**** BEGIN LOGGING AT Thu Jan 29 02:59:57 2009 Jan 29 08:49:48 03Robert Schuster  07robertschuster/amd64-target * r0aeb635f90 10openembedded.git/conf/machine/amdbarcelona-generic.conf: amdbarcelona: Fixed metadata. Jan 29 08:58:25 mickeyl: hi! :) what's the cause for the switch to py2.6? Jan 29 09:00:09 03Graeme Gregory  07org.openembedded.dev * r7c823edcc8 10openembedded.git/conf/machine/tosa.conf: tosa.conf : add to IMAGE_FSTYPES instead of just setting it Jan 29 09:00:51 morning emdete, well, apart from the usual updatetitis, a bunch of optimizations in how it talks to glib, and also some deprecations towards python 3 Jan 29 09:01:19 mickeyl: is it wise to do that /now/? Jan 29 09:01:40 well Jan 29 09:01:48 it's already done since one or two weeks Jan 29 09:02:07 why btw.? Jan 29 09:02:48 ...there was a branch for that. shure - i thought it's experimental. i am afraid to fight against some incompatiblities in libraries pyneo uses... Jan 29 09:03:07 definitely not experimental Jan 29 09:03:15 i have not seen any incompatibilities Jan 29 09:03:16 what are the "optimizations" you talk about? Jan 29 09:03:18 which ones do you see? Jan 29 09:03:27 wake up timers Jan 29 09:03:35 i currently have no 2.6 here. Jan 29 09:03:39 there's some bugs about that in python and/or gnome bugtracker, don't have the numbers offhand Jan 29 09:03:58 some excessive wakeups have been fixed Jan 29 09:04:00 is this related to the py version or to tha glib impl? Jan 29 09:04:10 combination of python and python-gobject Jan 29 09:04:54 gobject is a different thing.. do you have a link? Jan 29 09:05:39 not offhand Jan 29 09:05:57 mickeyl: not even debian unstable uses py2.6 - that a clear warning for me ;) Jan 29 09:06:02 heh Jan 29 09:06:06 feel free to continue with 2.5 Jan 29 09:06:27 with "end of support of oe"... Jan 29 09:06:44 yes :) Jan 29 09:06:48 see Jan 29 09:06:49 well opensuse 11.1 uses 2.6 :-) Jan 29 09:07:03 another warning ;) Jan 29 09:07:13 Jan 29 09:08:00 it is definitely the road ahead Jan 29 09:08:40 yep Jan 29 09:08:49 you should be glad that i didn't add python3 Jan 29 09:08:52 (which some folks requested) Jan 29 09:09:25 py3 sux Jan 29 09:09:33 adding it should be fine, but please do not make it preferred, as that could break a few things Jan 29 09:09:36 they made lots of things even worse Jan 29 09:09:41 heh Jan 29 09:09:49 well, all in all i like the changes Jan 29 09:10:05 which things do you mean? Jan 29 09:10:10 httplib got a timeout but not fully. stringformating begins to get a nightmare like in other languages... Jan 29 09:10:30 so the lib redesign didn't take place at all :/ Jan 29 09:11:57 03Michael 'Mickey' Lauer  07org.openembedded.dev * r975e14469b 10openembedded.git/packages/python/ (files/06-libffi-enable-default-mips.patch python_2.6.1.bb): python: fix libffi building for mips. Should close #5001 Jan 29 09:13:15 i don't think that was on topic for 3.0 Jan 29 09:13:22 topic for 3.0 was core language design Jan 29 09:13:30 later versions will bring stdlib changes Jan 29 09:13:36 (i think) Jan 29 09:14:00 anyways, if things work out, I'm going to do more and more in Vala Jan 29 09:14:14 i really like it so far Jan 29 09:16:06 emdete: how's pyneo progressing? Jan 29 09:16:18 didn't have a chance to flash it yet Jan 29 09:16:52 emdete: btw., do you plan incorporating abraxa's work? Jan 29 09:24:52 mickeyl: what bloomin usb cable fits a tosa Jan 29 09:25:46 i guess these new OTG things Jan 29 09:28:43 looks like a mini A and everything else is a mini B :-( Jan 29 09:30:04 mickeyl: shure, pim wasn't in sight till now. i had to implement some new ideas on the api. Jan 29 09:31:01 mickeyl: vala looks good but lacks alot of libraries. so you have to implement alot yourself Jan 29 09:31:16 agreed Jan 29 09:31:23 although you can use every C library on the planet :) Jan 29 09:31:38 mickeyl: so pyneos highlevel stuff wont be done in vala soon Jan 29 09:32:04 well Jan 29 09:32:09 then you'd have to rename the project anyways ;) Jan 29 09:32:15 mickeyl: also python is much better in non-threading programming ;) Jan 29 09:32:33 true Jan 29 09:32:37 although jürg is working on coroutines Jan 29 09:32:51 i will take a look when he's done Jan 29 09:33:14 yeah Jan 29 09:33:18 mickeyl: do you play with the siemens chip already? Jan 29 09:33:29 sure Jan 29 09:33:31 works nice Jan 29 09:33:43 didn't see any oddities yet Jan 29 09:33:46 but didn't use it a lot Jan 29 09:33:55 that'll change once we get the first 03 silicon Jan 29 09:34:03 right now i only have the standalone siemens board Jan 29 09:34:24 mickeyl: pyneo is such a nice name that it will remain - py is the j2me/Dalvik kill ;) Jan 29 09:34:37 mickeyl: would you borrow that? Jan 29 09:35:00 not before i have the first 03 silicon Jan 29 09:35:10 are you at the fosdem? Jan 29 09:35:12 no Jan 29 09:35:22 oh, sad... why not? Jan 29 09:36:33 are oe people there? Jan 29 09:46:28 emdete: there are a crowd of us at FOSDEM Jan 29 09:47:05 03Koen Kooi  07org.openembedded.dev * rf33eef09f1 10openembedded.git/packages/dsplink/codec-engine_2.21.bb: codec-engine: adjust staging for DMAI Jan 29 09:47:55 morning Jan 29 09:49:00 good morning Jan 29 09:50:09 mickeyl: no fosdem? ;( Jan 29 09:50:36 a bunch of reasons Jan 29 09:50:57 i was a bit disappointed wrt. to getting a meeting ogin Jan 29 09:51:00 s/ogin/going/ Jan 29 09:51:09 also, it's very early this year Jan 29 09:51:23 and in March travveling will start for me Jan 29 09:51:31 so i figured i try to rest in Feb. a bit Jan 29 09:52:33 XorA: how was merging tosa? Jan 29 09:54:32 hrw: I built yesterday kernels for collie and poodle...well, the gcc-2.95/embedix plague is still there...would you clean like you did with other Zaurus? Jan 29 09:55:22 03Koen Kooi  07org.openembedded.dev * ra3e236ed48 10openembedded.git/packages/dsplink/dmai_1.16.bb: dmai: make build get a bit further Jan 29 09:56:00 mickeyl: ok, I hope we will meet in 2009 Jan 29 09:56:53 for sure Jan 29 09:56:57 we need to arrange a meeting Jan 29 09:57:07 it's tough to get hold of RP these days Jan 29 09:57:13 and holger is also pretty apathic Jan 29 10:00:28 yep Jan 29 10:11:47 hi greentux Jan 29 10:12:00 hi florian Jan 29 10:12:09 mickeyl: We call this "Winterschlaf" ;) Jan 29 10:12:33 florian: bad news the xmodmap thing in GPE seems broken again... Jan 29 10:13:10 ant_work: hrm.... nasty, any idea which commit might be guilty? Jan 29 10:13:26 gpe.... Jan 29 10:13:28 hmm..koen put an hack in the initscripts Jan 29 10:13:36 which should hav esolved the issue Jan 29 10:13:41 we need to generate x11-nongpe-image Jan 29 10:13:59 yes..to separe the bugs... Jan 29 10:14:28 hrw: Some kind of xdmcp-terminal-image would be nice Jan 29 10:14:59 florian: I will need normal WM + terminal + panel image Jan 29 10:15:28 hrw: Something like the beagleboard demo image? Jan 29 10:15:52 lighter but probably yes Jan 29 10:16:03 bb in 10m Jan 29 10:16:25 florian: I'd start from here http://cgit.openembedded.net/cgit.cgi?url=openembedded/commit/&id=814da25dfa191482e1b0bf614eb064c85a3233b0 Jan 29 10:17:31 and the hrw fix Jan 29 10:17:33 http://cgit.openembedded.net/cgit.cgi?url=openembedded/commit/&id=42087db475b12c995755d430de5b59a5e1add199 Jan 29 10:22:25 ant_work: but you built angstrom, right? Jan 29 10:22:31 yep Jan 29 10:23:06 image of 2 days ago has wrong keymap in X Jan 29 10:23:42 hey ant_work Jan 29 10:24:01 btw shutting down x and going back to console revives the missing keys (/ one for all) Jan 29 10:24:08 XorA: morning Jan 29 10:24:25 ant_work: tosa is in and working Jan 29 10:24:31 ant_work: I had to patch kexecboot a little Jan 29 10:24:37 ant_work: but working nicely Jan 29 10:24:40 XorA: :-) I've seen the logs Jan 29 10:24:56 and the 'production' kernel? Jan 29 10:25:07 2.6.24 seems fully functional Jan 29 10:25:19 only thing I havent tested is wifi Jan 29 10:25:35 there was some discuss in OESF.... Jan 29 10:26:26 wifi seems to be working, iwlist scan picks up APs Jan 29 10:26:52 mickeyl: could you commit my patches for pyneo to oe? Jan 29 10:27:33 suspend/resume works Jan 29 10:27:35 XorA: that thread http://www.oesf.org/forum/index.php?showtopic=25057&st=45&start=45 Jan 29 10:27:43 hopefully all solved now Jan 29 10:27:43 the only slight problem is the yellow fbdev bug Jan 29 10:27:45 emdete: i can't. lets discuss that later, i have a TC atm. Jan 29 10:28:01 and the keymap is buggered which you already noticed Jan 29 10:28:03 :-) Jan 29 10:28:23 for a while was repaired... Jan 29 10:28:30 then broke again... Jan 29 10:28:55 XorA: tosa wifi driver is now part of 'staging' drivers in mainline Jan 29 10:29:06 hrw: awesome Jan 29 10:29:07 maybe one day it will be better usable Jan 29 10:30:06 XorA: I've seen you went straight and deployed full-featured kernel for Tosa Jan 29 10:30:19 our thesing did it already for collie Jan 29 10:30:32 and poodle iirc Jan 29 10:30:40 ant_work: I just applied the patches, I saw no reason not to Jan 29 10:30:51 as there seem to be only 5 people in the world with a tosa Jan 29 10:30:55 and most of them are here :-) Jan 29 10:31:08 yes, but now the kernel size is too big for nand Jan 29 10:31:13 :-) Jan 29 10:31:41 with my Angstrom hat on I would say "screw the nand" Jan 29 10:31:53 I don't see any good reason for respecting the old kernel layout in mtd1 Jan 29 10:32:02 here in .dev at least Jan 29 10:32:57 and yes..the embedix recipes too... Jan 29 10:33:15 old stuff...if one needs that can build openzaurus :-) Jan 29 10:33:52 ant_work: OZ used 2.4.18 only on collie Jan 29 10:34:41 ok, then one can use Sharp sources :-) Jan 29 10:34:55 if really really is obliged to Jan 29 10:34:56 ~seen thesing Jan 29 10:34:58 thesing was last seen on IRC in channel #oe, 40d 14h 22m 12s ago, saying: 'zecke, I thought you are one.'. Jan 29 10:36:00 ant_work: seems like its an agreement that at least Angstrom will use kexecboot :-) Jan 29 10:36:10 seems so Jan 29 10:36:21 ant_work: just asked koen and he agrees Jan 29 10:36:38 03Stefan Schmidt  07fso/milestone5 * r696b011d47 10openembedded.git/conf/distro/include/preferred-om-2008-versions.inc: preferred-om-2008-versions.inc: Fix rest of the outdated versions Jan 29 10:36:38 03Stefan Schmidt  07fso/milestone5 * rf1fa8a2c36 10openembedded.git/conf/distro/include/sane-srcrevs.inc: ane-srcrevs.inc: Bump frameworkd to get gta01 ogpsd fix Jan 29 10:36:39 03Stefan Schmidt  07fso/milestone5 * rc83c8a845f 10openembedded.git/: Merge branch 'fso/milestone5' of git@git.openembedded.net:openembedded into fso/milestone5 Jan 29 10:36:55 with so few maintainers for Z stuff, I don't see us supporting multiple paths Jan 29 10:37:20 XorA: perhaps .stable could keep the old layout...dunno Jan 29 10:37:31 'm for house-cleaning Jan 29 10:37:40 ant_work: I think the new stable should come from here, old stable can stay as it is Jan 29 10:37:51 ant_work: when/if we fork a new stable Jan 29 10:38:08 he..there are some magic days where all works... Jan 29 10:38:30 one should branch during these bright days... Jan 29 10:38:51 unfortunately there are still too many bugs.... Jan 29 10:38:56 ant_work: would be nice to get Z's upto 2.6.28 though Jan 29 10:39:13 I've read in LKML spitz support is almost done Jan 29 10:39:49 btw 2.6.26 has still some minor quirks (at least on c7x0) Jan 29 10:40:04 is about sensing the AC power Jan 29 10:40:28 if the unit is suspended the rcharging process won't start Jan 29 10:40:42 np if you branch the cable with unit on Jan 29 10:40:53 this is evident in kexecboot too Jan 29 10:41:30 try to branch the cable *before* starting the machine from a cold reboot: the charging led will shut off Jan 29 10:41:47 (never seen this on 2.6.24) Jan 29 10:42:08 probably is some pm quirk Jan 29 10:42:28 but as I said, it's for sure kernel Jan 29 10:45:15 Xora: 2.6.29 boots Jan 29 10:45:17 http://lkml.org/lkml/2009/1/28/102 Jan 29 10:45:35 I has no spitz Jan 29 10:48:04 Hello everyone, just building angstrom for beagleboard-demo-image and I ran into a a MD5 mismatch. Can I work around this or should I just wait? pastebin.com/m2fadc546 Jan 29 10:48:49 ~curse yellow corruption bug Jan 29 10:48:50 May you be reincarnated as a Windows XP administrator, yellow corruption bug ! Jan 29 10:49:45 XorA: bug 490 Jan 29 10:49:53 very old! Jan 29 10:50:01 ant_work: I know Jan 29 10:50:10 ant_work: been ignoring it for years :-) Jan 29 10:50:17 :-) Jan 29 10:51:47 XorA: most of the Zaurus bugs are probably fixed... Jan 29 10:52:27 these META bugs need a bit more care... Jan 29 10:53:54 * XorA would be tempted to blanket close lots of them and see what re-opens :-) Jan 29 10:54:03 03Stefan Schmidt  07fso/milestone5 * rd04c901fdf 10openembedded.git/ (2 files in 2 dirs): task-fso-compliance: Ship fso-monitord in the latest version Jan 29 10:55:40 XorA: I'd close all WPA related...it just works Jan 29 10:55:56 never found time to test usb-networking though Jan 29 10:59:51 I would like to try it on my spitz, but still no serial cable :/ Jan 29 11:01:48 ant_work: does kexecboot mmap the framebuffer? Jan 29 11:02:21 I think this is done by kernel Jan 29 11:02:57 ant_work: I dont think the kernel draws the menu :-) Jan 29 11:03:07 but this is exactly the point where it would possibly fail on c700 Jan 29 11:03:20 sorry, I mean giving address to fb Jan 29 11:03:47 c700 has only 32m and perhaps the fb needs to be relocated down Jan 29 11:07:16 Xora: yes, it is mmapped Jan 29 11:07:28 look at fb.c Jan 29 11:08:15 ant_work: am checking Jan 29 11:08:27 macneib: rm DL_DIR/glib-1.2.10.tar.gz* Jan 29 11:08:28 ant_work: kexecboot shows the yellow corruption real bad :-( Jan 29 11:08:33 macneib: your archive is broken Jan 29 11:08:40 XorA: interesting Jan 29 11:08:44 macneib: I just fetched it for comparison Jan 29 11:12:57 fgomez: morning Jan 29 11:17:10 hmm, looks like #490 has no fix except a shadow buffer in kernel :-( Jan 29 11:20:41 !oebug 490 Jan 29 11:20:45 hrw: morning Jan 29 11:21:38 ant_work: Pavel Machek and his Zaurus kernel skills :) Jan 29 11:21:46 XorA: sound scary Jan 29 11:22:35 hrw: is Pawel from PL? Jan 29 11:23:19 no...pavel@ucw.cz Jan 29 11:23:26 ant_work: Czech Jan 29 11:23:52 he..nowadays only Eastern hackers Jan 29 11:23:55 he has knowledge but sometimes use it in totally weird/wrong way Jan 29 11:24:21 :-D Jan 29 11:24:35 dark side of the force Jan 29 11:27:16 Thank hrw! have it working again. Jan 29 11:29:21 macneib: can you suggest better message for md5 checksums being wrong? Jan 29 11:33:55 hrw: are you suggesting that my MD5 problem description was crappy? Jan 29 11:34:00 no Jan 29 11:34:21 macneib: you just did not know what '# Jan 29 11:34:28 NOTE: The MD5Sums did not match. Wanted: '6fe30dad87c77b91b632def29dd69ef9' and Got: 'd41d8cd98f00b204e9800998ecf8427e' Jan 29 11:34:36 meant. Jan 29 11:34:49 macneib: can you suggest better message for such situations? Jan 29 11:40:24 hrw: oh... sorry... i understand now.. I've been up working all night and not thinking clearly. I'm not sure if there is a better message because if the checksum is mismatched on account of user's system corrupting it then it would be fine to direct the user to do someplace in the oe download directory, but if the mismatch occurs on account of there original server side file being compromised then that would be up to the server admin to fix. On account of Jan 29 11:40:25 someone potentially injecting malicious code at the server side. Jan 29 11:41:11 let me think for a minute Jan 29 11:43:51 * hrw wants to kill blueprobe Jan 29 11:44:48 now that i think about it, perhaps it resemble something like this: 'please delete file and MD5 Checksum and run bitbake again' Jan 29 11:45:11 I just removed 9 archives from my DL_DIR due that Jan 29 11:45:39 contrib/source-checker/oe-source-checker.py is great script Jan 29 11:45:48 hrw: that's a lot of corruptions Jan 29 11:45:57 in 6128 files? Jan 29 11:46:20 this is my first time I was kinda confused. Jan 29 11:47:11 i'm a newbie though so what do I know... Jan 29 11:48:27 thanks again... nap time Jan 29 11:51:06 03Stanislav Brabec  07org.openembedded.dev * r7fcb561081 10openembedded.git/packages/mysql/mysql_4.1.22.bb: Jan 29 11:51:06 mysql: Provide config cache for ps checks: Jan 29 11:51:06 * Scripts need target ps, not build host ps. Jan 29 11:51:06 * It makes possible to build it on host chroots without /proc mounted. Jan 29 11:56:08 03M. Dietrich  07org.openembedded.dev * r392db7f27c 10openembedded.git/packages/images/ (epydial-image.bb moko-underground-image.bb zad-image.bb): epydial.image: split from moko-underground-image; also zad-image Jan 29 11:57:37 emdete: i can't add pyneo_git.bb. we only add SCM recipes if there is an SCM behind it. We better use fixed versioning until then Jan 29 11:59:32 03M. Dietrich  07org.openembedded.dev * rd8230ba5ab 10openembedded.git/packages/bluez/bluez-utils_3.33.bb: bluez-utils: enable more tools Jan 29 12:01:46 mickeyl: good morning Jan 29 12:01:53 morning pb! Jan 29 12:07:38 hrw: Hi, a little Koen sized birdie tells me you have been looking at UBIFS images as an OE target? Is that true? Jan 29 12:09:56 heh Jan 29 12:10:34 03M. Dietrich  07org.openembedded.dev * r5a5ef46cc6 10openembedded.git/packages/gtk+/gdk-pixbuf_2.10.14.bb: gdk-pixbuf: remove no-longer-applying patch Jan 29 12:15:48 sort of Jan 29 12:16:01 DJWillis: I switched my devboard to ubifs yesterday Jan 29 12:16:07 emdete: 014 no longer necessary Jan 29 12:16:19 emdete: 011 should not be necessary Jan 29 12:16:29 emdete: fixed 010 Jan 29 12:16:54 hrw: looking into it for the Pandora you see, trying to get the measure on the OE hacks needed rather then post building the image from the TAR ;-) Jan 29 12:17:18 emdete: 015 i can't apply (see above) Jan 29 12:18:11 hey DJWillis Jan 29 12:18:50 Hi XorA, not forgotten about sending your post out ;-) just got wrapped up in some real life work ;-) Jan 29 12:18:59 DJWillis: no probs Jan 29 12:19:05 DJWillis: playing with a tosa at the moment Jan 29 12:20:23 DJWillis: I will look into problem later Jan 29 12:20:32 XorA: cool Jan 29 12:20:40 DJWillis: the problem is that you need to know target flash size ;( Jan 29 12:21:00 hrw: just trying to work out the work needed, yep, that is what is screwing me up. Jan 29 12:23:30 03Koen Kooi  07org.openembedded.dev * r2b27f2c769 10openembedded.git/packages/bluez/bluez-utils_3.36.bb: bluez-utils 3.36: sync with 3.33 Jan 29 12:24:09 <_boto> hello Jan 29 12:24:27 hi _boto Jan 29 12:24:46 <_boto> hi DJWillis, how are things going? Jan 29 12:25:09 _boto: next ;-) Jan 29 12:25:15 <_boto> hehe Jan 29 12:25:41 <_boto> the usual frustration? Jan 29 12:25:53 yep, guess so Jan 29 12:25:59 <_boto> hehe, what's up? Jan 29 12:26:46 TODO list 5 times longer then time avalable and a full time day job 'on the side' Jan 29 12:32:32 mickeyl: i just ask for the pyneo related patches. Jan 29 12:33:45 emdete: sure, 01 is in Jan 29 12:33:55 03 and 015 bring in _git.bb which is problematic Jan 29 12:34:05 014 and 011 no longer necessary Jan 29 12:34:07 010 in Jan 29 12:37:37 mickeyl: so everything but 3, 11, 14, 15 are in? Jan 29 12:38:10 no Jan 29 12:38:19 04-08 are problematic as well Jan 29 12:38:23 since they change policy Jan 29 12:38:31 in a global way Jan 29 12:38:57 so either you need to do create an override Jan 29 12:39:04 or try to run that through oe-devel Jan 29 12:39:08 mickeyl: i am alone with pyneo and have no resources to setup a scm Jan 29 12:39:41 sure, i don't question that. its just that we can't apply a recipe where the contents of the tarball changes without changing the version Jan 29 12:39:49 btw., there are free hosters Jan 29 12:39:54 perhaps you want to take a look there Jan 29 12:40:31 i need reliability - use enough time elsewhere so: no external free hoster which btw take time also Jan 29 12:41:11 sure. i just try to make the point clear. no problem with versioned recipes Jan 29 12:41:14 like... pyneo 1.7.1 Jan 29 12:41:15 1.7.2 Jan 29 12:41:16 etc. Jan 29 12:41:36 can we go with HEAD until 2.0 is released? that is hopefully soon. Jan 29 12:42:08 we can name the tarball head, but the version needs to be something meaningful otherwise you destroy update path Jan 29 12:42:20 we just seemed to struggled into another problem that again steels time: sysfs names changed again for current kernel :/ did you notice that too? Jan 29 12:42:36 sure Jan 29 12:42:42 03Graeme Gregory  07org.openembedded.dev * rd11d4ce87b 10openembedded.git/ (conf/checksums.ini packages/connman/connman-gnome_0.4.bb): Jan 29 12:42:42 connman-gnome_0.4.bb : add connman-gnome so we can attempt to control Jan 29 12:42:42 connman Jan 29 12:42:50 you need to track openmoko-kernel until the kernel is stable Jan 29 12:43:03 mickeyl: the "HEAD" in the tarball is because the file will change ... so a version in .bb would be a false assumption Jan 29 12:44:09 well, what you want will be completely nondeterminstic Jan 29 12:44:20 it will depend on the contents of your tarball Jan 29 12:44:27 and all users will have different versions Jan 29 12:44:30 since your tarball changes Jan 29 12:44:33 that will lead to chaos Jan 29 12:44:39 exactly: unstable Jan 29 12:44:43 no Jan 29 12:44:45 worse than unstable Jan 29 12:44:47 unpredictable Jan 29 12:44:51 since it won't redownload Jan 29 12:44:55 once it's in the source cache Jan 29 12:45:05 are you shure oe is in all places that predictable? i dont think so Jan 29 12:45:22 XorA: about connman, is there a way to avoid to pull-in bash ? resolv.conf was depending on it last time I checked) Jan 29 12:45:28 well, at least we try to make it so :) Jan 29 12:45:40 ant_work: dont use resolv.conf or dhclient Jan 29 12:45:45 mickeyl: doubt that ;) Jan 29 12:45:53 ant_work: plugins Jan 29 12:46:20 but hal doesnt seem to recognise the tosa wifi :-( Jan 29 12:46:48 hal does not recognise many things Jan 29 12:48:00 of course we have little control over upstream Jan 29 12:48:10 but we try not to add packages that link to source files that change Jan 29 12:48:18 and usually upstream authors don't do that Jan 29 12:48:56 pyneod/zad is in a quite changing state currently, tagging does not make any sense at al Jan 29 12:50:47 well, in that case you might better keep it as a local patch Jan 29 12:51:03 i'm not trying to block it, but please see the point Jan 29 12:51:41 we might as well ask on oe-devel Jan 29 12:51:46 if the others don't mind, i'm fine Jan 29 12:51:56 hey, shure i see the point... it would be so much easier for others to follow the development Jan 29 12:52:20 i promise there will be a "names" release soon :) Jan 29 12:53:33 I would just use github D: Jan 29 12:54:27 http://rafb.net/p/6vMo7m85.html Jan 29 12:54:50 can someone suggest how I can modify EXTAR_OECONF based on machine Jan 29 12:55:34 DUMMY_OECONF = "" Jan 29 12:55:41 DUMMY_OECONF_machine = "blah" Jan 29 12:55:44 I really do not want to hack at gnuradio autofoo Jan 29 12:55:56 EXTRA_OECONF = "xxx yyy ${DUMMY_OECONF}" Jan 29 12:56:01 hmmm, so I can't just add a bit Jan 29 12:56:10 ah Jan 29 12:56:12 ok Jan 29 12:56:17 I'll try this Jan 29 12:56:52 thanks Jan 29 12:57:06 Crofton: youll need to set that package to MACHINE_ARCH as well Jan 29 12:57:13 as its different for beagle to everything else Jan 29 12:57:54 yeah, it is a nuisance because I do not know if armv7 gaurantees NEON Jan 29 12:58:08 how do I set it to machine arch Jan 29 12:58:30 there x86 code has runtime detection of instruction sets Jan 29 12:58:33 PACKAGE_ARCH_beagle = "${MACHINE_ARCH}" Jan 29 12:58:34 I'm not that smart Jan 29 12:58:44 I think from memory Jan 29 12:58:45 even better Jan 29 12:59:11 koen was saying something about cut price armv7 with missing bits Jan 29 12:59:21 like the cheaposs msm7200 series for armv6 Jan 29 12:59:41 more like armv7 can have bits hung off it a random Jan 29 13:00:31 hopefully this gets less messy, but I really need a NEON version of gnuradio for show and tell Jan 29 13:02:43 Koen: the current _armv7a packages assume neon Jan 29 13:02:52 Koen: so if it's using neon it's safe to make it _armv7a Jan 29 13:02:57 urg Jan 29 13:03:04 I am not koen Jan 29 13:03:10 Koen: and angstrom even has that as an override Jan 29 13:03:14 hmm, that will bite us one day Jan 29 13:03:30 Crofton|work: not really, we can make armv7-lite as an arch Jan 29 13:03:31 ok I won't worry aout package arch Jan 29 13:03:39 heh Jan 29 13:03:52 the nomenclature is questionable Jan 29 13:03:54 like for msm7200 we had to make an armv6-msm arch Jan 29 13:04:04 OMAP should be armv7-heavy Jan 29 13:04:45 I'm hust glad not to worry about package arch Jan 29 13:04:56 03Koen Kooi  07org.openembedded.dev * re26046fa3c 10openembedded.git/packages/vlc/libdvdcss_1.2.10.bb: libdvdcss: disable docs Jan 29 13:07:06 03Graeme Gregory  07org.openembedded.dev * r9a2593530b 10openembedded.git/contrib/angstrom/build-feeds.sh: angstrom/build-feeds.sh : add connman and connman-gnome Jan 29 13:10:42 Python/graminit.c:878: error: expected expression at end of input Jan 29 13:10:44 grrrr Jan 29 13:16:38 bbl, bye Jan 29 13:20:32 03Graeme Gregory  07org.openembedded.dev * r0685559e6c 10openembedded.git/packages/connman/connman-gnome_0.4.bb: connman-gnome_0.4.bb : use gtk-icon-cache class Jan 29 13:22:48 ~hail debian win32 installer Jan 29 13:22:50 * ibot bows down to debian win32 installer and chants, "I'M NOT WORTHY!!" Jan 29 13:30:54 03ghost  07org.openembedded.dreambox * r4e63d335fd 10openembedded.git/packages/gstreamer/gst-plugin-dvbmediasink.bb: gst-plugin-dvbmediasink.bb: push SRCDATE Jan 29 14:23:17 mickeyl: wb Jan 29 14:23:22 hi Jan 29 14:23:44 03Stefan Schmidt  07fso/milestone5 * r67ff029b0d 10openembedded.git/packages/images/fso-image.bb: fso-image: Ship paroli Jan 29 14:23:55 03Stefan Schmidt  07fso/milestone5 * r91dbe7892c 10openembedded.git/conf/distro/include/sane-srcrevs.inc: sane-srcrevs.inc: Bump zhone for TTFF display Jan 29 14:24:31 since some days i have a strange effect when building. installing kernel module bails out with "tmp/staging/i686-linux/usr/bin/install-sh: ./-m0644 does not exist. Jan 29 14:24:54 does someone have an idea what happens? Jan 29 14:24:55 ibot: botmail for ant_work it might be a good idea to update xserver-kdrive in oe first - 1.23 comes with most of the necessary patches anyway. Jan 29 14:26:23 emdete: I have seen this several times but don't know why it works sometimes and why ist fails. Jan 29 14:26:42 florian: pfew, i can tell you: here it fails every time ;) Jan 29 14:26:56 florian: always after a complete clean Jan 29 14:27:13 emdete: Adding a space char between the '-m' and the number fixes it. Jan 29 14:27:22 it seams an var is not defined in that shell script Jan 29 14:28:20 no, for some reason the "-m0644" is interpreted as a file name Jan 29 14:28:20 florian: it's not my script... and i think the problem is before the "-m". i think there is a var $INSTALL which is empty Jan 29 14:28:27 ah Jan 29 14:28:51 right that may be the cause Jan 29 14:29:06 is it possible that different install.sh are used? Jan 29 14:29:36 somethgin like this... Jan 29 14:30:06 hm, strange. nobody else has this problem? Jan 29 14:30:43 i also have a problem with the version of intltool, which is build in version 0.35 but some packages request at least 0.40 Jan 29 14:31:43 florian: do you know of the intltool problem too? Jan 29 14:32:35 not, this one is new to me Jan 29 14:32:56 is thes a native package maybe? Jan 29 14:34:37 as far as i remember it complains about both Jan 29 14:34:45 what would that mean? Jan 29 14:36:10 if its a native package it meight be possible it complains about the one on your build machine Jan 29 14:39:23 florian: you mean xserver-kdrive-common 1.23? Jan 29 14:40:17 hrw: no xserver-common. I just noticed that ant_work used 1.22 Jan 29 14:42:25 xserver-common are ugly Jan 29 14:42:45 I think that I know what I will work on in near time Jan 29 14:42:48 cleaning... Jan 29 14:42:52 florian: you might be right, i have both, 40&35 :/ Jan 29 14:43:02 my devices do not need blueprobe for example Jan 29 14:43:03 hrw: patches are very very welcome Jan 29 14:43:12 florian: so pinpointig the native one to .40 is okay..? Jan 29 14:43:30 florian: when there is a native bb why doesn bb grab that? Jan 29 14:43:44 hrw: yes I have a device hiere that might wnat to get rid of this too Jan 29 14:44:05 florian: blueprobe is ipaq+acer n30 only Jan 29 14:44:11 ok. and few htc. Jan 29 14:44:11 emdete: i have no idea, that was just a guess. Jan 29 14:44:23 florian: which seem to be correct :) Jan 29 14:44:45 hrw: we could make it depend on if there is a bluetooth entry in the features Jan 29 14:44:57 hmm mo Jan 29 14:45:02 florian: it currently is that way Jan 29 14:45:13 i get the point... Jan 29 14:45:38 florian: I am rather thinking more about MACHINE_FEATURES_RDEPENDS_BLUETOOTH += "blueprobe" for them Jan 29 14:45:58 hrw: that would be an option Jan 29 14:46:13 hi prpplague^2 Jan 29 14:47:07 prpplague^2: hi, is your battery... eh device ready for sale now? :) Jan 29 14:49:20 hrm the "branding" of my crosscompiler starts to annoy me Jan 29 14:50:04 * hrw -> afk Jan 29 14:50:13 hrw|afk: cu Jan 29 14:54:08 florian: grep finds 65 occurences of a -m directly followed by the mode in openembeddeds tree Jan 29 14:56:50 hi prpplague Jan 29 14:56:58 florian: hey bud Jan 29 14:57:15 prpplague: you seem to have some kind of fork here :) Jan 29 14:58:13 florian: ?? re my nick? Jan 29 14:58:53 prpplague: yes Jan 29 14:59:00 hi cbrake Jan 29 14:59:17 florian: yea, forgot and left my pc at the house logged in Jan 29 14:59:38 :-) Jan 29 15:02:00 florian: you where right: its a -m0... problem, which gets solved by a patch (openembedded/packages/linux/linux-omap-2.6.28/fix-install.patch) but that didn't get applied :/ Jan 29 15:22:36 hi florian Jan 29 15:22:53 pb_: hi Jan 29 15:23:27 thebohemian: hi - any news for the e.V.? Jan 29 15:27:12 03Sergey Lapin  07org.openembedded.dev * r73fc517dec 10openembedded.git/packages/libpcap/ (4 files in 2 dirs): Jan 29 15:27:12 libpcap: added 1.0.0 Jan 29 15:27:12 Signed-off-by: Sergey Lapin Jan 29 15:27:13 03Sergey Lapin  07org.openembedded.dev * r755d08e98c 10openembedded.git/packages/tcpdump/ (3 files in 2 dirs): tcpdump: added 4.0.0, minimal support for IEEE802.15.4 capture Jan 29 15:35:40 does someone know why tune2fs left the system? Jan 29 15:40:20 03Philip Balister  07org.openembedded.dev * rb93e93b336 10openembedded.git/packages/gnuradio/gnuradio.inc: Jan 29 15:40:20 gnuradio : Pass machine specific optimization selection information to Jan 29 15:40:20 gnuradio configure script. Jan 29 15:40:20 NOTE: the current arm option needs changing to neon in the Jan 29 15:40:20 gnuradio neon patch. Jan 29 15:40:31 03Richard Purdie  07ieee80215 * r58fb511d5a 10openembedded.git/classes/siteinfo.bbclass: siteinfo.bbclass: Add cygwin and mingw entries Jan 29 15:40:32 03Richard Purdie  07ieee80215 * r5563af5f6c 10openembedded.git/ (4 files in 2 dirs): bitbake.conf: Add EXEEXT variables for BUILD/HOST/TARGET so we can start to handle other architectures Jan 29 15:40:34 03Richard Purdie  07ieee80215 * r56ff792ea6 10openembedded.git/conf/bitbake.conf: Add a shortcut for the EXEEXT we'll use in packaging Jan 29 15:40:34 03Tom Rini  07ieee80215 * rc460a93935 10openembedded.git/classes/insane.bbclass: insane.bbclass: Fix thinko in package_qa_check_staged logic for iscrossnative Jan 29 15:40:36 03Richard Purdie  07ieee80215 * raabadc9673 10openembedded.git/classes/insane.bbclass: insane.bbclass: Add handling for canadian classes Jan 29 15:40:41 03Tom Rini  07ieee80215 * rdbd20e4123 10openembedded.git/packages/mingw/ (4 files): Jan 29 15:40:44 mingw-w32api: Add main and -headers package for 3.13 Jan 29 15:40:46 We use some .inc files for the real logic for easy upgrades Jan 29 15:40:48 03Tom Rini  07ieee80215 * re602f0edfa 10openembedded.git/classes/ (4 files): canadian sdk: Add classes/canadian{,-cross,-native,-sdk}.bbclass Jan 29 15:40:51 03Tom Rini  07ieee80215 * r6c7b5a8183 10openembedded.git/packages/binutils/mingw-binutils_2.17.50-20060824-1.bb: mingw-binutils: Add a special version of binutils for building MinGW Jan 29 15:40:54 03Tom Rini  07ieee80215 * r5484efc246 10openembedded.git/packages/mpfr/mpfr-canadian_2.3.1.bb: mpfr-canadian: Add a copy of mpfr for use in Canadian builds Jan 29 15:40:57 03Tom Rini  07ieee80215 * rfaf3b90495 10openembedded.git/packages/tasks/task-sdk-canadian-host.bb: task-sdk-canadian-host: Add a task for the Canadian host packages Jan 29 15:41:08 03Tom Rini  07ieee80215 * r9fd597b68e 10openembedded.git/packages/gcc/ (4 files in 2 dirs): mingw-gcc-cross, mingw-gcc-cross-initial: Bring in the special gcc for MinGW Jan 29 15:41:11 03Tom Rini  07ieee80215 * r84c71aebc0 10openembedded.git/packages/ (2 files in 2 dirs): mingw-binutils-canadian-cross, mingw-gcc: Add missing Canadian SDK recipes Jan 29 15:41:14 03Tom Rini  07ieee80215 * r73fba6a84a 10openembedded.git/packages/mingw/ (4 files): Jan 29 15:41:17 mingw-runtime: Add main and -headers package for 3.15.1 Jan 29 15:41:19 We use some .inc files for the real logic for easy upgrades Jan 29 15:41:21 03Tom Rini  07ieee80215 * r2378bfca9b 10openembedded.git/packages/gcc/ (8 files in 2 dirs): Jan 29 15:41:24 gcc-canadian-sdk: Add a version of gcc 4.2.4 for building Canadian SDKs. Jan 29 15:41:26 With this, add some patches specific to fixing up MinGW issues on 4.2.x Jan 29 15:41:38 03Tom Rini  07ieee80215 * rc90263a63f 10openembedded.git/packages/gdb/ (5 files in 2 dirs): Jan 29 15:41:41 gdb-canadian-sdk: Add a version of gdb 6.8 for building Canadian SDKs. Jan 29 15:41:43 With this, add some patches specific to fixing up MinGW issues on some targets Jan 29 15:41:45 03Tom Rini  07ieee80215 * r04c9615b93 10openembedded.git/ (4 files in 3 dirs): Jan 29 15:41:47 SDK: Rename 'SDK_PREFIX' to 'SDK_PATH' so that it has a meaning analogous to BUILD/TARGET/HOST Jan 29 15:41:49 This isn't a big deal until we introduce the rest of the SDK variables for Jan 29 15:41:51 doing Canadian builds. Jan 29 15:41:53 03Stefan Schmidt  07ieee80215 * r410e41ec32 10openembedded.git/ (2 files in 2 dirs): angstrom.inc: Catch up with openmoko kernel recipe renames. Jan 29 15:41:54 * Tartarus wonders what CIA-2 is up to... Jan 29 15:41:58 03Tom Rini  07ieee80215 * r09a97024fa 10openembedded.git/packages/gmp/gmp-canadian_4.2.4.bb: gmp-canadian: Add a copy of gmp for use in Canadian builds Jan 29 15:42:07 03Frans Meulenbroeks  07ieee80215 * ra554b2571b 10openembedded.git/ (3 files in 2 dirs): libtorrent: updated to libtorrent 0.12.4 and rtorrent 0.8.4 Jan 29 15:42:12 03Philip Balister  07ieee80215 * rf430bac83b 10openembedded.git/packages/gnuradio/gnuradio.inc: gnuradio : Install python files in the proper directory. Jan 29 15:42:15 03Koen Kooi  07ieee80215 * r5015fa7658 10openembedded.git/packages/linux/ (9 files in 5 dirs): linux-handhelds: remove some stale versions Jan 29 15:42:18 03Koen Kooi  07ieee80215 * r26f47dc8f2 10openembedded.git/ (conf/checksums.ini packages/vlc/x264_0.0.bb): x264: add 20090127 snapshot Jan 29 15:42:21 (112 lines omitted) Jan 29 15:42:23 03Sergey Lapin  07ieee80215 * r73fc517dec 10openembedded.git/packages/libpcap/ (4 files in 2 dirs): Jan 29 15:42:30 libpcap: added 1.0.0 Jan 29 15:42:32 Signed-off-by: Sergey Lapin Jan 29 15:42:34 03Sergey Lapin  07ieee80215 * r755d08e98c 10openembedded.git/packages/tcpdump/ (3 files in 2 dirs): tcpdump: added 4.0.0, minimal support for IEEE802.15.4 capture Jan 29 15:42:43 03Koen Kooi  07org.openembedded.dev * rccf05d43d6 10openembedded.git/conf/machine/beagleboard.conf: beagleboard: remove kernel-modules meta-package , those have no place in machine.conf Jan 29 15:46:36 cia-2 got nuts, it is barfing out alot of messages it did before Jan 29 15:46:53 watch the branch Jan 29 15:47:35 ah mickeyl, you're right (as usual) Jan 29 15:48:23 ;) Jan 29 15:50:17 mickeyl: any idea where tune2fs is gone? Jan 29 15:50:30 no Jan 29 15:50:32 where was it Jan 29 15:50:35 and where is it missing? Jan 29 15:50:55 in e2fs-utils or similar. its gone since some days Jan 29 15:51:11 may be ...-tools Jan 29 15:53:08 Packaged contents of e2fsprogs-tune2fs into /local/pkg/oe/fic-gta02/tmp/deploy/glibc/ipk/armv4t/e2fsprogs-tune2fs_1.38-r17.1_armv4t.ipk Jan 29 15:53:23 mickey@localhost:/local/pkg/oe/fic-gta02$ ipkgfiles /local/pkg/oe/fic-gta02/tmp/deploy/glibc/ipk/armv4t/e2fsprogs-tune2fs_1.38-r17.1_armv4t.ipkdrwxr-xr-x root/root 0 2009-01-29 16:53 ./ Jan 29 15:53:23 drwxr-xr-x root/root 0 2009-01-29 16:52 ./sbin/ Jan 29 15:53:23 -rwxr-xr-x root/root 89800 2009-01-29 16:52 ./sbin/tune2fs Jan 29 15:53:23 lrwxrwxrwx root/root 0 2009-01-29 16:52 ./sbin/findfs -> tune2fs Jan 29 15:53:23 lrwxrwxrwx root/root 0 2009-01-29 16:52 ./sbin/e2label -> tune2fs Jan 29 15:53:28 looks pretty 'there' to me Jan 29 15:56:30 03Michael 'Mickey' Lauer  07org.openembedded.dev * rc866349385 10openembedded.git/packages/e2fsprogs/ (e2fsprogs/e2fsprogs_1.34.bb e2fsprogs_1.41.2.bb): e2fsprogs: remove bogus recipe in files directory; move 1.41.2 to nonworking (patches not found/not applying) Jan 29 15:59:19 florian: unfortunately not. I am waiting for a letter from the court :| Jan 29 16:03:17 * cbrake notes a good bit of the opensuse buildservice is written in rails ... Jan 29 16:04:02 thebohemian: if they want to get somethign german officials are faster usually ;) Jan 29 16:14:13 * florian thinks kc needs a better internet connection Jan 29 16:22:26 re Jan 29 16:29:33 03Stefan Schmidt  07fso/milestone5 * rede43b1be9 10openembedded.git/packages/tasks/task-fso-compliance.bb: task-fso-compliance: Leave out connman-ethernet as it breaks the usb ethernet setup Jan 29 16:29:44 03Stefan Schmidt  07fso/milestone5 * rf272038356 10openembedded.git/conf/distro/include/sane-srcrevs.inc: sane-srcrevs.inc: bump frameworkd for sound profile fix Jan 29 16:29:44 03Stefan Schmidt  07fso/milestone5 * r5ea40b4298 10openembedded.git/packages/numptyphysics/numptyphysics/numptyphysics.desktop: numptyphysics: Remove xrandr calls from desktop file as they break xglamo Jan 29 16:29:58 stefan_schmidt: I hate you. Jan 29 16:30:01 I just regenerated an image. Jan 29 16:30:11 Ainulindale: thanks Jan 29 16:30:13 :-) Jan 29 16:30:17 How's life stefan_schmidt? Jan 29 16:30:25 * stefan_schmidt just cames from a massage, totally relaxed :D Jan 29 16:30:56 Was it a real massage or a "quickest way to pleasure" kind of message? Jan 29 16:31:03 Ainulindale: We postponed MS5 until monday so I feel more confident that we actually test it Jan 29 16:31:29 Ainulindale: nah, just a statement that I'm relaxed. Jan 29 16:31:31 stefan_schmidt: We reverted everything here but two patches, fastboot wise. Jan 29 16:31:36 could people committing to .dev stop building with checksums turned off? Jan 29 16:31:49 I'll build, check that everything is working then push to testing. Jan 29 16:32:11 Ainulindale: ok, good to know. We need to sort this out after the release. The revert was for the release branch only. Fastest method. Jan 29 16:32:35 I looked at the patches for over an hour but did not find why they break !gta02 devices here Jan 29 16:32:39 stefan_schmidt: IMHO your problem isn't related Jan 29 16:32:48 Because we built GTA01 images with the fastboot patches Jan 29 16:32:58 Something broke, but I'm not sure it's fastboot Jan 29 16:33:25 Ainulindale: Well, related in terms of no working image before and after reverting the merge it works again. :) Jan 29 16:33:49 Ainulindale: Of course it can be stuff that is related with it. Jan 29 16:33:49 Yes indeed but I think it's coincidental somehow Jan 29 16:33:53 sure Jan 29 16:34:11 As we can check with our git and builds that we had fastboot patches in december and built several GTA01 images Jan 29 16:34:14 That's to sort out when I bring more stuff back from the ms5 branch Jan 29 16:34:27 Yep :-) Jan 29 16:34:40 Ainulindale: That was one the reasons why I did not get it. It should have worked Jan 29 16:35:04 stefan_schmidt: Did you think about changing the kernel rev? Jan 29 16:35:09 For the USB BSOD on windows fix Jan 29 16:35:55 Ainulindale: Not for the release. There are more good bug fixes in it, but I don 't trust them enough for not bringing regressions in. Jan 29 16:36:02 Risc management. :D Jan 29 16:36:06 Heh Jan 29 16:36:11 Can't you just cherry pick it? Jan 29 16:36:18 (I'm m irroring the discussion in cdevel here :-) ) Jan 29 16:36:22 After MS5 we will need to bump kernel, EFL and other stuff to have breakage again :) Jan 29 16:36:50 stefan_schmidt: I don't think you should do that you know Jan 29 16:36:53 We're doing that in SHR Jan 29 16:36:54 Ainulindale: We have no own kernel tree. Would be patches in OE then. Not really eager to do it. Jan 29 16:37:04 You should focus on development and ask what you need Jan 29 16:37:16 Maybe Jan 29 16:37:20 Would be better Jan 29 16:37:20 (w.r.t. EFL/Kernel bumps) Jan 29 16:37:25 I'm sure it'll save some time Jan 29 16:37:32 And as we're going to commit real soon Jan 29 16:37:35 But people are always screaming if we are not bleeding edge :) Jan 29 16:38:07 Same here Jan 29 16:38:12 That's why we have unstable and testing Jan 29 16:38:30 We should spent less time on distro things. We agree on this. Sadly OM does not do it for us either Jan 29 16:39:00 That's why we're here Jan 29 16:39:05 I don't think that's OM job Jan 29 16:39:42 I think it is. But that will not change it. :D Jan 29 16:39:55 stefan_schmidt: well IMHO Jan 29 16:40:02 I don't think that producing software at all is OM job Jan 29 16:40:18 Hardware, kernel, and nothing more. Jan 29 16:41:21 Ainulindale: ok, if you go that way then it matches. But once they do software above the kernel they also should care about distro. Jan 29 16:41:36 And trust me. They will not stop to do software Jan 29 16:41:39 Well, I disagree again :-) Jan 29 16:41:41 (about distro) Jan 29 16:41:46 (but that's only my opinion, again) Jan 29 16:41:49 As for the software, I know Jan 29 16:41:51 I think it's dumb Jan 29 16:42:06 Funding software, ok Jan 29 16:42:09 * stefan_schmidt flashes a new kernel and rootfs to test if the closed bugs are really fixed Jan 29 16:42:15 Producing software in OM name, dumb :-) Jan 29 16:42:39 It's a product strategy in sense of hardware+software = product Jan 29 16:42:55 Well Jan 29 16:43:03 I think I told you my feelings about that quite at length Jan 29 16:43:06 But hey, we don't need to discuss this. We are on the same page. Jan 29 16:43:11 (heh :-) ) Jan 29 16:43:15 sure, no need to repeat Jan 29 16:43:25 Anyway, to sum up Jan 29 16:43:33 SHR is currently using FSO plus some patches Jan 29 16:43:56 Ainulindale: That good. Jan 29 16:44:10 So it'll speed up the "debugging process" I hope Jan 29 16:44:11 Ainulindale: btw, any tweak for numpty in SHR? Jan 29 16:44:21 Yep Jan 29 16:44:25 Namely the xrandr thing Jan 29 16:44:28 Ainulindale: I had to remove the xrandr calls as they let xglamo fail Jan 29 16:44:29 Which you removed Jan 29 16:44:34 hmm, ok Jan 29 16:44:41 it still did not start here Jan 29 16:44:46 The RDEPENDS on libpng too Jan 29 16:44:46 Complaining about an image Jan 29 16:44:51 !logs Jan 29 16:44:53 Which I should have fixed long ago Jan 29 16:44:54 Ah Jan 29 16:44:57 ~logs Jan 29 16:44:58 All conversations are logged to http://ibot.rikers.org/channel, where "channel" is replaced by the URL-encoded channel name, such as %23freenode for #freenode. Lines starting with spaces are not logged. Jan 29 16:45:03 Will check this, thanks Jan 29 16:45:08 stefan_schmidt: what sort of problems were observed with fastboot? Jan 29 16:45:09 stefan_schmidt: well that's the thing you're missing Jan 29 16:45:12 stefan_schmidt: http://git.shr-project.org/git/?p=shr-overlay.git;a=blob;f=patches/general/0004-numptyphysics-depend-on-libpng.patch;h=afe9f0daca657404e2bd677a73197ee4a60ff95f;hb=HEAD Jan 29 16:45:28 mwester: GTA01 image not compiling Jan 29 16:45:43 compilation errors with fastboot? Odd. Jan 29 16:45:44 mwester: On gta01 and other non-gta02 devices I had no busybox installed neither sysvinit Jan 29 16:45:52 wow. Jan 29 16:45:53 Ok. Jan 29 16:46:02 sysvinit that's mostly normal Jan 29 16:46:08 busybox, I can't see why Jan 29 16:46:08 mwester: I tested on gta01 and a780 Jan 29 16:46:13 livelog (http://hentges.net/tmp/logs/irc/livelogs/%23oe.livelog) is still broken btw Jan 29 16:46:20 Then I will not bother to integrate fastboot stuff into SlugOS until this gets sorted. Jan 29 16:46:24 Me neither, I really banged my head on the table Jan 29 16:46:44 mwester: We need to find out what happened later. Jan 29 16:46:52 Jay7: There are logs maintained on nslu2-linux.org as well. Jan 29 16:46:56 It's not a good situation right now Jan 29 16:47:20 right. then the revert was a good action. Jan 29 16:48:21 mwester: Well, I only did it in the fso/ms5 branch right now as I was uncomfortable with breaking other things Jan 29 16:48:36 mwester: It seems only distro=openmoko related to only we suffer anyway Jan 29 16:49:03 Ainulindale: The RDEPENDS is really missing. I'll take the SHR patch and push it in ms5 and dev Jan 29 16:49:31 Nice, thanks Jan 29 16:49:49 stefan_schmidt: we're going to build a real OE distro by the way Jan 29 16:50:25 Ainulindale: A complete distro? Jan 29 16:50:50 stefan_schmidt: well we'll fork distro=openmoko Jan 29 16:51:18 Ainulindale: hmm, ok Jan 29 16:51:23 Ainulindale: For what? Jan 29 16:51:24 Is there some xorg-kdrive experts ? I Have a question about keymaps and modifiers keymaps : I don't understand why on my embedded target, the keymap and modifierskeymap is empty. Whereas on my host with my Xorg server, it's full of keysyms. Since I would like to send fakekeys to Xserver, I need first to fill in the keymap. Is there a common way to do this ? Jan 29 16:52:00 stefan_schmidt: to own our future :-p Jan 29 16:52:08 03Klaus Kurzmann  07fso/milestone5 * r5793c57c07 10openembedded.git/packages/numptyphysics/numptyphysics_svn.bb: numptyphysics: depend on libpng Jan 29 16:52:12 Ainulindale: heh Jan 29 16:52:48 03Klaus Kurzmann  07org.openembedded.dev * r0ac38d6951 10openembedded.git/packages/numptyphysics/numptyphysics_svn.bb: numptyphysics: depend on libpng Jan 29 16:54:05 03Michael 'Mickey' Lauer  07fso/milestone5 * re8fc49f319 10openembedded.git/packages/bluez/ (2 files in 2 dirs): bluez4: add compile patch fixing some usb declaration mismatch Jan 29 16:55:25 * stefan_schmidt takes a break Jan 29 16:57:54 drakou: what is your machine? Jan 29 16:58:12 and distro Jan 29 16:58:26 Arm9 custom board with an angstrom linux built with OE Jan 29 16:58:37 and no keyboard Jan 29 16:58:58 maybe I should use xmodmap to load a default keymap Jan 29 17:00:02 03Koen Kooi  07org.openembedded.dev * re58876afd0 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Jan 29 17:00:06 try to use xmodmap -pke > xmodmaprc on desktop and xmodmap - < xmodmaprc on your board Jan 29 17:00:09 03Koen Kooi  07org.openembedded.dev * r7abf3f1827 10openembedded.git/packages/tasks/task-base.bb: task-base-usbhost: add asix and pegasus modules Jan 29 17:00:43 03Koen Kooi  07org.openembedded.dev * rdbf490f1fb 10openembedded.git/conf/checksums.ini: checksums: add libpcap Jan 29 17:02:30 Jay7, Is it normal to have an empty keymap ? I just launch Xfbdev by hand then my programs Jan 29 17:03:42 drakou: I'm not sure but in general no.. Jan 29 17:04:38 03Stefan Schmidt  07fso/milestone5 * r2a4c3b782e 10openembedded.git/packages/numptyphysics/numptyphysics_svn.bb: numptyphysics: Bump PR for xrandr and libpng change Jan 29 17:04:48 03Stefan Schmidt  07fso/milestone5 * rbf035c64c4 10openembedded.git/: Merge branch 'fso/milestone5' of git@git.openembedded.net:openembedded into fso/milestone5 Jan 29 17:16:55 Jay7, If i launch xmodmap -pke on the target the result is an empty list of keycodes. Maybe it's possible to include a default keymap at xorg-kdrive build time ? Jan 29 17:17:33 drakou: do xmodmap -pke on your desktop :) Jan 29 17:17:52 then transfer file and do xmodmap - < file on board Jan 29 17:18:59 btw, when you have no keyboard attached to your board why do you needed xmodmap? Jan 29 17:21:59 Jay7, because I want to use a virtual keyboard such as matchbox-keyboard and this type of keyboard use XTest extensions to send fakekeys to Xserver, so we need a mapping between keycodes and keysyms. Jan 29 17:22:15 ah.. Jan 29 17:22:29 ok, try to transfer xmodmap file from desktop then Jan 29 17:23:27 or better run xserver 1.3.0.0 on target, dump xmodmap, use 1.4 and give that xmodmap Jan 29 17:24:50 hrw, what's the difference between 1.3 and 1.4 ? Jan 29 17:25:02 1.3 has internal keymap Jan 29 17:26:12 hrw: you have any contacts for the openmoko stuff? Jan 29 17:27:30 prpplague: no Jan 29 17:28:06 hrw: hmm, ok thanks Jan 29 17:29:14 My problem is : On target There are 4 KeySyms per KeyCode; KeyCodes range from 8 to 255. And on my desktop : There are 10 KeySyms per KeyCode; KeyCodes range from 8 to 255 Jan 29 17:30:02 I am really surprised of that issue... Everybody here using kdrive and a virtual keyboard should have that problem also Jan 29 17:30:24 I have 5 per keycode Jan 29 17:32:12 Is it possble to change that ? I didn't make any modification to kdrive bb files from angstrom distribution. Jan 29 17:35:06 I loaded the desktop file , It changed to 8 keysyms by keycode. Jan 29 17:45:45 see you in ~1-1.5h Jan 29 18:01:34 03Jeremy Lainé  07org.openembedded.dev * rd28c359e62 10openembedded.git/conf/checksums.ini: checksums.ini: add entry for tcpdump 4.0.0 Jan 29 19:18:29 did bitbake ever get moved to git? Jan 29 19:19:36 kergoth: not that I've heard Jan 29 19:19:52 aw Jan 29 19:31:16 03Koen Kooi  07org.openembedded.dev * r685a829091 10openembedded.git/packages/hal/ (hal.inc hal_0.5.11.bb): Jan 29 19:31:16 hal, fix packaging: Jan 29 19:31:16 NOTE: the following files were installed but not shipped in any package: Jan 29 19:31:16 NOTE: /usr/bin/hal-disable-polling Jan 29 19:31:16 NOTE: /usr/bin/hal-setup-keymap Jan 29 19:37:08 re Jan 29 20:01:03 http://www.youtube.com/watch?v=ofpoyXyfi2U Jan 29 20:01:37 http://www.youtube.com/watch?v=zymOmduNWyI Jan 29 20:09:32 that is cool. where do i buy one. Jan 29 20:09:39 the projector Jan 29 20:09:40 :) Jan 29 20:14:10 digikey :) Jan 29 20:14:20 im looking at the resellers already :) Jan 29 20:21:01 03Philip Balister  07org.openembedded.dev * re3792ea343 10openembedded.git/packages/images/gnuradio-image.bb: gnuradio-image : Add image to simplify testing gnuradio on OMAP3 boards. Jan 29 20:29:44 hmm, might just get a beagle too. Jan 29 20:30:08 ive had a hole in the gadget collection since i got rid of the gumstix. Jan 29 20:30:43 can it actually play a HD video? Jan 29 20:37:56 03Koen Kooi  07org.openembedded.dev * r6a6d6bab51 10openembedded.git/ (3 files in 2 dirs): bigbuckbunny: add recipes for 480p and 720p mpeg4 asp versions Jan 29 20:42:51 I'm patching packages/grub/grub_0.97.bb Jan 29 20:45:39 in do_install_append() I want to copy a file from openembedded/packages/grub/grub-0.97/menu.lst to ${D}/boot/grub Jan 29 20:46:01 what is proper syntax for openembedded/packages/grub/grub-0.97/menu.lst ? Jan 29 20:53:16 maybe {PN} ? Jan 29 20:57:02 Well Jan 29 20:57:08 it needs to be in SRC_URI Jan 29 20:57:15 then it's ${WORKDIR}/menu.lst Jan 29 20:58:13 Tartarus: ${WORKDIR} is openembedded/packages/grub/grub-0.97 ? AFAIK is /tmp/work Jan 29 20:58:57 menu.lst is a part of recipe I added in openembedded/packages Jan 29 20:59:11 If menu.lst is in SRC_URI it gets copied to ${WORKDIR} Jan 29 20:59:17 So you refer to it as ${WORKDIR}/menu.lst Jan 29 20:59:51 Tartarus: so SRC_URI is openembedded/packages/grub/grub-0.97 ? Jan 29 20:59:55 No Jan 29 21:00:00 in SRC_URI you add file://menu.lst Jan 29 21:04:11 Tartarus: I read SRC_URI is the list of source files, what is the path? Jan 29 21:05:57 There is no variable that refers back to it, that you can use Jan 29 21:05:59 maybe I simly have to use cp file://menu.lst ${D}/boot/grub Jan 29 21:06:09 What you do is list your file in there so it gets copied into WORKDIR and you refernce that Jan 29 21:06:37 You put SRC_URI += "file://menu.lst" Jan 29 21:06:49 then in do_install_append cp ${WORKDIR} ${D}/boot/grub/ Jan 29 21:06:52 aha, I understand Jan 29 21:06:56 er ${WORKDIR}/menu.lst Jan 29 21:09:54 Tartarus: thanks Jan 29 21:39:31 how would one make KERNEL_CCSUFFIX_armv7a dependent on kernels <2.6.27? Jan 29 21:40:00 03Stefan Schmidt  07fso/milestone5 * r57d160dddf 10openembedded.git/packages/tasks/task-fso-compliance.bb: task-fso-compliance: Bump PR to really avoid connman-plugin-ethernet in the image Jan 29 21:40:25 ant__: hey Jan 29 21:40:30 Is it a problem on newer ones or just not needed? Jan 29 21:42:33 Jay7: hi there Jan 29 21:43:05 ant__: I'm refactoring kexecboot :) Jan 29 21:43:18 Tartarus, was that for me? Jan 29 21:43:25 created separate util.c/h Jan 29 21:43:27 :1 Jan 29 21:44:12 Jay7: I'm thinking that perhaps is the moment to switch to kernel in /boot for clamshells too... Jan 29 21:44:31 yep Jan 29 21:44:38 Xora committed tosa today that way Jan 29 21:44:46 there is trend to move to kexecboot at all :) Jan 29 21:44:53 collie and poodle aready there Jan 29 21:44:56 for zaurus machines Jan 29 21:44:58 :-) Jan 29 21:45:03 not only... Jan 29 21:45:04 read here Jan 29 21:45:27 my name was used :-) Jan 29 21:45:31 tharvey, yes Jan 29 21:45:32 http://www.mail-archive.com/coreboot@coreboot.org/msg04413.html Jan 29 21:45:38 XorA: hey Jan 29 21:46:32 XorA: what about +require conf/machine/include/kexecboot.inc Jan 29 21:46:39 in zaurus-2.6.inc ? Jan 29 21:47:06 ant__: probably makes sense now all Z's are kexecbooted Jan 29 21:47:09 Advanced Micro Devices, Inc.? Jan 29 21:47:24 Jay7: make a call there Jan 29 21:47:27 :-) Jan 29 21:47:37 ant__: that is good news :) Jan 29 21:47:54 was months ago..even Jan 29 21:47:55 hey - lookit me! Jan 29 21:48:19 ant__: yeah, but now kexecboot is better :) Jan 29 21:48:37 Tartarus, ah... well I'm told that the csl toolchain is required for kernels <2.6.27 and not otherwise, and b/c I'm also told that the csl toolchain is pretty broken was wondering how to specify KERNEL_CCSUFFIX_armv7a (which angstrom-2008.1.conf defines as '-4.2.1+csl-arm-2007q3-53') something that differs depending on kernel version Jan 29 21:49:06 What's magic about newer kernels? Jan 29 21:49:20 Can someone explain to a dump, me in this context, what kexecboot is being used for? Jan 29 21:49:26 I would think it's more of a "we now have a sane toolchain" thing than a kernel thing Jan 29 21:49:28 ;) Jan 29 21:49:44 And fwiw - I still think kexecboot would be happy in coreboot land Jan 29 21:50:01 Tartarus, not sure... just something koen mentioned when I was asking why csl was used for linux-omap kernel Jan 29 21:50:42 right... perhaps whatever triggered the issue <2.6.27 was simply changed in source Jan 29 21:50:46 Well that's not a full answer from him :( Jan 29 21:51:16 s,dump,dumb, Jan 29 21:51:40 otavio_: kexecboot is used to load and boot 'big kernel' from any supported by linux kenel media Jan 29 21:51:40 Sounds like a gcc patch or 12 is needed on older gccs Jan 29 21:52:10 Jay7: so you have small partition that is loaded Jan 29 21:52:12 otavio_: we using it on zauruses where max kernel size with sharp loader is ~1.4Mb Jan 29 21:52:34 Jay7: it has a linux kernel and this one loads another one? Jan 29 21:52:37 Jay7: ok, got it Jan 29 21:52:47 but in general, how would I specify such a thing if I wanted a var dependent on a specific kernel version? just override it in the various kernel recipes? Jan 29 21:53:08 What about the fastboot work done by Moblin people? Jan 29 21:53:13 otavio_: yes, it just small excerpt of linux kernel Jan 29 21:53:26 Is someone trying to make it work in OE distros? Jan 29 21:53:34 otavio_: which kexec's other big kernel from selected (in GUI) partition Jan 29 21:54:00 Jay7: that is really interesting :-) how about stability? Jan 29 21:54:13 otavio_: looking good :) Jan 29 21:54:16 Jay7: or is mostly looks like a raw boot? Jan 29 21:54:24 hm.. Jan 29 21:54:57 I should make some 'screenshots' :) Jan 29 21:55:05 Jay7: for kernel POV it can't be the same, I imagine ... so few stability issues might be involved. :-) Am I wrong? Jan 29 21:55:08 tharvey: I believe that's the only way... Jan 29 21:55:10 Jay7: I might need to fiddle with the fb_render function for tosa Jan 29 21:55:11 Jay7: yes, indeed Jan 29 21:55:53 XorA: It's not my code but I can try to help :) Jan 29 21:56:19 Jay7: looks like that memcpy is doing 8bit writes, that breaks tosa framebuffer Jan 29 21:56:36 fwiw I'm pretty sure I stumbled in Dmitry's patches for that yellow lines Jan 29 21:56:42 otavio_: It's kexec.. I have not seen/heard any problems with kexec at last time.. Jan 29 21:56:44 somewhere Jan 29 21:56:58 ant__: for the kernel? Jan 29 21:57:08 ant__: there is a patch for kdrive that is a work around Jan 29 21:57:14 he..now that you say i..I've seen 8bit is disabled Jan 29 21:58:27 XorA: hm.. memcpy should just take one area and move to other... Jan 29 21:59:06 Jay7: looks like the klibc one is doing 8 bit writes from the mess on the screen :-) Jan 29 21:59:24 Jay7: I need to investigate more Jan 29 21:59:30 I'll look at Dmitry's patches then into klibc Jan 29 21:59:33 one moment Jan 29 21:59:49 I applied all the patches in the bug tracker Jan 29 22:00:11 hm.. Jan 29 22:00:11 anything more here :-) Jan 29 22:00:27 there is some unapplied patches for kexecboot.. Jan 29 22:00:30 03pieterg  07org.openembedded.dreambox * rfe1302d630 10openembedded.git/packages/initscripts/ (initscripts-1.0/umountnfs.sh initscripts-opendreambox_1.0.bb): umount network filesystems in if-down, instead of from runlevels 0 and 6, when the network is down already (and umount might block) Jan 29 22:00:32 03Felix Domke  07org.openembedded.dreambox * r45d4e28e31 10openembedded.git/packages/tuxbox/tuxbox-common.bb: tuxbox-common: upgrade tuxbox-common to get new satellites.xml and fix typo in timezone list Jan 29 22:00:33 03Felix Domke  07org.openembedded.dreambox * rf268009dca 10openembedded.git/: Merge branch 'opendreambox-pli' of http://git.grimmerink.nl/opendreambox-1.5 into org.openembedded.dreambox Jan 29 22:01:48 XorA: could it be fine as text? Jan 29 22:01:49 +# uncomment this if you want to override the legacy layout Jan 29 22:01:49 +# and deploy kernel in /boot Jan 29 22:01:49 +require conf/machine/include/kexecboot.inc Jan 29 22:02:12 and uncomment it :-) Jan 29 22:02:22 in zaurus-2.6.inc Jan 29 22:02:52 then remove the line from machine.conf Jan 29 22:03:12 well I just checked, klibc uses 8 bit writes for memcpy Jan 29 22:03:20 hmm.. Jan 29 22:03:57 easy enough to fix, I know framebuffer is always 16 bit alligned so can just alter fb_render to not call memcpy just do it in 16 bit chuncks internet Jan 29 22:03:59 internal Jan 29 22:04:28 isn't any command line option like for pc? mtrrd and so on? Jan 29 22:04:32 for arm? Jan 29 22:04:46 ?? Jan 29 22:05:14 http://en.wikipedia.org/wiki/Memory_Type_Range_Registers Jan 29 22:05:45 I was thinking to force it somewhere Jan 29 22:06:05 should be write-combing alike Jan 29 22:06:26 XorA: is void pointers 8bit? Jan 29 22:06:28 but..I'm not exert in the realm Jan 29 22:06:48 Jay7: it internally casts void to char * Jan 29 22:06:59 that is bad.. Jan 29 22:07:04 Jay7: klibc is obviously not optimised for speed Jan 29 22:07:06 ah yes Jan 29 22:07:08 I see Jan 29 22:07:16 I shall just write fb_render using 16 bit transfers Jan 29 22:07:28 wont break anything else, will fix tosa Jan 29 22:07:29 ok, let's do it by self :) Jan 29 22:07:36 gimme 5 mins Jan 29 22:08:12 anyone else used ubifs with the latest kernels? Jan 29 22:08:14 is there any internal type which is 16bit? Jan 29 22:08:23 uint16 Jan 29 22:08:35 Jay7: http://syslinux.zytor.com/wiki/index.php/SYSLINUX#How_do_I_Configure_SYSLINUX.3F Jan 29 22:08:46 is the simplest I've found Jan 29 22:08:47 Jay7: working on the patch, so let me play :-) Jan 29 22:09:27 XorA: ok, I'll wait :) Jan 29 22:09:42 Jay7: should be even easier to parse Jan 29 22:10:09 ant__: anyway we needed to add some our options Jan 29 22:10:14 e.g. order Jan 29 22:10:26 Jay7: but the tags at beginning looks fine Jan 29 22:10:38 KERNEL xyc Jan 29 22:10:43 APPEN blah blah Jan 29 22:10:50 *APPEND Jan 29 22:10:57 but it looks good Jan 29 22:13:52 Jay7: Im cleaning a bit the defconfigs. Trying to 'standardize' the features a bit Jan 29 22:15:00 I'd need to know if akita suffers the 'recharging doesn't start when suspended' bug with linux-rp-2.6.26 Jan 29 22:15:11 and spitz too... Jan 29 22:17:32 dont ask me, sold mine Jan 29 22:17:35 Jay7: other thing, after resume (with kexecboot kernel) the screen is black on c7x0..I have to press one key...would it be possible to come back to bootmenu? Jan 29 22:18:42 XorA: we'll ask Omegamoon/Cortez Jan 29 22:18:51 for spitz Jan 29 22:18:58 ant__: I don't know.. Jan 29 22:19:06 btw did you try his Zubuntu on tosa? Jan 29 22:19:19 ant__: you at FOSDEM? Jan 29 22:19:27 no Jan 29 22:19:39 I'm in Bruxelles until friday 13.30 Jan 29 22:19:42 he Jan 29 22:19:42 ant__: is screen back with 2.6.26 while in normal usage? Jan 29 22:19:58 after suspend Jan 29 22:20:07 yes Jan 29 22:20:31 CoreDump: hello there Jan 29 22:20:33 ant__: leaving 1hr before I arrive then :-) Jan 29 22:20:45 oh..I have friday moring free Jan 29 22:21:03 biz appointment being on 5th Jan 29 22:21:39 we should show kexecboot at FOSDEM :) Jan 29 22:22:50 be careful..remember this cases http://www.theregister.co.uk/2008/08/13/olympics_blue_screen_death/ Jan 29 22:23:07 Ill have a kexecboot tosa at fosdem most likely Jan 29 22:23:24 psychedelic with yellow lines ? Jan 29 22:23:49 8-) Jan 29 22:25:18 just fixed that, nice and clean now Jan 29 22:25:33 XorA: cool! give me a patch :) Jan 29 22:25:33 bravo Jan 29 22:26:13 btw, I should incorporate other kexecboot patches for tosa from Dmitry's mail Jan 29 22:26:25 Jay7: resuming in GPE brings desktop back Jan 29 22:26:30 Jay7: have you got a tosa? Jan 29 22:26:44 03Graeme Gregory  07org.openembedded.dev * rfe90a08540 10openembedded.git/packages/kexecboot/ (kexecboot-0.4/fb-render-16bit.patch kexecboot_0.4.bb): Jan 29 22:26:44 kexecboot_0.4.bb : add patch to always use 16bit writes, this fixes tosa Jan 29 22:26:44 screen corruption and wont harm other machines Jan 29 22:26:48 XorA: no, still have only akita :) Jan 29 22:26:56 Jay7: forward me email then Jan 29 22:27:33 Jay7: and there is your patch Jan 29 22:28:40 XorA: should I forward to this email? ^^^ Jan 29 22:29:19 Jay7: the one in commit message will work :-) Jan 29 22:30:01 XorA: sent Jan 29 22:30:19 but this is relatively old patches Jan 29 22:30:26 so you can already have them Jan 29 22:31:03 they are the patches I applied last night Jan 29 22:32:27 XorA: crofton hinted there could be an issue using oe logo as bootlogo Jan 29 22:32:39 Angstrom has own Jan 29 22:32:55 do you see any issues? Jan 29 22:33:09 ant__: not really kexecboot is distro agnostic Jan 29 22:33:51 XorA: patch is incorporated :) Jan 29 22:34:10 next release should be working out of box :) Jan 29 22:34:35 someone should test other machines than Zaurus too Jan 29 22:37:02 need a machine that is not zaurus first :-) Jan 29 22:37:21 anyway, well, what to do with the Zaurus clamshells? the change is big...you could perhaps take the responsability Jan 29 22:37:40 and go forth Jan 29 22:37:43 not really I dont have any clamshells any more Jan 29 22:38:11 he..official man is RP but ... Jan 29 22:38:21 ant__: post it as an RFC if no-one objects in a week we just do it :-) Jan 29 22:38:33 is there since a week ;-) Jan 29 22:38:44 oh, must have missed the email Jan 29 22:38:51 only started paying attention last night :-) Jan 29 22:38:55 03Koen Kooi  07org.openembedded.dev * r5b9a8a2245 10openembedded.git/packages/angstrom/ (bigbuckbunny-480p.bb bigbuckbunny-720p.bb): bigbuckbunny: fix typos Jan 29 22:39:01 I'd have avoided too...14 commits..scary ;-) Jan 29 22:41:51 XorA: I'll post remaining/open patches as attachment to 4545 Jan 29 22:44:47 done Jan 29 22:45:21 that last 7 patches? :) Jan 29 22:45:37 :-] Jan 29 22:46:47 btw, where I can get /proc/cpuinfo from all zaurus models supported by angstrom? Jan 29 22:47:07 hmm.. Jan 29 22:47:11 bootlog I know Jan 29 22:47:23 not sure about the rest Jan 29 22:47:26 mom Jan 29 22:47:42 can you remember bootlogs url? Jan 29 22:47:57 http://digit.que.ne.jp/visit/index.cgi?59414d41 Jan 29 22:49:41 Jay7: I propose to copy the line about standard mtdparts in machine.conf Jan 29 22:50:04 and assign ROOT_FLASH_SIZE Jan 29 22:50:44 is that needed? Jan 29 22:51:01 03Michael 'Mickey' Lauer  07org.openembedded.dev * rb87d74dc9c 10openembedded.git/conf/machine/include/motorola-ezx-base.inc: motorola-ezx-base.inc: openezx-kernel is the new kernel for all EZX machines Jan 29 22:51:03 03Michael 'Mickey' Lauer  07org.openembedded.dev * r7b598436b8 10openembedded.git/packages/linux/ (218 files in 23 dirs): linux-ezx all: remove unmaintained kernels Jan 29 22:51:10 hardcoded values from sharpsl.c is good for most peoples Jan 29 22:51:24 for me there is no way yet :) Jan 29 22:51:57 ok, we'll try to hack the bootloader oneday Jan 29 22:52:08 cacko-style Jan 29 22:52:34 yes I hope Jan 29 22:53:00 ant__: can you say what is Hardware line in your cpuinfo? Jan 29 22:53:14 Shepherd? Jan 29 22:53:21 no, Husky Jan 29 22:54:21 Machine: SHARP Husky Jan 29 22:54:50 (mine would be Boxer, even) Jan 29 22:55:01 hmhm.. Jan 29 22:55:17 I just trying to clean up machine detection Jan 29 22:55:18 in RP sources some machines are missing Jan 29 22:55:33 and prepare to add other machines (non-sharp) Jan 29 22:56:36 # Jan 29 22:56:36 CPU: XScale-PXA255 [69052d06] revision 6 (ARMv5TE), cr=0000397f Jan 29 22:56:36 # Jan 29 22:56:36 Machine: SHARP Husky Jan 29 22:58:23 ant__: always did think it was cute that sharp had the dog names for their products Jan 29 22:59:21 probably internally-only at the beginning Jan 29 23:00:12 ant__: yea, code names for products so the engineers know which product it is before marketing comes up with gwhiz model numbers Jan 29 23:00:19 but IT is a precise science (where dog's races aren't) Jan 29 23:00:25 :-) Jan 29 23:00:43 hm.. how is Collie detected? Jan 29 23:00:49 woof woof Jan 29 23:01:13 here is no collie: linux/arch/arm/boot/compressed/head-sharpsl.S Jan 29 23:01:16 Jay7: from long hair? Jan 29 23:01:24 ant__: :) Jan 29 23:02:03 isn't SA1100 or smthg? Jan 29 23:02:16 03Stefan Schmidt  07fso/milestone5 * r02bab9ff35 10openembedded.git/packages/images/fso-image.bb: fso-image: Put vagalume into an enterprise distro condition as it needs gst-mp3 Jan 29 23:04:22 yep... there is Jan 29 23:05:46 looks like this: Machine: Sharp-Collie Jan 29 23:06:10 ah, SHARP != Sharp Jan 29 23:06:50 03Stefan Schmidt  07fso/milestone5 * r79a7cb9ef4 10openembedded.git/packages/vagalume/vagalume_0.7.bb: vagalume: Add dbus-x11 as RRECOMMENDS Jan 29 23:12:32 ant__: Converting old-style param struct to taglist Jan 29 23:12:39 in bootlog's :) Jan 29 23:12:50 Linux version 2.4.18-rmk7-pxa3-embedix-021129 Jan 29 23:13:45 yes, that old-style paramlist Jan 29 23:14:27 there was some converting code.. Jan 29 23:14:47 CONFIG_SHARPSL_BOOTLDR_PARAMS Jan 29 23:15:11 never found the patch alone Jan 29 23:16:05 03Michael 'Mickey' Lauer  07fso/milestone5 * rd48a49ba1e 10openembedded.git/packages/linux/ (218 files in 23 dirs): linux-ezx all: remove unmaintained kernels Jan 29 23:16:05 03Michael 'Mickey' Lauer  07fso/milestone5 * rfc521da1d8 10openembedded.git/conf/machine/include/motorola-ezx-base.inc: motorola-ezx-base.inc: openezx-kernel is the new kernel for all EZX machines Jan 29 23:18:04 Hardware : Compaq iPAQ H3600 Jan 29 23:18:20 hehe.. iPAQ's can be detected easily Jan 29 23:28:37 well, I'll go sleep :) Jan 29 23:29:12 nite then Jan 29 23:29:42 good night Jan 30 00:24:43 anyone still up? Jan 30 00:25:08 barely Jan 30 00:25:26 are you actually at work? Jan 30 00:26:11 anyone know if I can put IMAGE_INSTALL += blah in my local.conf, is this the way to add packages now? Jan 30 00:27:21 not really Jan 30 00:27:28 drinking beer and thinking about dinner Jan 30 00:27:34 what image? Jan 30 00:28:27 lots of them Jan 30 00:28:33 msotly x11-gpe-image though Jan 30 00:28:58 but i'd prefer if I can add stuff to whatever image i'm building at the time Jan 30 00:30:16 you need to look in the image bb file Jan 30 00:30:47 ok, was hoping there was something general Jan 30 00:30:55 ANGSTROM_EXTRA_INSTALL Jan 30 00:31:00 yeah, ANGSTROM_EXTRA_INSTALL Jan 30 00:31:10 yea, thats the only one i've found works at all Jan 30 00:31:22 it works for x11-gpe-image but not for the openmoko images Jan 30 00:31:30 except we really shuold make it IMAGE_EXTRA_INSTALL since images really should not be distro specific ... Jan 30 00:31:50 what is it on the om ones? Jan 30 00:31:58 no idea Jan 30 00:32:06 only been playing with them today and yesterday Jan 30 00:32:21 then there is the adding packages to images from local.conf is harmful camp Jan 30 00:33:16 hmm, i figured i'd put stuff i think should be in an image in the image (like by making a new overriding image) Jan 30 00:33:33 i put stuff thats needed for my platform in the machine def in AMCHINE_EXTRA_RDEPENDS Jan 30 00:33:42 and put stuff that i presonally just want in my local.conf Jan 30 00:33:47 that seemed to make sense **** ENDING LOGGING AT Fri Jan 30 02:59:57 2009