4501af159b 10OE.dev/packages/kxml2/kxml2_2.3.0.bb: kxml2 2.3.0: Fixed SRC_URI. Jul 28 12:38:37 03  07master * rec09758c34 10OE.dev/: Jul 28 12:38:37 merge of '2fb25ef6fb3cb91ddaf8dd62069a538d6f50df58' Jul 28 12:38:37 and 'c62ede010b1359ffbadc321f747e6f44d9a83dd5' Jul 28 12:46:29 What package provides arm-angstrom-linux-gnueabi-gcc? I can't build gcc-cross because glibc can't find that executable. Jul 28 12:47:09 gcc-native maybe? Jul 28 12:47:11 on host or target? Jul 28 12:47:18 diego_at_work: On host Jul 28 12:47:27 gcc-cross Jul 28 12:47:37 hrw: Yeah, that's what I tried to compile. Jul 28 12:48:13 piroko: try to pastebin what are you trying to do ... Jul 28 12:48:24 hrw: I ran bitbake gcc-cross, and it failed on compiling glibc after it couldn't find arm-angstron-linux-gnueabi-gcc Jul 28 12:49:03 then start from scratch? Jul 28 13:06:29 Hi. I'm using an AT91SAM9260 ARM8 cpu for my project. I'm using the SSC controller to talk to a TLV320AIC23b codec. However, can I use I2S for more than one device ? I have another codec I would like to connect. Jul 28 13:06:40 ARM9 cpu of course Jul 28 13:06:51 hi guys Jul 28 13:10:54 http://rafb.net/p/GkrBSM69.html Jul 28 13:13:24 package.bbclass notice for shared librairy that are not provided , but bitbake doesn't seems to log this important message , do i miss something ? Jul 28 13:25:07 * * OE Bug 4457 has been created by jazz_bass(AT)mail.ru Jul 28 13:25:09 * * HP iPAQ hx4700 128MB RAM Jul 28 13:25:11 * * http://bugs.openembedded.net/show_bug.cgi?id=4457 Jul 28 13:35:08 mne: yes Jul 28 13:35:12 oh, duh, he quit Jul 28 13:37:43 Genesis: afaik, bitbake doesn't log any of the warning or error messages that it generates. I don't think this particular one is any different. Jul 28 13:38:01 obviously you can send bitbake's stderr to a file if you want Jul 28 13:39:43 The shell functions get logged, the python ones don't. This has changed in bitbake trunk but that brings its own problems Jul 28 13:44:38 oki , noticed Jul 28 13:45:16 thx Jul 28 14:12:43 How are postinst scripts supposed to work on images? In my case, a script is supposed to add a group and a user, but the utilities to manipulate /etc/passwd (part of tinylogin) are all powerpc executables, while the host is a x86. Jul 28 14:13:21 I see a line in the do_rootfs log executing them, but no further detailed output. It doesn't seem to fail either, but the users aren't there afterwards. I suspect it's that. Jul 28 14:16:40 stephank: check the pulseaudio .bb file for an example of how to do that Jul 28 14:17:01 stephank: packages/pulseaudio/pulseaudio.inc Jul 28 14:26:26 XorA: PulseAudio seems to be doing the same as the dbus package, which is what I was looking at. I just noticed the offline check in it though, but when are they supposed to run online? Jul 28 14:26:54 stephank: on first boot Jul 28 14:27:09 there is an ipkg configure in the init scripts Jul 28 14:27:40 XorA: Ah, I think I see what I'm missing. Thanks for the pointers. :) Jul 28 14:29:18 (I derived my image bb from an example, and screwed up that part) Jul 28 14:30:23 stephank: heh heh Jul 28 14:52:55 mickeyl: good morning Jul 28 14:52:59 hi Jul 28 14:54:17 hi mickeyl Jul 28 14:54:25 mickeyl: Are you the admin for the oe-commits list? Jul 28 14:54:51 mickeyl: I suspect the git commit emails are being bounced off there atm... Jul 28 14:56:04 mickeyl: my fso-image build completed overnight but I haven't had much luck getting it to boot in qemu yet. I'll poke at that again later this evening. Jul 28 14:57:21 ok, thanks for the heads up Jul 28 14:57:33 RP: likely yes. what's the exact sender address? Jul 28 14:59:28 mickeyl: which will become git@ Jul 28 14:59:35 mickeyl: Could you whitelist both please? Jul 28 15:02:26 done Jul 28 15:02:34 held mails shoud appear now Jul 28 15:02:42 * XorA waits for the flood Jul 28 15:17:51 hmm, how did the old ts-detect thing work Jul 28 15:17:55 I think I need it Jul 28 15:18:26 for? Jul 28 15:19:01 hrw: for not using udev which can peek at more sysfs stuff Jul 28 15:21:13 ahah, I find it in gpe SVN Jul 28 15:36:28 would an sdk be expected to build a kernel image or just userland apps and perhaps kernel modules? I'm wondering if the sdk I'm putting together for my machine should include mkimage (required to build uImage type kernels) Jul 28 15:44:33 tharvey: if I would have to use sdk to build kernel for device I would expect mkimage in it Jul 28 15:48:03 mickeyl: thanks Jul 28 15:50:06 hrw, yes I guess it depends on what you are trying to support with your sdk. I haven't seen any recipes that put mkimage into the sdk - they put it in staging - so I would want to create my own sdk recipe that would extend another and place mkimage in the appropriate place correct? Jul 28 15:53:39 hmm, somewhere I think I misunderstand mdev.conf Jul 28 15:54:15 XorA: there isn't much to it, it's just a list of devices Jul 28 15:54:31 and, uncharacteristically for busybox, there is even documentation! Jul 28 15:56:57 03jeremy_laine 07org.oe.dev * r8003a340... 10/ (1 packages/imagemagick/imagemagick_6.3.5-10.bb): imagemagick: move *.la files out of dev package, they are needed for modules Jul 28 16:00:44 pb__: the docs are well out of date Jul 28 16:01:01 pb__: Im trying to get device renaming to work, but it seems I fail Jul 28 16:01:38 pb__: ahah, misleading email threw me down wrong path Jul 28 16:05:21 03  07org.openembedded.dev * rc7a6ff7c0d 10OE.dev/packages/imagemagick/imagemagick_6.3.5-10.bb: imagemagick: move *.la files out of dev package, they are needed for modules Jul 28 16:05:27 03  07master * rc7a6ff7c0d 10OE.dev/packages/imagemagick/imagemagick_6.3.5-10.bb: imagemagick: move *.la files out of dev package, they are needed for modules Jul 28 16:09:50 bye Jul 28 16:10:06 03thebohemian 07org.oe.dev * r36d1e788... 10/ (5 files in 2 dirs): Jul 28 16:10:06 libmatthew: Removed 0.5, added 0.7.1 Jul 28 16:10:06 - added a libmatthew.inc for use for 0.7 and 0.7.1 Jul 28 16:10:11 03thebohemian 07org.oe.dev * refd5f8a9... 10/ (3 files in 2 dirs): dbus-java: Adjust DEPENDS to accommodate changes in libmatthew recipe. Jul 28 16:12:24 03  07org.openembedded.dev * r33d72c641b 10OE.dev/packages/libmatthew/ (4 files): Jul 28 16:12:24 libmatthew: Removed 0.5, added 0.7.1 Jul 28 16:12:24 - added a libmatthew.inc for use for 0.7 and 0.7.1 Jul 28 16:12:24 03  07org.openembedded.dev * r861ca6340e 10OE.dev/: Jul 28 16:12:24 merge of '8003a3400630ec53f3ce15fc048a60a29d5e7bed' Jul 28 16:12:25 and 'efd5f8a984f8580641b9e6e68ee6d933d93e6fc2' Jul 28 16:12:29 03  07org.openembedded.dev * r5d78b48cdc 10OE.dev/packages/dbus-java/ (dbus-java_2.3.bb dbus-java_2.4.bb): dbus-java: Adjust DEPENDS to accommodate changes in libmatthew recipe. Jul 28 16:12:32 03  07master * r33d72c641b 10OE.dev/packages/libmatthew/ (4 files): Jul 28 16:12:34 libmatthew: Removed 0.5, added 0.7.1 Jul 28 16:12:36 - added a libmatthew.inc for use for 0.7 and 0.7.1 Jul 28 16:12:38 03  07master * r5d78b48cdc 10OE.dev/packages/dbus-java/ (dbus-java_2.3.bb dbus-java_2.4.bb): dbus-java: Adjust DEPENDS to accommodate changes in libmatthew recipe. Jul 28 16:12:43 03  07master * r861ca6340e 10OE.dev/: Jul 28 16:12:45 merge of '8003a3400630ec53f3ce15fc048a60a29d5e7bed' Jul 28 16:12:47 and 'efd5f8a984f8580641b9e6e68ee6d933d93e6fc2' Jul 28 16:33:45 03mickeyl 07org.oe.dev * rdf71ad28... 10/ (4 files in 2 dirs): vte 0.16.13 ship termcap seperately, so it can be used from the library Jul 28 16:33:49 03mickeyl 07org.oe.dev * r200abe8b... 10/ (1 conf/distro/include/preferred-om-2008-versions.inc): preferred-om-2008-versions.inc: use vte 0.16.13 Jul 28 16:33:54 03mickeyl 07org.oe.dev * r09258eaf... 10/ (1 packages/meta/distro-feed-configs.bb): distro-feed-configs: specify CONFFILES Jul 28 16:37:51 03  07org.openembedded.dev * rf4e8300a50 10OE.dev/packages/vte/ (vte.inc vte_0.16.8.bb vte_0.16.9.bb): vte 0.16.13 ship termcap seperately, so it can be used from the library Jul 28 16:37:59 03  07org.openembedded.dev * r92fa5df829 10OE.dev/packages/meta/distro-feed-configs.bb: distro-feed-configs: specify CONFFILES Jul 28 16:37:59 03  07org.openembedded.dev * r3118af8b95 10OE.dev/conf/distro/include/preferred-om-2008-versions.inc: preferred-om-2008-versions.inc: use vte 0.16.13 Jul 28 16:38:00 03  07org.openembedded.dev * ra60669639b 10OE.dev/: Jul 28 16:38:00 merge of '09258eafc3e7182a15978e372af12800698031ce' Jul 28 16:38:02 and '96d88977407dc82a4987de86d47802dd8322556f' Jul 28 16:38:04 03  07master * r3118af8b95 10OE.dev/conf/distro/include/preferred-om-2008-versions.inc: preferred-om-2008-versions.inc: use vte 0.16.13 Jul 28 16:38:07 03  07master * rf4e8300a50 10OE.dev/packages/vte/ (vte.inc vte_0.16.8.bb vte_0.16.9.bb): vte 0.16.13 ship termcap seperately, so it can be used from the library Jul 28 16:38:10 03  07master * ra60669639b 10OE.dev/: Jul 28 16:38:12 merge of '09258eafc3e7182a15978e372af12800698031ce' Jul 28 16:38:14 and '96d88977407dc82a4987de86d47802dd8322556f' Jul 28 16:38:18 03  07master * r92fa5df829 10OE.dev/packages/meta/distro-feed-configs.bb: distro-feed-configs: specify CONFFILES Jul 28 16:48:42 I'm having a stupid patch issue Jul 28 16:49:08 I'm creating a patch with svn diff, but when I apply it, patch is refusing to creat efiles Jul 28 16:49:16 anyone have any thoghts? Jul 28 16:54:24 urg Jul 28 16:54:35 I think I see the problem Jul 28 17:03:07 Low hanging fruit: libgpg-error and libgcrypt are missing a gettext inheritance. :) Jul 28 17:22:53 what's the best way to go about changing the timezone info on a default OE install Jul 28 17:23:06 i think i'm missing something, as setting /etc/TZ doesn't seem to have any effect Jul 28 17:24:02 Laibsch, fixed it!! now dumping jawiki :D Jul 28 17:24:32 great! Jul 28 17:24:38 looking forward to it Jul 28 17:25:02 :D Jul 28 17:25:15 prepare you squashfs-lzma modules for tomorrow! Jul 28 17:37:16 sw17ch: /etc/localtime, probably. I don't think there is anything in a standard install that reads /etc/TZ. Jul 28 17:38:44 RP: let's patch this change out in OE Jul 28 17:38:47 pb_: i suppose i assumed that the absence of /etc/localtime seemed to point that it wasn't to be used Jul 28 17:39:01 RP: lets try to continue using opkg as long as they do not invent more crazy things Jul 28 17:39:15 i think i've decided to use UTC (the default) for the system, and inform individual processes which time zone they should oeprate/convert to Jul 28 17:39:46 ok, that makes sense Jul 28 18:04:25 ow wow, the entire paxutils directory just dissappeared on ftp.gnu.org and ibiblio! XD Jul 28 18:10:05 mickey|walking: My concern is the political pressure this brings? I'm going to get asked time and again why Poky doesn't support/use .opk from now on :( Jul 28 18:14:48 well, then perhaps we can fix it otherwise Jul 28 18:15:05 i have just received a mail explaining some things Jul 28 18:15:19 i was less annoyed by the technical thing, rather than the way how it was carried out Jul 28 18:15:26 and i have expressed my annoyance Jul 28 18:15:30 feeling better now Jul 28 18:15:53 I wasn't too happy about the way it was carried out either. Doesn't do much for the "open" idea Jul 28 18:16:23 mickey|walking: You mean Holger's email or a different one? Jul 28 18:16:44 no, a different one. Jul 28 18:16:48 forwarding to you in a minute Jul 28 18:16:58 keep it for yourself please Jul 28 18:17:06 will do Jul 28 18:17:39 sent Jul 28 18:28:26 Laibsch, http://img366.imageshack.us/img366/6987/screenshotjoschjoschff2.png Jul 28 18:28:51 Laibsch, care to join #mokopedia again? then i dont have to sit around at #oe posting OT stuff :P Jul 28 18:47:21 can anybody give me some advice on how to make mtdblock emulation work in qemu-system-arm using the "qemuarm" kernel from oe? It seems like you should just be able to say "-mtdblock filename" but that doesn't seem to have any perceptible effect. Jul 28 19:00:09 josch: naisu (that is Japanese, but just read it out loud ;-)) Jul 28 19:06:16 hi Jul 28 19:06:31 is a quil expert here? Jul 28 19:06:45 whats the easiet way to include whole new file with quilt Jul 28 19:08:32 quilt add Jul 28 19:08:32 cp Jul 28 19:08:35 Jul 28 19:08:37 quilt refresh Jul 28 19:08:48 ah hm okay Jul 28 19:08:51 thanks Jul 28 19:08:54 the important thing is to add it to the stack before copying it Jul 28 19:09:15 yes now I understand it Jul 28 19:09:38 I have updated the uclibc-stdlibc++-local patches Jul 28 19:12:41 for gcc-4.3.1 Jul 28 19:12:43 sorry Jul 28 19:12:45 *g* Jul 28 19:31:43 Laibsch, ah... now i get the naisu pun ^^ Jul 28 19:31:59 (reading out lout helped :P ) Jul 28 19:32:03 *loud Jul 28 19:39:16 morning all Jul 28 19:44:10 FYI: http://beagleboard.org/ Jul 28 19:47:23 hi thesing Jul 28 19:47:46 thesing building an kernel with nfsroot support is now easy? Jul 28 19:48:52 I don't know I didn't do anything about nfsroot. Jul 28 19:49:01 anyone know anything about this : http://www.zeromq.org/ Jul 28 19:49:36 Crofton, the pandora gaming handheld (based on OMAP) guys already use oe for their os Jul 28 19:50:11 josch, yeah, some of them hang out in #beagle Jul 28 19:50:28 josch: Told you to read out loud Jul 28 19:50:31 ;-) Jul 28 19:50:34 Laibsch, :D Jul 28 19:55:26 josch: Actually, this is no pun. The loan word exists and is frequently used. ナイス Jul 28 19:56:41 Laibsch, i know ;) Jul 28 19:57:04 Laibsch, i told you about my anime obsession so this is one of the things that you learn from there :P Jul 28 20:01:13 re Jul 28 20:07:38 re Jul 28 20:08:12 re florian Jul 28 20:14:43 so are you resyncing the gumstix branch? Jul 28 20:26:02 frikker, no oe.dev Jul 28 20:27:18 crap Jul 28 20:27:20 wrong window Jul 28 20:28:43 hi florian Jul 28 20:32:02 hi pb Jul 28 20:32:07 hi gnutoo Jul 28 20:32:16 woglinde, hi Jul 28 20:34:05 woglinde, what's going on? Jul 28 20:34:32 hm Jul 28 20:34:34 nothing Jul 28 20:35:21 hi woglinde Jul 28 20:44:04 Crofton: thats still rev.B of beagleboard? the one without EHCI port? Jul 28 20:44:07 03daniel 07org.oe.dev * rc11624e1... 10/ (5 files in 3 dirs): Jul 28 20:44:07 tangogps-fso_0.9.2.bb: Add a tangogps version with experimental patches that Jul 28 20:44:07 support ogpsd. This will hopefully be merged once the Gypsy API is completely Jul 28 20:44:07 supported. Jul 28 20:46:08 03  07org.openembedded.dev * rceed7ab05a 10OE.dev/packages/tangogps/ (4 files in 2 dirs): Jul 28 20:46:08 tangogps-fso_0.9.2.bb: Add a tangogps version with experimental patches that Jul 28 20:46:08 support ogpsd. This will hopefully be merged once the Gypsy API is completely Jul 28 20:46:08 supported. Jul 28 20:46:14 03  07master * rceed7ab05a 10OE.dev/packages/tangogps/ (4 files in 2 dirs): Jul 28 20:46:14 tangogps-fso_0.9.2.bb: Add a tangogps version with experimental patches that Jul 28 20:46:14 support ogpsd. This will hopefully be merged once the Gypsy API is completely Jul 28 20:46:15 supported. Jul 28 20:46:32 okay... Jul 28 20:47:37 hehe Jul 28 20:47:45 ping rp Jul 28 20:54:35 hrw|gone, right Jul 28 20:55:17 woglinde: But I didn't break anything, right? Jul 28 20:55:29 dont know Jul 28 20:55:33 didnt build tangops Jul 28 21:00:27 woglinde: pong Jul 28 21:00:56 rp in gcc-configure-common.inc Jul 28 21:01:27 we define EXTRA_OECONF = "${@['--enable-clocale=generic', ''][bb.data.getVar('USE_NLS', d, 1) != 'no']} Jul 28 21:01:53 hm is there a reason why generic and not gnu for glibc or uclibc for uclibc? Jul 28 21:01:54 does OE have mirror configs? I would like to make it behave such that if it can't find a src package in sources, it first looks to a specific mirror (for all packages), before looking to the SRC_URI the package specifies Jul 28 21:02:15 otherwise the uclibc-locale patches are useles Jul 28 21:03:44 woglinde: No reason I know of Jul 28 21:04:31 tharvey: You probably want PREMIRRORS Jul 28 21:05:54 G'night all Jul 28 21:09:54 03crofton 07org.oe.dev * r6a6b5625... 10/ (5 files in 3 dirs): gumstix-kernel : Update from gumstix svn so kernel builds again. Jul 28 21:11:01 RP, ah perfect thanks - PREMIRROS will be tried first when fetching a file if that fails the actual file will be tried and finally all MIRRORS will be tried - http://bitbake.berlios.de/manual/ch03s04.html Jul 28 21:11:18 03  07org.openembedded.dev * r22d554706d 10OE.dev/packages/linux/ (4 files in 2 dirs): gumstix-kernel : Update from gumstix svn so kernel builds again. Jul 28 21:11:18 03  07org.openembedded.dev * r7a6710f169 10OE.dev/: Jul 28 21:11:18 merge of '6a6b5625ce1a2f68e15757a541a724d13945155a' Jul 28 21:11:18 and 'c11624e19576778237cea5d125bbc9f0b39d26cc' Jul 28 21:11:25 03  07master * r22d554706d 10OE.dev/packages/linux/ (4 files in 2 dirs): gumstix-kernel : Update from gumstix svn so kernel builds again. Jul 28 21:11:25 03  07master * r7a6710f169 10OE.dev/: Jul 28 21:11:25 merge of '6a6b5625ce1a2f68e15757a541a724d13945155a' Jul 28 21:11:26 and 'c11624e19576778237cea5d125bbc9f0b39d26cc' Jul 28 21:12:58 where's the info on mtn key generation & such on the site? Jul 28 21:13:22 03daniel 07org.oe.dev * rd59dea16... 10/ (3 files in 3 dirs): frameworkd: Don't forward accelerometer data via dbus Jul 28 21:17:26 woglinde: wouldn't you be setting USE_NLS to true if you wanted working locale support? Jul 28 21:23:19 pb hehe yes Jul 28 21:23:40 otherwise looking at the m4 file with or yes you get auto Jul 28 21:23:45 +no Jul 28 21:24:05 and auto selects from dawrin linux generic or pachted uclibc Jul 28 21:24:09 right Jul 28 21:24:14 so generic is wron anyway Jul 28 21:24:28 so, the line you quoted is just there to turn locale support off if USE_NLS isn't set Jul 28 21:24:32 and != is confusing Jul 28 21:24:43 if USE_NLS is set, then you don't get any --enable-clocale option, and it uses the default of "auto" Jul 28 21:25:15 I really wonder why it worked the whole time Jul 28 21:25:37 what's the actual problem you are seeing? that code looks right to me. Jul 28 21:25:53 I hit uclibc compile error Jul 28 21:25:57 kergoth: deleted, apparently. but you can find it through the wayback machine: Jul 28 21:25:59 http://web.archive.org/web/20070501165706/www.openembedded.org/wiki/MonotonePhraseBook Jul 28 21:26:15 but only after the toolchain patches from khem Jul 28 21:26:24 and the bug is really old Jul 28 21:26:27 and are you trying to build with locale support on or off? Jul 28 21:26:32 yes Jul 28 21:26:36 which? Jul 28 21:26:39 I dont need NLS Jul 28 21:26:41 uclibc Jul 28 21:26:54 so, off, then? Jul 28 21:27:06 yes Jul 28 21:27:12 in that case, --enable-clocale=generic sounds like the right thing Jul 28 21:27:24 hm why? Jul 28 21:27:38 that just means that libstdc++ won't make any attempt to use the libc locale functions, it will use its built-in copy of the C locale Jul 28 21:27:51 so you can use it on a libc with no locale support at all Jul 28 21:27:56 ah okay Jul 28 21:28:16 but then I wonder why we didnt hit the bug Jul 28 21:28:18 before Jul 28 21:28:33 http://gcc.gnu.org/bugzilla/attachment.cgi?id=6132 is the patch Jul 28 21:28:38 mickeyl: wb Jul 28 21:28:48 re Jul 28 21:28:55 he mickeyl Jul 28 21:29:32 pb but thanks that you enlight me with this stuff Jul 28 21:29:37 hi woglinde Jul 28 21:30:49 pb but I will change the != 'no' Jul 28 21:30:54 thats confusing Jul 28 21:31:22 yeah, it is a bit Jul 28 21:39:43 pb_ do you know how USE_NLS really works Jul 28 21:39:50 there is one global Jul 28 21:39:56 and then per package Jul 28 21:44:05 morning all Jul 28 21:47:41 hi ant Jul 28 21:48:10 woglinde: wie geht es? Jul 28 21:48:45 ant I am fine Jul 28 21:49:02 reinventing the wheel (useflags)? Jul 28 21:49:47 a global/package USE would be a nice addition... Jul 28 21:57:36 I have created recipe for uclibc arm-nptl for testing the arm-nptl merge upstream. I wonder if it will be useful to have this recipe committed Jul 28 21:58:11 khem hm Jul 28 21:58:16 I would test it Jul 28 21:58:44 khem after your toolchain Jul 28 21:58:58 gcc-4.3.1 libstdc++ build fails Jul 28 21:59:08 a patch is here and some how old Jul 28 21:59:10 http://gcc.gnu.org/bugzilla/attachment.cgi?id=6132 Jul 28 21:59:12 woglinde: hmmm for uclibc or glibc ? Jul 28 21:59:19 but I wonder why we dont hit before Jul 28 21:59:31 khem uclibc USE_NLS=no Jul 28 21:59:56 gcc 4.3.1 is new recipe realtively Jul 28 22:00:13 may be we did not try enough so far. Jul 28 22:00:35 ???? Jul 28 22:00:47 USE_NLS=no then you can use clocale for libstdc++ Jul 28 22:00:50 before your changes libstdc++ build fine Jul 28 22:00:56 yes Jul 28 22:01:00 its generic than Jul 28 22:01:04 and hit this error Jul 28 22:01:40 it must have been a latent bug then Jul 28 22:01:52 hm anyway I can fix it Jul 28 22:01:59 yes you should Jul 28 22:02:04 I must Jul 28 22:02:06 *g* Jul 28 22:02:11 but I really wonder Jul 28 22:02:53 I think it could be another patch that I did to fix SDK build for gcc 4.3.1 Jul 28 22:03:02 that could have brought this to fore Jul 28 22:03:12 hm Jul 28 22:04:29 khem hm the patch repos should be cleanded somehow Jul 28 22:04:41 there are serval doubles Jul 28 22:05:04 602-sdk-libstdc++-includes.patch and sdk-libstdc++-includes.patch Jul 28 22:05:08 for instance Jul 28 22:05:11 woglinde: I know this patch I have backported it before Jul 28 22:05:48 woglinde: we maintain a lot of versions of gcc one reason Jul 28 22:08:50 hmmm we need __mempcpy patch for uclibc as well for coreutils to build Jul 28 22:09:38 hm gcc? Jul 28 22:09:41 or in uclibc Jul 28 22:10:28 uclibc or may be better in coreutils Jul 28 22:21:04 03mickeyl 07org.oe.dev * r9e8d77e2... 10/ (1 packages/openmoko-3rdparty/epiano_0.3.1.bb): add epiano, an EFL-based piano application Jul 28 22:21:08 03mickeyl 07org.oe.dev * ra9043ecc... 10/ (3 files in 3 dirs): remove specific feed-configs, superseded by distro-feed-configs.bb Jul 28 22:21:14 03daniel 07org.oe.dev * r4c7ac645... 10/ (1 packages/gypsy/gypsy_svn.bb): gypsy_svn.bb: Gypsy needs libxslt-native for its documentation Jul 28 22:21:19 03daniel 07org.oe.dev * rd3210ecb... 10/ (1 packages/images/fso-image.bb): fso-image: Use tangogps-fso Jul 28 22:21:41 mtn wants to update 76 revs and takes 23 mins to tell me that there are multiple heads Jul 28 22:22:35 pull again Jul 28 22:24:27 woglinde: yeah the point is 23mins Jul 28 22:28:53 03  07org.openembedded.dev * rf2ad26788b 10OE.dev/packages/freesmartphone/ (frameworkd/om-gta02/frameworkd.conf frameworkd_git.bb): frameworkd: Don't forward accelerometer data via dbus Jul 28 22:28:54 03  07org.openembedded.dev * ra9e8f136fc 10OE.dev/: Jul 28 22:28:54 merge of '154d82740f4a13ef53553ca5cd9af7689511b994' Jul 28 22:28:54 and 'd59dea16aa8c4ba3bb6565204bc02a02a9e906ff' Jul 28 22:28:56 03  07org.openembedded.dev * r6d39a43a4b 10OE.dev/packages/openmoko-3rdparty/epiano_0.3.1.bb: add epiano, an EFL-based piano application Jul 28 22:28:57 03  07org.openembedded.dev * r1818631ea6 10OE.dev/packages/images/fso-image.bb: fso-image: Use tangogps-fso Jul 28 22:29:00 03  07org.openembedded.dev * r467c2828b1 10OE.dev/packages/gypsy/gypsy_svn.bb: gypsy_svn.bb: Gypsy needs libxslt-native for its documentation Jul 28 22:29:03 03  07org.openembedded.dev * r2de67e24fe 10OE.dev/packages/ (2 files in 2 dirs): remove specific feed-configs, superseded by distro-feed-configs.bb Jul 28 22:29:08 03  07org.openembedded.dev * r90e2451db7 10OE.dev/: Jul 28 22:29:10 merge of '2159dc39693944f0d5864c14aa1d9513f3cbefa0' Jul 28 22:29:12 and 'd3210ecbfcc795109a513bb1c8bb6469047ed55d' Jul 28 22:29:14 03  07org.openembedded.dev * ra512e5acb5 10OE.dev/: Jul 28 22:29:16 merge of '0ad423d3482b73008fa7b39e78b0dbc3658cbc50' Jul 28 22:29:18 and 'a9043ecc52370e4ecf1fd689e40cf12dac122cb9' Jul 28 22:29:20 03  07master * r90e2451db7 10OE.dev/: Jul 28 22:29:24 merge of '2159dc39693944f0d5864c14aa1d9513f3cbefa0' Jul 28 22:29:26 yeah, mtn is sloww at the moment. Jul 28 22:29:26 and 'd3210ecbfcc795109a513bb1c8bb6469047ed55d' Jul 28 22:29:28 03  07master * rf2ad26788b 10OE.dev/packages/freesmartphone/ (frameworkd/om-gta02/frameworkd.conf frameworkd_git.bb): frameworkd: Don't forward accelerometer data via dbus Jul 28 22:29:31 03  07master * r1818631ea6 10OE.dev/packages/images/fso-image.bb: fso-image: Use tangogps-fso Jul 28 22:29:33 khem: but *somebody important* likes it, so it stays... Jul 28 22:29:33 whoa Jul 28 22:29:38 03  07master * r2de67e24fe 10OE.dev/packages/ (2 files in 2 dirs): remove specific feed-configs, superseded by distro-feed-configs.bb Jul 28 22:29:41 03  07master * r467c2828b1 10OE.dev/packages/gypsy/gypsy_svn.bb: gypsy_svn.bb: Gypsy needs libxslt-native for its documentation Jul 28 22:29:44 03  07master * r6d39a43a4b 10OE.dev/packages/openmoko-3rdparty/epiano_0.3.1.bb: add epiano, an EFL-based piano application Jul 28 22:29:49 03  07master * ra512e5acb5 10OE.dev/: Jul 28 22:29:51 merge of '0ad423d3482b73008fa7b39e78b0dbc3658cbc50' Jul 28 22:29:53 and 'a9043ecc52370e4ecf1fd689e40cf12dac122cb9' Jul 28 22:29:55 03  07master * ra9e8f136fc 10OE.dev/: Jul 28 22:29:57 merge of '154d82740f4a13ef53553ca5cd9af7689511b994' Jul 28 22:29:59 and 'd59dea16aa8c4ba3bb6565204bc02a02a9e906ff' Jul 28 22:33:23 mwester: I hope someday there is a SCM system which sucks less Jul 28 22:34:41 it does not stay Jul 28 22:34:46 well, it does Jul 28 22:34:48 but it will not Jul 28 22:34:58 mtn is soon history Jul 28 22:35:06 as you can see with the git revs in the changelogs Jul 28 22:35:12 we're right in test mode Jul 28 22:35:23 so please have a little more patience Jul 28 22:37:59 mickeyl: than I hope git sucks less :) Jul 28 22:38:21 i hope so too Jul 28 22:39:18 mickeyl: is it planned an abrupt change or will both mtn and git live together for a while before git takes over Jul 28 22:50:44 [OT]: /tmp is tmpfs for me. I have a single task that requires slightly more space than usual. Is it possible to divert /tmp (not OE TMPDIR!) to a different location just for this task? ("TMPDIR=/bigpartition bash $task" ?) Jul 28 22:51:26 mount -o bind /harddisk/tmp /tmp Jul 28 22:52:20 XorA|gone: good idea Jul 28 22:52:32 but I am afraid that messes up processes already running Jul 28 22:52:38 that are accessing /tmp Jul 28 22:52:43 Laibsch: already open files should stay open Jul 28 22:52:51 Laibsch: new file will end up in your new tmp Jul 28 22:53:01 OK, let's give it a try Jul 28 22:53:11 There's always a reboot if things fail ;-) Jul 28 22:53:16 Thanks, XorA|gone Jul 28 22:54:08 anyway I should really attempt to sleep, maybe some night it shall happen Jul 28 22:54:40 xora haha Jul 28 22:55:24 that mickey|zzZZzz always copies me Jul 28 22:55:59 is it the distro or the image that causes busybox to be used for specific apps? Jul 28 22:56:29 khem: i'd like to keep mtn read-only for a while Jul 28 22:59:09 so nice libstdc++ compiles again Jul 28 23:02:00 I'm trying to figure out how to create an image that uses full apps vs busybox Jul 28 23:17:32 nite Jul 28 23:45:53 tharvey: I did not get your question Jul 28 23:46:18 it seems you want to use sysvinit and coreutils etc Jul 28 23:46:26 and no busybox ? Jul 28 23:48:56 I suppose so yes Jul 28 23:50:05 khem, does coreutils use busybox or provided alternatives to busybox applets? Jul 28 23:54:26 tharvey: coreutils have alternatives Jul 28 23:54:58 so simply including coreutils in your image will replace busybox? Jul 28 23:55:17 tharvey: I dont think so Jul 28 23:59:06 tharvey: you need to replace INITSCRIPTS stuff Jul 28 23:59:09 as well Jul 29 00:06:19 thesing: ping Jul 29 00:08:36 khem: pong Jul 29 00:09:48 thesing: I did not try BB_POLICY stuff though Jul 29 00:10:05 I thought this should have worked Jul 29 00:10:49 I didn't do a clean build. But I don't think this matters. Jul 29 00:11:30 i.e. I did a normal build switched the BB_POLICY stuff on, and tried to build sth. Jul 29 00:11:33 circular dependency is happening because earlier glibc was built before gcc-cross and there was no dep for cross-gcc on glibc instead it used glibc-intermediate Jul 29 00:12:07 but now glibc is build prior to gcc-cross so there is a dep for cross-gcc on glibc Jul 29 00:12:52 I guess this change is confusing bitbake RP might tell more Jul 29 00:13:05 but a clean build should work Jul 29 00:13:29 I think there is a whitelist somewhere to solve such stuff. Jul 29 00:13:47 hmmm yeah let me look for it Jul 29 00:13:56 a clean build only works until glibc is changed. then it will fail again. Jul 29 00:14:44 hmm Jul 29 00:19:21 ok time for bed. Bye Jul 29 00:19:30 thesing: good night ! Jul 29 00:19:35 are u in europe Jul 29 00:20:29 so if I use BB_STAMP_POLICY="whitelist" I should see the problem. I will try it out thx **** ENDING LOGGING AT Tue Jul 29 02:59:56 2008