**** BEGIN LOGGING AT Tue Aug 23 02:59:56 2011 Aug 23 07:59:33 hi guys! **** BEGIN LOGGING AT Tue Aug 23 09:06:07 2011 Aug 23 09:35:46 hi all Aug 23 09:36:37 hi bluelightning Aug 23 09:55:09 hi Aug 23 09:57:16 can somebody point me to new patch header where it is defined Aug 23 09:58:15 Noor: Aug 23 09:58:45 thanks PaulePanter .... so nice of you and helping me out Aug 23 10:01:38 Noor: No problem. When I know something, I gladly help. Like others do help me. Aug 23 10:03:52 PaulePanter: right know I am going through the guidelines .... one thing that was in my mind that it should be there in header/commit message (not sure it is there in the guideline or not) which is someone fixs any problem and send the patch he should mentione the scenarion which he fixed in the commit summary Aug 23 10:04:51 PaulePanter: I am seen some of the commits messages but could not figured the scenario that is fixed by that patch Aug 23 10:05:39 not sure it is necessary or not but this will help down the road to reowork the paches when new versions are there Aug 23 10:05:53 *rework Aug 23 10:16:02 ka6sox-zzzz: any word on when the website will be back? Aug 23 10:26:05 bluelightning, which website? Aug 23 10:26:22 ka6sox-zzzz: www.openembedded.org Aug 23 10:26:38 wiki.openembedded.org works fine, but www.openembedded.org gives an error Aug 23 10:27:57 (sorry, I figured someone would have let you know by now...) Aug 23 10:28:36 nope...did not know that. Aug 23 10:28:47 how long has this been going on? Aug 23 10:29:25 ka6sox-zzzz: since some time this weekend AFAIK Aug 23 10:29:37 well, last weekend Aug 23 10:29:50 nope...I didn't hear about this till 3:12am my time now :D Aug 23 10:30:25 are there two separate instances of mediawiki running then? Aug 23 10:31:28 ~dig www.openembedded.org Aug 23 10:31:38 hmm..only works in my channel...jas Aug 23 10:33:14 well..you are correct...there *are* 2 instances running...and www is broken. Aug 23 10:33:19 so lets see why. Aug 23 10:33:42 FWIW, I did edit wiki.openembedded.org; hopefully any edits there can be preserved Aug 23 10:35:47 they are the same DB Aug 23 10:40:17 ah ok, cool Aug 23 10:41:36 okay, I have put in a ticket to the folks who handle that stuff (mediawiki,php and stuff) asking them to fix. Aug 23 10:41:48 and now...I must retire before the sun comes up. Aug 23 11:37:15 morning all Aug 23 11:44:51 hi pb_ Aug 23 12:38:06 PaulePanter: you recommended me to more '"' to the next line and use '\' where I have used '"' .... may I know the reason why we move '"' after the last file in SRC_URI Aug 23 15:14:23 ericben, hi Aug 23 15:38:01 morning kergoth_ Aug 23 15:40:26 * Tartarus waves Aug 23 15:50:22 morning guys Aug 23 16:16:49 morning folk... Aug 23 16:17:08 <<< not PhP fan...(or mediawiki this am) Aug 23 16:27:32 bluelightning, can you please test www.openembedded.org? Aug 23 16:47:45 ka6sox: working, awesome :) Aug 23 16:47:45 thanks! **** BEGIN LOGGING AT Tue Aug 23 16:51:08 2011 Aug 23 16:55:20 hello Aug 23 16:56:55 is anyone here? Aug 23 16:58:14 i need to figure out how to edit a .conf file in caco Aug 23 17:03:40 03Joshua Lock  07master * rfc3bb3a4f0 10bitbake.git/lib/bb/ui/crumbs/hobprefs.py: Aug 23 17:03:40 bb/ui/crumbs/hobprefs: fix setting IMAGE_FSYTPES Aug 23 17:03:40 We were setting the value as a list when a string is expected Aug 23 17:03:40 Signed-off-by: Joshua Lock Aug 23 17:03:40 03Joshua Lock  07master * rc969857c7f 10bitbake.git/lib/bb/ui/hob.py: Aug 23 17:03:40 bb/ui/hob: fix package only build Aug 23 17:03:41 Use the models new get_selected_pn method to return a list of PN's for the Aug 23 17:03:41 selected items and pass that for a package only build, rather than passing Aug 23 17:03:42 a list of package names (which buildTargets can't handle). Aug 23 17:03:42 Fixes [YOCTO #1385] Aug 23 17:03:43 Signed-off-by: Joshua Lock Aug 23 17:03:43 03Joshua Lock  07master * r023edbe6d0 10bitbake.git/lib/bb/ui/hob.py: Aug 23 17:03:44 bb/ui/hob: warn and prevent image build if no IMAGE_FSTYPE is set Aug 23 17:03:44 If IMAGE_FSTYPE is empty and the user has opted to build an image warn Aug 23 17:03:45 them and prevent the build. Aug 23 17:03:45 Fixes [YOCTO #1267] Aug 23 17:03:57 bb/ui/crumbs/tasklistmodel: more robust checking for substrings Aug 23 17:03:57 Relying on the count() method of Str to determine whether the string Aug 23 17:03:58 contains a substring is error prone purely because the return value doesn't Aug 23 17:03:58 strictly convert to a Boolean value. Aug 23 17:03:59 To that end take the more pythonic approach of using the in operator (foo Aug 23 17:03:59 in bar) to check for the substring. Aug 23 17:04:00 03Joshua Lock  07master * r70aebebc38 10bitbake.git/lib/bb/ui/crumbs/tasklistmodel.py: Aug 23 17:04:08 (30 lines omitted) Aug 23 17:15:00 I am using the combo-layer but I am having some ugly conflicts Aug 23 17:40:31 Is someone having trouble with it too? Aug 23 17:41:40 can someone please help me get my clie wl110 wifi card to work on my new c700???? Aug 23 17:49:06 bluelightning: hi :-) I need a hand with comp-layer :-D Aug 23 17:49:20 otavio: sure, what's up? Aug 23 17:49:36 bluelightning: I am having some conflicts that makes no sense for me Aug 23 17:49:42 i think i need a new wifi card Aug 23 17:50:02 bluelightning: and I am wondering how people does the fixes on non-working patches Aug 23 17:50:42 the idea was that you'd keep most things in separate directories within the combo layer Aug 23 17:50:49 so you shouldn't get too many conflicts... Aug 23 17:51:06 bluelightning: I do Aug 23 17:51:36 if you have some conflicts with known cause that are likely to come up again you can use a hook script to modify the patches as they come through Aug 23 17:52:20 bluelightning: http://paste.debian.net/127156/ Aug 23 17:54:15 bluelightning: cleaned up http://paste.debian.net/127157/ Aug 23 17:55:50 otavio: can I have a look at your conf file pls? Aug 23 17:56:10 http://paste.debian.net/127158/ Aug 23 17:57:51 bluelightning: update log http://paste.debian.net/127159/ Aug 23 18:03:03 bluelightning: another nice feature would be to be allowed to update a single component, not all them Aug 23 18:03:36 otavio: yes, that would be useful, I'll add it to my list Aug 23 18:04:24 bluelightning: do you see any problem on how I am using it? Aug 23 18:04:39 not really, it looks OK at a glance Aug 23 18:04:47 I was going to give it a try with your config myself :) Aug 23 18:04:57 bluelightning: did you look at the log? Aug 23 18:05:05 yes, it doesn't tell me a lot though Aug 23 18:05:21 bluelightning: the failing patch is to apply a rm to two files Aug 23 18:05:37 bluelightning: I did check and those two match on the raw repo Aug 23 18:05:47 bluelightning: so no clear reason why it fails Aug 23 18:05:53 hmm Aug 23 18:07:05 heh Aug 23 18:20:36 testing with a replicated version of your setup now Aug 23 18:22:25 bluelightning: cool Aug 23 18:24:29 bluelightning: have you been able to reproduce it? Aug 23 18:24:36 03Philip Balister  07master * r54aaa2d908 10openembedded.git/recipes/glib-2.0/glib-2.0_2.28.5.bb: Aug 23 18:24:36 glib-2.0: explicitly disable dtrace and systemtap for native variant Aug 23 18:24:36 Copied from oe-core : 0d883b5df25635fbad45191d297cbdf78a6c1fe0 Aug 23 18:24:36 Thanks to Joshua Lock for pointing out the fix. Aug 23 18:24:36 Signed-off-by: Philip Balister Aug 23 18:25:26 otavio: yes Aug 23 18:27:19 I wonder if the files in the combo layer are somehow different... Aug 23 18:27:42 bluelightning: they are not Aug 23 18:28:05 bluelightning: I did a diff using the raw repositories by hand and they match Aug 23 18:29:41 bluelightning: I failed to see why they failed to apply Aug 23 18:36:25 is it possible to point to the toolchain in tmp/sysroots/i686-linux (using OE-core) to perform external builds? do I need to build gcc-cross-sdk? Aug 23 18:37:04 generally speaking teh recommendation is to generate an sdk with meta-toolchain, rather than advising poking into the bitbake/oe internals Aug 23 18:41:12 k Aug 23 18:42:57 I don't get why so much needs to be rebuilt between cross and crosssdk, but I'll trust the tools for now. :) Aug 23 18:44:30 heh, ideally, I wouldn't expect the difference to need to exist. if we always packaged our recipes, regardless of whether they're native/cross/etc, it'd simplify things Aug 23 18:45:37 otavio: found the problem... it's the combo-layer hook mangling the inner patch :/ Aug 23 18:46:27 I'm wondering how well the combo layer tool is working for koen in his experiments Aug 23 18:46:30 kergoth_: so, you suspect it is just a matter of needing to do the packaging and I'm pretty much just repeating the same build? :( Aug 23 18:46:34 I'm rather skeptical about it Aug 23 18:46:46 jkridner: probably, but I'm no expert on the sdk classes, could be wrong Aug 23 18:47:01 unfortunately, a fetch of eglibc failed. :( Aug 23 18:47:10 odd Aug 23 18:47:23 I'll bet it is a firewall issue. Aug 23 18:47:39 probably my first subversion fetch. Aug 23 18:49:35 bluelightning: humm Aug 23 18:49:51 bluelightning: so the bug is on the hook script? Aug 23 18:50:00 I think so, yes Aug 23 18:50:36 but I can't see how, it's supposed to only replace the first instance of "Subject: [PATCH]" Aug 23 18:55:27 otavio: ok, I have a fix... just add ^ at the beginning of both regexes in combo-layer-hook-default.sh Aug 23 18:55:30 bluelightning: maybe it has a hash somewhere Aug 23 18:55:33 I will prepare a patch Aug 23 18:55:52 bluelightning: so prepare it and I give it a try Aug 23 18:56:11 ok Aug 23 19:01:43 bluelightning: please send me it by mail; i have a commitment now but once I am back I give it a try Aug 23 19:01:53 otavio: ok Aug 23 19:49:38 hi, the devshell is gone in oe-core and it has been replaced by a task:do_devshell but how do I use itm it spawns a shell but oe_runmake, oe_runconf etc... are absent from it Aug 23 19:58:03 huh? Aug 23 19:58:17 devshell hasn't gone anywhere, and it was a task in oe too Aug 23 19:58:22 it's just implemented differently Aug 23 19:59:01 hmm Aug 23 19:59:27 ok Aug 23 19:59:42 if you want to run those, I'd advise sourcing the run script for the task you care about from ../temp/ Aug 23 19:59:57 ok Aug 23 20:00:06 thanks a lot Aug 23 20:02:24 bluelightning: hi Aug 23 20:02:38 I don't think the oe one had oe_runmake, etc without sourcing it either, actually Aug 23 20:02:45 hmm Aug 23 20:04:48 hi anarsoul Aug 23 20:05:15 bluelightning: you tried linux 3.0 on your hx4700, right? Aug 23 20:11:41 anarsoul: yes I did Aug 23 20:12:30 bluelightning: did you see any issues with tmio_mmc driver? Aug 23 20:34:19 sourcinf the run.do* fails for me Aug 23 20:34:42 before to quick test something: no need of a recipe Aug 23 20:35:03 jusr oe_runconf and oe_runmake Aug 23 20:35:15 now you must modify run.do* Aug 23 20:35:26 which is way harder.... Aug 23 20:39:04 ? Aug 23 20:39:12 the run script is exactly what the shell runs when the task is run Aug 23 20:39:18 if you can't source that, your shell is broken Aug 23 20:43:08 i know....it's just not what i'm looking for Aug 23 20:43:38 shell is not broken....i tried that script on another source Aug 23 20:43:49 for quick turnarround times Aug 23 20:59:41 bluelightning: worked fine :-D Aug 23 21:00:19 bluelightning: one thing I'd like to see improved is it to try to detect when you are appling an already included patch Aug 23 21:56:24 otavio: why would you be doing that? **** ENDING LOGGING AT Wed Aug 24 02:59:57 2011