**** BEGIN LOGGING AT Tue Oct 26 03:00:11 2010 Oct 26 06:31:45 gm Oct 26 06:35:30 obi which one got lost? (the lua one got pushed by khem maybe he already set it to archived before pushing it) Oct 26 06:36:02 btw I did not touch your patches to the bbclass files, they are potentially more dangerous and they are in an area I'm not too familiar with Oct 26 07:28:18 good morning Oct 26 07:40:54 hi mckoan Oct 26 07:41:09 what package provides ldd ? Oct 26 07:42:04 glibc/eglibc Oct 26 07:42:33 khem, thanks, didn't find it in sysroot Oct 26 07:42:49 so guessed I missed it somewhere Oct 26 07:43:48 eFfeM_work: actually I am unable to find it in my eglibc based rootfs too Oct 26 07:44:32 hm, strange Oct 26 07:44:54 i didn't find it in my rootfs nor in sysroots will peek in the eglibc dir Oct 26 07:47:11 mckoan: it is a separate package Oct 26 07:48:00 that's why it was not in my rfs either Oct 26 07:49:03 and there is no host version, thats why it is not in sysroots Oct 26 07:49:20 * eFfeM_work thinks it could be useful to have ldd on host too, to examine deps Oct 26 08:06:20 eFfeM_work: heh .. is one of the first packages to install Oct 26 08:06:39 ant_work: yeah, added it to my image Oct 26 08:07:25 eFfeM_work: so what recipe is required? Oct 26 08:07:54 I'm still working on pre-boot environments so I'm exhempted atm but I'm pretty sure there is something wrong with ts calibration wrt linkages Oct 26 08:07:56 mckoan: it is in the eglibc recipe, but it is stored in a separate package, add ldd to your image Oct 26 08:09:52 eFfeM_work: second package to install if mc you have to do serious editing-work on target Oct 26 08:10:05 *is mc if Oct 26 08:10:57 it even works whit brken keymaps, where vi has no chanche ;) Oct 26 08:12:37 btw you guys are all preparing for OEDEM? Seems a bit empty those days... Oct 26 08:14:56 i'm still at work, will fly to uk tonight (only 1 hr flight for me) Oct 26 08:15:56 eFfeM_work: clear, thx Oct 26 08:17:07 hm there is no sys/io.h in eglibc ???? Oct 26 08:18:32 ah nevermind error between chair and kbd Oct 26 08:53:14 hi florian Oct 26 08:53:19 are you already in the UK? Oct 26 08:54:07 good morning Oct 26 08:54:47 eFfeM_work: no not yet, i'll arrive wednesday morning - unluckily i didn't manage to get a ticket for elc Oct 26 08:54:58 oh, pity Oct 26 08:55:41 i'm flying this evening,fortunately i was in time to get a ticket (generally I tend to be a late decision maker) Oct 26 09:03:03 heh..last week first ICE went to London Oct 26 09:03:36 (was pulled in the Channel by an 'authorized' train though) Oct 26 09:04:06 florian: maybe in a couple of years Oct 26 09:04:50 heh right Oct 26 09:11:45 ant_work: currently it takes about eight hours by train from here and they do not tell you about the cost Oct 26 09:11:51 hi lrg Oct 26 09:14:47 03Martin Jansa  07master * r7071906aad 10openembedded.git/recipes/estardict/ (5 files in 2 dirs): Oct 26 09:14:47 estardict: add patch for efl api change and remove older versions Oct 26 09:14:47 Signed-off-by: Martin Jansa Oct 26 10:04:04 hi obi, typed this earlier today but guess you were not in Oct 26 10:04:05 (08:29:38 AM) eFfeM_work: obi which one got lost? (the lua one got pushed by khem maybe he already set it to archived before pushing it) Oct 26 10:04:05 (08:30:10 AM) eFfeM_work: btw I did not touch your patches to the bbclass files, they are potentially more dangerous and they are in an area I'm not too familiar with Oct 26 10:05:50 btw anyone an idea about this: e2fsprogs is doing a bin/sh /usr/bin/texi2dvi ./libext2fs.texinfo which starts etex txiversion.tex whichis aready running for 300+ mins Oct 26 10:09:05 JaMa: thanks for looking at introspection; seems this created quite a mess Oct 26 10:13:00 hi eFfeM_work Oct 26 10:14:03 eFfeM_work: http://patchwork.openembedded.org/patch/3418/ Oct 26 10:24:52 obi: I applied that, thougth I pushed it too Oct 26 10:25:39 hm it is not in the log Oct 26 10:27:33 will repush later Oct 26 10:27:46 ok :) Oct 26 10:29:32 currently busy with a build Oct 26 10:29:53 obi: have you already sent your key? Oct 26 10:31:03 not yet, in a minute Oct 26 10:39:27 btw., what's the difference between master and org.openembedded.dev? do you have to keep them in sync manually? Oct 26 11:16:56 obi, no they are autosynced, we used to have org.openembedded.dev and a while ago master was attached to it because it is more common (and preferred by some) Oct 26 11:39:13 03Martin Jansa  07master * r92b407d9e1 10openembedded.git/recipes/ (3 files in 2 dirs): Oct 26 11:39:13 mplayer,gst-plugins-ugly: bump PR to rebuild after libdvdread3 -> libdvdread4 upgrade Oct 26 11:39:13 Signed-off-by: Martin Jansa Oct 26 11:39:30 03Martin Jansa  07master * r3cd16a427e 10openembedded.git/recipes/libgee/libgee_0.6.0.bb: Oct 26 11:39:30 libgee_0.6.0: add gobject-introspection-native also to DEPENDS_virtclass-native, because DEPENDS handling in native(sdk).bbclass is broken now Oct 26 11:39:30 Signed-off-by: Martin Jansa Oct 26 11:39:30 03Martin Jansa  07master * r4ab99f2b10 10openembedded.git/recipes/xorg-lib/ (14 files in 3 dirs): Oct 26 11:39:30 pixman: move from 0.19.4 to 0.19.6 Oct 26 11:39:30 Signed-off-by: Martin Jansa Oct 26 11:40:08 hi, I've a device with a broken touchscreen which was linear and isn't anymore Oct 26 11:40:27 so how do I use ts_harvest Oct 26 12:15:13 obi: there's your patch Oct 26 12:15:41 if cia wants to deliver Oct 26 12:17:06 thanks :) Oct 26 12:17:40 http://cgit.openembedded.org/cgit.cgi/openembedded/commit/?id=55fb0f174f63760b0483099aab9deddadd8ace9c Oct 26 12:17:52 ~curse cia Oct 26 12:17:52 May the fleas of a thousand camels infest your most sensitive regions, cia ! Oct 26 12:28:12 CIA is not interested in giving my patches Oct 26 13:11:06 heyho Oct 26 13:34:20 03Frans Meulenbroeks  07org.openembedded.dev * rb678eb611e 10openembedded.git/recipes/hdparm/ (4 files in 3 dirs): Oct 26 13:34:20 hdparm: removed legacy versions 6.3 and 6.6 Oct 26 13:34:20 we still have 6.9 and 9.35 Oct 26 13:34:20 Signed-off-by: Frans Meulenbroeks Oct 26 13:34:55 03Frans Meulenbroeks  07org.openembedded.dev * r4d2cb79dce 10openembedded.git/recipes/hdparm/hdparm_9.15.bb: Oct 26 13:34:55 hdparm: updated to 9.35 Oct 26 13:34:55 also added a separate package for the wiper script (previously unpackaged) Oct 26 13:34:55 rationale for making it an additional package is because the script Oct 26 13:34:55 depends on gawk and bash etc. Oct 26 13:34:55 Signed-off-by: Frans Meulenbroeks Oct 26 13:34:56 03Andreas Oberritter  07org.openembedded.dev * r55fb0f174f 10openembedded.git/classes/oestats-client.bbclass: Oct 26 13:35:23 ok, pushed that more thanan hour ago Oct 26 13:35:36 oestats-client.bbclass: Fix "AttributeError: 'NoneType' object has no attribute 'startswith'" if OESTATS_SERVER is not set Oct 26 13:35:36 Signed-off-by: Andreas Oberritter Oct 26 13:35:36 Acked-by: Chris Larson Oct 26 13:35:36 Signed-off-by: Frans Meulenbroeks Oct 26 13:35:36 03Martin Jansa  07master * r27e60dd004 10openembedded.git/recipes/xorg-lib/ (2 files in 2 dirs): Oct 26 13:35:36 pixman_0.19.6: add missing patch Oct 26 13:35:37 * sorry I've tested it only on armv4t before (where it doesn't fail, because of disabled neon) Oct 26 13:35:37 Signed-off-by: Martin Jansa Oct 26 14:42:53 hello Oct 26 14:43:53 I'm trying to build wt3 for beagleboard. Is it possible to do so without X11? Oct 26 15:33:18 hi , is alsamixer having microphone control inside ? i vent seen any analog microphone controls there , does someone know any solution ? Oct 26 16:31:01 ok stupid question - can't seem to figure out how to get the oe-testing branch from git - its not documented on http://wiki.openembedded.org/index.php/Testing as its likely assumed there people know how to use git well Oct 26 16:31:21 git checkout remotes/origin/testing-next is what I thought I should do Oct 26 16:31:32 03Khem Raj  07master * rf0a0500df0 10openembedded.git/recipes/gcc/ (15 files in 2 dirs): Oct 26 16:31:32 gcc-4.5: Import recent linaro patches Oct 26 16:31:32 * These are selected patches that are applied on latest linaro Oct 26 16:31:32 4.5 gcc Oct 26 16:31:32 Signed-off-by: Khem Raj Oct 26 16:31:42 03Khem Raj  07master * r7032f48ad0 10openembedded.git/recipes/uclibc/ (uclibc-git/isnan.patch uclibc_git.bb): Oct 26 16:31:42 uclibc_git.bb: Bump to latest git rev Oct 26 16:31:42 Signed-off-by: Khem Raj Oct 26 16:32:30 khem: ping Oct 26 16:32:54 tharvey: git checkout -b testing-next origin/testing-next, assuming you want testing-next rather than testing Oct 26 16:33:11 tharvey: with recent git, the -b bit isn't necesary, just git checkout testing-next, it'll automatically go based upon the upstream one Oct 26 16:33:29 -b creates the local branch based on the remote tracking branch Oct 26 16:33:44 ah... 'based on' remote - ok thx Oct 26 16:34:06 I assume testing-next is what I want - I want to run a test build on what cliff announced he merged the other day for testing Oct 26 16:34:35 ah, I'm reading the testing page wrong - I need to use 'testing' right? Oct 26 16:34:52 'testing' is more 'tested', it's what was testing-next last time, it gets moved after people do their build testing with testing-next Oct 26 16:35:01 khem, what's in the linaro patches? Oct 26 16:35:04 so it depend son whether you want a stable baseline, or want to contribute to the testing of the next one Oct 26 16:35:22 no, I want to contribute to the testing of the next one - so testing-next Oct 26 16:35:26 yep Oct 26 16:35:32 seems like testing should be called 'tested' - think that was discussed last week Oct 26 16:35:36 confusing Oct 26 16:35:38 yeah, it was Oct 26 16:35:41 on the liast Oct 26 16:35:42 list Oct 26 16:35:58 i expect its called testing because thats what, say, debian calls their testing distribution Oct 26 16:36:00 but it is confusing for us Oct 26 16:39:33 thx Oct 26 16:41:51 np Oct 26 16:43:57 03Khem Raj  07master * r2c93b831a7 10openembedded.git/recipes/aircrack/ (aircrack_2.1.bb aircrack_2.41.bb): Oct 26 16:43:57 aircrack_2.1.bb,aircrack_2.41.bb: Update HOMEPAGE Oct 26 16:43:57 Signed-off-by: Khem Raj Oct 26 16:55:49 anyone doing much with bootspeed tuning here? I'm trying to get fast boot from my OMAP3 target. I have a decent handle on optimizing x-load, u-boot, and kernel but not userspace - seems like sysvinit and udev are slow but commonly used in the images in oe Oct 26 16:56:17 I was surprised to find that upstart had very outdated recipes in oe so obviously people aren't using it? Oct 26 16:56:51 khem: what about bumping gcc-4.5 SRCREV to 164911 as linaro does? to stay as close as possible to them? currently I'm bisecting which linaro patch is causing that broken libcairo and armv7 so maybe it's already fixed in FSF upstream or one of those new patches you've added today Oct 26 18:26:10 Anyone run into undefined reference to __stack_chk_guard when building eglibc? Oct 26 18:26:16 khem: perhaps you've seen this? Oct 26 18:27:46 wait, nevermind, think i got it Oct 26 18:40:28 JaMa|Off: yes one by one the update is next in queue I did not want to change so much at one time Oct 26 18:44:57 kergoth: ssp issue seems Oct 26 18:48:54 hmm, so it seems, no idea what though, odd Oct 26 18:50:18 did you get over it ? Oct 26 18:52:19 hey guys. i have a difficult problem :(. I'm using a gumstix overo board and our USB Device won't turn on if it is plugged in BEFORE we boot up the system. Oct 26 18:52:30 if we plug it in after the system is up and running, no problem. works fine. Oct 26 18:52:48 We have to "unplug/replug" the cable after bootup for it to initialize. Oct 26 18:52:58 lsusb won't list it either Oct 26 18:53:22 does anyone have any suggestions on where to look? I feel like it is a kernel issue, but I'm really not sure Oct 26 18:54:46 khem: verifying it wasnt the autotools branch causing it first, heh Oct 26 19:08:01 kergoth: what freebsd versions we will test and support? Oct 26 19:08:14 I vote for 8.1 and later :) Oct 26 19:12:26 Jay7: whatever you work on latest is always ok Oct 26 19:12:39 I mean latest stable release which is 8.1 Oct 26 19:13:15 I'll setup 8.1/i386 to kvm then Oct 26 19:13:26 cool Oct 26 19:17:30 Jay7: no idea, i don't have any bsd installs, the last bsd i touched was an openbsd in 2000 or so.. Oct 26 19:17:33 heh Oct 26 19:17:41 well, other than darwin, of course Oct 26 19:17:46 * khem deals with freebsd on daily bases Oct 26 19:19:36 03Khem Raj  07master * r699938902c 10openembedded.git/recipes/gcc/ (gcc-4.5.inc gcc-4.5/linaro/gcc-4.5-linaro-r99362.patch): Oct 26 19:19:36 gcc-4.5.inc: Update to latest SVN revision on 4.5 branch Oct 26 19:19:36 Signed-off-by: Khem Raj Oct 26 19:19:59 03Chris Larson  07master * rfe76d10990 10openembedded.git/conf/bitbake.conf: Oct 26 19:19:59 bitbake.conf: add SAVANNAH_MIRROR Oct 26 19:19:59 Signed-off-by: Chris Larson Oct 26 19:20:01 03Chris Larson  07master * r77c46353c6 10openembedded.git/ (classes/base.bbclass lib/oe/unpack.py): Oct 26 19:20:01 Handle .lz/.tar.lz URIs Oct 26 19:20:02 Signed-off-by: Chris Larson Oct 26 19:20:04 03Chris Larson  07master * re2a741b3d2 10openembedded.git/recipes/lzip/lzip_1.11.bb: Oct 26 19:20:05 lzip: add recipe, version 1.11 Oct 26 19:20:05 Signed-off-by: Chris Larson Oct 26 19:20:06 03Chris Larson  07master * r46cb2fc362 10openembedded.git/recipes/ed/ (ed.inc ed_0.5.bb ed_1.5.bb): Oct 26 19:20:07 ed: add recipes, version 0.5 and 1.5 Oct 26 19:20:08 0.5 was the last version to be GPLv2+, so I'm including that in addition to Oct 26 19:20:09 the latest GPLv3+ version. Oct 26 19:20:12 Signed-off-by: Chris Larson Oct 26 19:21:17 kergoth: I think we should discuss the machine layering model more the one that koen proposed Oct 26 19:22:28 I agree. bbappend makes it easier to do such structuring Oct 26 19:23:02 hi guys... my bitbake does not do anything useful, it just stalls...saying: NOTE: Cache is clean, not saving. Oct 26 19:23:02 NOTE: Resolving any missing task queue dependencies Oct 26 19:23:09 /scratch/oe/work/armv7a-oe-linux-gnueabi/scim-1.4.9-r1/temp/run.do_compile.20737 hangs forever Oct 26 19:23:12 hmmm Oct 26 19:23:42 Phlogi: what version? Oct 26 19:23:47 kergoth: git Oct 26 19:24:34 BitBake Build Tool Core version 1.11.0, bitbake version 1.11.0 Oct 26 19:24:58 never seen that, dunno. try cranking up the debugging Oct 26 19:24:59 -DDD Oct 26 19:25:27 Phlogi: what is your bitbake command that u are using Oct 26 19:25:54 http://pastebin.ca/1973913 khem Oct 26 19:25:58 good point, i forgot about the hang when you try to bake something that's in ASSUME_PROVIDED Oct 26 19:26:14 Phlogi: hah world Oct 26 19:26:18 good luck Oct 26 19:26:23 it'll take a long time before it even starts Oct 26 19:26:26 it isn't hung, just working Oct 26 19:26:28 khem: its a micro distribution... Oct 26 19:26:35 Phlogi: distro doesn't control whats buitl. Oct 26 19:26:35 Phlogi: how muscular is your system Oct 26 19:26:40 kergoth: no... no cpu, no network, no hdd activity Oct 26 19:26:41 only particular policy things Oct 26 19:26:58 Phlogi: that build will take a week or so, depending on your machine. Oct 26 19:27:05 kergoth: why? Oct 26 19:27:07 you're asking it to build every recipe in the oe repository. Oct 26 19:27:11 which is, what, 7,000? Oct 26 19:27:16 phlogi for bitbake world on my system it takes about 30 min to get thru the runqueue Oct 26 19:27:17 kergoth: really? :D Oct 26 19:27:19 not all, but one of each provider, of each version Oct 26 19:27:20 yes Oct 26 19:27:22 thats what world is Oct 26 19:27:28 oh oops Oct 26 19:27:38 I thought that are all packages in the distrubution, so like a minimal system Oct 26 19:27:39 * khem has bitbake world going for last 3 days Oct 26 19:27:42 non stop Oct 26 19:27:44 how can I do my micro distri then? Oct 26 19:27:51 it depends on what you want Oct 26 19:27:56 disto, again, just controls policy Oct 26 19:27:58 not what you're building Oct 26 19:28:03 mickey|zzZZzz: ping Oct 26 19:28:05 hmm I see Oct 26 19:28:06 see recipes/images/, recipes/meta/, and recipes/tasks/ Oct 26 19:28:23 micro-image would be a good choice, perhaps Oct 26 19:28:24 kergoth: so I have to build everything on my own? ok using taks or meta packages/recipies I think I get it :) Oct 26 19:28:27 but again, deepnds on what you want to build Oct 26 19:28:40 what I want first is just a minimal linux with bash terminal to see if I can really get it working :D Oct 26 19:28:44 what would you build for this? Oct 26 19:28:51 OE is structured this way intentionally, with orthogonnal pieces Oct 26 19:29:12 machine controls the device being targeted, distro controls the policies, and neither fully control what you build Oct 26 19:29:23 like i said, micro-image would likely be a good choice Oct 26 19:29:28 its basically busybox in a filesystem Oct 26 19:29:39 there's console-image too, but it ends up building X and crap due to deps from bluez Oct 26 19:30:17 kergoth: ok that sucks... don't need bluetooth :D Oct 26 19:30:42 look around images and tasks, there are plenty for various needs Oct 26 19:31:18 Running task 55963 of 79600 Oct 26 19:31:21 images produce filesystem images suitale for flashing to a device, or tarballs suitable for extraction as an nfsroot Oct 26 19:31:35 kergoth: how do I build this? just the name is not enough, why? Doesn't it search for the bb files? Oct 26 19:31:39 tasks are just recipes that produce ipks that depend on things, pulls those things into the build, similar to debian's tasks Oct 26 19:31:42 Phlogi: ? Oct 26 19:31:44 the name is just fine. Oct 26 19:31:46 bitbake micro-image Oct 26 19:31:57 ERROR: Nothing PROVIDES 'micro-image' Oct 26 19:32:07 then your OE is ancient, or your setup is wrong Oct 26 19:32:17 oh Oct 26 19:32:21 again, look in openembedded/recipes/images/ Oct 26 19:32:22 BBFILES = is wrong Oct 26 19:32:24 see what images exist Oct 26 19:32:26 thatd do it Oct 26 19:32:59 bbfiles is how it finds recipes, without recipes its hard to build ;) Oct 26 19:33:07 jup I know Oct 26 19:33:14 I thought I've set this, sorry! Oct 26 19:33:25 now it does some stuff finally :) Oct 26 19:33:34 thank you guys for this starting help Oct 26 19:33:50 it will build the needed cross toolchain first right? Oct 26 19:33:54 yep Oct 26 19:33:59 neat Oct 26 19:34:07 03Chris Larson  07master * r0450e32748 10openembedded.git/MAINTAINERS: Oct 26 19:34:07 MAINTAINERS: add myself for lzip, ed Oct 26 19:34:07 Signed-off-by: Chris Larson Oct 26 19:34:26 recipes state what they depend on, for the most part -- the crosscopmiler is special, it gets added to recipe deps automatically unless they opt out Oct 26 19:34:32 bitbake obeys those Oct 26 19:34:35 oki Oct 26 19:35:04 khem: i'm gonna add the freshmeat release rss feeds for the shit i maintain, see if that helps keep an eye on things :) Oct 26 19:35:14 s/for/to my google reader for/ Oct 26 19:36:37 kergoth: yeah thats a nice think imp Oct 26 19:36:39 imo Oct 26 19:36:48 we have too much unmanagable stuff Oct 26 19:37:11 and I am going to propose to drop old gcc and binutils Oct 26 19:37:20 nice Oct 26 19:37:25 I will send an email later today and ask for which ones to keep Oct 26 19:37:39 if I dont get replies then I will keep what I like Oct 26 19:42:01 kergoth: that override patch I proposed does not work Oct 26 19:43:09 kergoth: the problem is that if I do a setVar of OVERRIDES somehow it evaluates some vars like prefix prematurely and they dont get the override they should get from say native.bbclass Oct 26 19:43:51 and hence the CONFIGUREOPTS get values of --prefix --bindir etc. pointing to /usr even for native recipes Oct 26 19:44:39 you're doing getVar with True as the second argumnet Oct 26 19:44:40 I saw that if I read OVERRIDES using getVar its ok it only happens when I update it Oct 26 19:44:43 which is expanding it right then Oct 26 19:44:48 no I changed those Oct 26 19:44:55 to False that was my first fix Oct 26 19:44:58 wwell, dunno what to tell you then, no idea Oct 26 19:45:01 still did not work Oct 26 19:45:14 would need it in front of me, i'll try doing it locally Oct 26 19:45:27 I was not able to understand why half of vars will get pre evaluated Oct 26 19:45:40 like --execdir gets the right value but not --prefix Oct 26 19:45:43 its really weird Oct 26 19:45:54 weird Oct 26 19:45:58 yeah Oct 26 19:46:15 I was trying jamvm-native_x.x.x.bb recipe to test btw. Oct 26 19:46:31 it inherits native and does not use BBCLASSEXTEND Oct 26 19:47:04 I moved it to base.bbclass into anonymous python but no use Oct 26 19:47:46 Does data smart evaluate stuff before updating ? Oct 26 19:48:05 I thought it just did a blind copy Oct 26 19:48:13 meaning string to string Oct 26 19:48:25 things only get expanded when used, at getVar() time, it doesn't happen at finalize time when overrides are collapsed Oct 26 19:48:29 afaik, anyway Oct 26 19:48:49 of course, OVERRIDES will get expanded at finalize time, since its used there Oct 26 19:49:25 hmmm but then it should have already inherited native by then and those values should be used Oct 26 19:49:31 and not the ones from bitbake.conf Oct 26 19:51:13 kergoth: on incremental builds bitbake knows if a recipe was built last time right thru stamps ? Oct 26 19:52:14 It would be nice thing to add a keyword which told bitbake to ignore incremental build of a given recipe i.e. only build one time Oct 26 19:52:18 unless cleaned Oct 26 19:52:47 I would add that keyword to -initial and -intermediate recipes in toolchain Oct 26 19:53:11 and let bitbake only build gcc-cross when I have a fully built and staged tmpdir Oct 26 19:54:11 SIGFOOD bbiab Oct 26 20:11:04 jo Oct 26 20:11:58 jo hrw Oct 26 20:23:12 03Khem Raj  07master * r78ad338687 10openembedded.git/recipes/wrt/loader_0.04.bb: Oct 26 20:23:12 loader_0.04.bb: Its a mips 32-bit application build it for mips 32bit platforms only. Oct 26 20:23:12 Signed-off-by: Khem Raj Oct 26 20:36:09 can anybody try to build scim ? Oct 26 20:43:30 I just moved from 32 to 64 bit kubuntu host, copied over my whole oe tree and noticed that sysroots/x86_64-linux is now in my PATH. I could probably symlink my way out of this, or rebuild all. Is there a better way (rebuilding just native stuff)? Oct 26 20:44:45 khem mom Oct 26 20:49:09 khem did you update uclibc? Oct 26 20:52:02 okay Oct 26 21:07:00 woglinde: yes I did Oct 26 21:07:35 hoj: please checkout stuff fresh Oct 26 21:08:10 khem: why did you drop gcc-4.5-linaro-r99362.patch ? the faulty patch (from cairo pov) is somewhere between 99347 and 99369 so it could be it :) Oct 26 21:08:41 JaMa|Off: no thats not the one Oct 26 21:08:48 its a test fix Oct 26 21:08:58 which got pulled into 4.5 branch Oct 26 21:09:08 so had to be dropped Oct 26 21:09:32 JaMa|Off: hmm so you are bisecting it thats cool Oct 26 21:09:42 *g* Oct 26 21:09:51 thats sick for gcc Oct 26 21:10:31 JaMa|Off: btw. I have also reverted 99384 Oct 26 21:10:39 good that cairo is testcase, not ie qt :) Oct 26 21:10:40 with a different patch Oct 26 21:11:14 JaMa|Off: give it a try with latest gcc-cross may be it fixed the issue Oct 26 21:11:46 khem: yup I started build with those new patches when I pinged you about SRCREV.. Oct 26 21:12:00 now I'll rebuild again with newer SRCREV too Oct 26 21:12:43 JaMa|Off: let me see what patches could be suspects between 347 and 369 Oct 26 21:14:36 JaMa|Off: 99348 looks a candidate try to remove that one Oct 26 21:16:32 JaMa: also 99353 Oct 26 21:18:26 now tested old SRCREV + all patches to 99415 -> broken Oct 26 21:18:46 JaMa: ok now remove the patches selectively Oct 26 21:18:58 as I suggested Oct 26 21:19:22 running OE head build (new SRCREV) on shr buildhost and -99348 -99353 on local Oct 26 21:21:14 btw I'm storing gcc-cross staging-pkg and whole workdir from cairo, anything else which could be usefull later to compare? Oct 26 21:21:45 thats enough Oct 26 21:22:23 when we figure out which patch it is in gcc second thing would be to find what this changed in cairo Oct 26 21:22:39 right fix would be to understand the cairo code and go from there Oct 26 21:22:49 as it could be a gcc bug or cairo bug as well Oct 26 21:22:58 but first lets pin the gcc patch Oct 26 21:23:17 99357 is also a suspect Oct 26 21:24:00 99363 Oct 26 21:25:40 99368 and 99369 Oct 26 21:28:13 khem: thanks. I'll rm -rf tmp and rebuild Oct 26 21:31:36 he ant Oct 26 21:32:34 khem scim -> found. Your intltool is too old. You need intltool 0.33 or later. Oct 26 21:33:01 hello woglinde Oct 26 21:37:02 woglinde: thats a better error then my infinite look in make Oct 26 21:37:07 s/look/loop/ Oct 26 21:37:22 okay Oct 26 21:37:26 have to go Oct 26 21:52:24 khem: 99369-99348-99353 still broken trying without those other 4 Oct 26 21:52:47 hmmm 48 was my guess Oct 26 22:26:02 khem: 99369-99348-99353-99357-99363-99368-99369 still broken, any other guess? Oct 26 22:29:55 JaMa: really hmmm Oct 26 22:30:01 JaMa: how about 99347 Oct 26 22:31:04 and 99349 Oct 26 22:31:25 ok will try.. but results tomorrow.. I'll start build and go to sleep Oct 26 22:31:31 alright Oct 26 22:31:33 np Oct 26 22:31:34 thx Oct 26 22:31:36 and gn Oct 26 22:35:51 03Andrea Adami  07org.openembedded.dev * r6e93a515f1 10openembedded.git/recipes/linux/ (11 files in 2 dirs): linux-rp: readd boot logo to zaurus machines (for 2.6.24 + 2.6.26) Oct 26 22:35:58 03Andrea Adami  07org.openembedded.dev * r82cf0abfd5 10openembedded.git/recipes/linux/linux-kexecboot/ (4 files in 4 dirs): Oct 26 22:35:58 linux-kexecboot: move logos in recipes/linux/files Oct 26 22:35:58 * custom-sized distro logos can be shared including linux.inc Oct 26 22:35:58 * and e.g. adding SRC_URI += "file://${LOGO_SIZE}/logo_linux_clut224.ppm.bz2" Oct 26 22:35:59 * to the recipe. Oct 26 22:35:59 03Andrea Adami  07org.openembedded.dev * rbfe6bce386 10openembedded.git/recipes/linux/linux-2.6.30/tosa/defconfig: linux-2.6.30: readd boot logo for zaurus tosa Oct 26 22:36:00 03Andrea Adami  07org.openembedded.dev * r4e2b3a6355 10openembedded.git/recipes/linux/linux_2.6.32.bb: linux-2.6.32: remove obsoleted zaurus customizations Oct 26 22:36:01 03Andrea Adami  07org.openembedded.dev * rf000274e8b 10openembedded.git/recipes/linux/linux-2.6.34/ (4 files in 3 dirs): linux-2.6.34: remove obsoleted zaurus defconfigs Oct 26 22:36:02 03Andrea Adami  07org.openembedded.dev * r57ade72912 10openembedded.git/recipes/linux/linux-2.6.29/tosa/defconfig: linux-2.6.29: readd boot logo for zaurus tosa Oct 26 22:36:04 03Andrea Adami  07org.openembedded.dev * r44cf2c404d 10openembedded.git/recipes/linux/linux-2.6.33/spitz/defconfig: linux-2.6.33: readd boot logo to spitz defconfig Oct 26 22:36:25 03Andrea Adami  07org.openembedded.dev * rb82cd64b60 10openembedded.git/recipes/linux/linux-2.6.33/spitz/defconfig: linux-2.6.33: remove obsoleted zaurus spitz defconfig Oct 26 22:36:26 03Andrea Adami  07org.openembedded.dev * r91d7fb6c05 10openembedded.git/recipes/linux/linux-2.6.28+2.6.29-rc6/tosa/ (2 files): linux-2.6.28+2.6.29-rc6: remove obsoleted tosa defconfig. Oct 26 22:36:27 03Andrea Adami  07org.openembedded.dev * r2974cb5763 10openembedded.git/recipes/linux/ (28 files in 3 dirs): linux-2.6.28: remove obsoleted zaurus collie and tosa patchsets Oct 26 22:39:49 03Chris Larson  07master * r203bc975c0 10openembedded.git/ (4 files in 2 dirs): Oct 26 22:39:49 genromfs: add recipes, version 0.5.2 and git Oct 26 22:39:49 Also mark myself as the maintainer of this. Oct 26 22:39:49 Signed-off-by: Chris Larson Oct 26 22:47:07 03Andrea Adami  07org.openembedded.dev * rf1335197b0 10openembedded.git/recipes/linux/ (4 files in 4 dirs): linux-kexecboot-2.6.35: remove temporary dev version Oct 26 22:47:18 03Andrea Adami  07org.openembedded.dev * rb4eb083f5e 10openembedded.git/recipes/linux/linux-kexecboot-2.6.34/ (9 files in 7 dirs): linux-kexecboot-2.6.35: remove temporary dev version Oct 26 22:57:19 03Andrea Adami  07org.openembedded.dev * r428fd9c5aa 10openembedded.git/recipes/linux/ (10 files in 8 dirs): linux-kexecboot-2.6.34: remove temporary dev version Oct 26 22:57:22 03Andrea Adami  07org.openembedded.dev * r03cba6d5ae 10openembedded.git/recipes/linux/linux-kexecboot-2.6.34/ (9 files in 7 dirs): Oct 26 22:57:22 Revert "linux-kexecboot-2.6.35: remove temporary dev version" Oct 26 22:57:22 This reverts commit b4eb083f5efe0d1c24dee527c545d8e82a6c65b7. Oct 26 22:57:22 * misnamed Oct 26 22:57:29 03Andrea Adami  07org.openembedded.dev * r9b6e9ed52d 10openembedded.git/recipes/linux/ (5 files in 4 dirs): linux-kexecboot-2.6.32: remove obsoleted zaurus patchsets and defconfigs Oct 26 23:07:52 03Andrea Adami  07org.openembedded.dev * r0a31ce3009 10openembedded.git/recipes/linux/linux-kexecboot/fix.module.loading.16310.patch: linux-kexecboot: remove obsolete 2.6.35 specific patch. Oct 26 23:12:32 hmm unity UI is default desktop for next version of ubuntu codenames natty narwhal Oct 26 23:13:27 Hi Oct 26 23:13:57 I'm now in a Arch Linux operational system. Oct 26 23:14:16 i get a error when the OE compiles FLEX on A1200 arch with Angstrom distro: Oct 26 23:14:26 http://pastebin.com/Zgx5RCGT Oct 26 23:18:28 any help? Oct 26 23:21:26 03Andrea Adami  07org.openembedded.dev * r0e49c49428 10openembedded.git/recipes/linux/linux/fix.module.loading.16310.patch: linux-2.6.35: move version specific patch in /linux-2.6.35 Oct 26 23:23:39 angelox_123: seems somehow it does not like autotools Oct 26 23:23:52 angelox_123: is maintainer-mode selected during configure Oct 26 23:24:11 (Hi khem :) ) Oct 26 23:24:15 khem: strange Oct 26 23:26:25 khem: i've tried some times clean package,but no effect Oct 26 23:27:14 khem: Do you know anyhow to fix Oct 26 23:37:00 angelox_123: nothing on top of my head Oct 26 23:37:15 khem: Ok :) Oct 26 23:37:16 I dont see this issue locally so its hard to come up with some analysis Oct 26 23:37:32 flex_****.bb is small Oct 26 23:39:49 good night Oct 26 23:45:52 my friend are asking if he can build packages for him HOST PC with openembedded Oct 26 23:46:34 " i'm saying by me because my friend doesn't speak english " Oct 26 23:47:55 angelox_123: yes he can but he better use the build system thats used for host distro Oct 26 23:47:58 and not OE Oct 26 23:48:10 if he has OE installed as distro on host then its a different story Oct 26 23:49:31 is bitbake 1.10.1 stable, or should I use 1.8.18? Oct 26 23:49:58 T0mW: 1.10.1 is good Oct 26 23:50:05 thanks Oct 26 23:50:14 T0mW: I would recommend git master though Oct 26 23:50:37 khem: i didn't understand...he can use oe to build packages(e.g. nano) to him pc? Oct 26 23:50:46 ? I've git installed, what is "git master"? Oct 26 23:51:37 heh, last time I used bitbake was with svn Oct 26 23:51:38 T0mW: master of bitbake Oct 26 23:53:19 oh, meant that I was using monotone the last time for the database, seems things have changed since. Oct 26 23:54:16 T0mW: yes quite a bit Oct 26 23:55:51 * T0mW sucks in git.openembedded.org Oct 26 23:56:38 T0mW: what makes you work again on OE Oct 26 23:57:16 khem: new project, have to do some research before I submit a proposal to the customer (s3c2440 based board). Oct 26 23:58:14 khem: time critical delivery, want to get situated again before committing to a hard delivery date. Oct 26 23:59:11 OOO! more kernel sources now (recipies/linux) Oct 26 23:59:32 last time I looked there was only about a dozen Oct 27 00:08:13 Hmmm, looks like I'll have to update my kernel recipe to EABI from OABI Oct 27 00:40:23 one thing that has not changed, bitbake's biggest weakness is the URLs in the recipes. Oct 27 01:14:00 T0mW: why is that a weakness Oct 27 01:14:26 khem, site goes offline, then ... Oct 27 01:14:42 T0mW: we have mirrors now Oct 27 01:14:59 angstrom mirror and a source mirror of openembedded Oct 27 01:15:18 heh, "bitbake world" then keep the tarballs around Oct 27 01:15:25 so this should minimize such issues Oct 27 01:15:40 bitbake -c fetchall Oct 27 01:15:51 will fetch all the tars you need for a given image Oct 27 01:15:52 ftp.ayamura.org is missing, ftp timeouts Oct 27 01:16:00 khem, thanks Oct 27 01:16:34 T0mW: btw OE is good choice for quick delivery project so you chose right :) Oct 27 01:16:52 and if its arm then you have a lot of support Oct 27 01:16:57 khem, yessir, beats the heck out of buildroot Oct 27 01:18:00 thing that messed me up initially with OE was that I was using the dev branch, too much weird stuff with that branch. Finally woke up to the stable branch existence Oct 27 01:18:35 T0mW: these days master/dev branch is always in pretty good shape Oct 27 01:18:40 we have weekly testing Oct 27 01:18:47 and if you want you can checkout that tag Oct 27 01:19:04 ok Oct 27 01:19:10 I am planning on some sort of releases too but I dont know how much time I could put into it Oct 27 01:19:14 in coming months Oct 27 01:19:27 desire is to have a release atleast once a year Oct 27 01:19:54 I built this board about 3 years ago, now the customer is interested in the design. so, I'm back to OE. Oct 27 01:20:04 http://wiki.openembedded.org/index.php/Testing Oct 27 01:20:14 T0mW: ah Oct 27 01:20:23 are you in westcoast Oct 27 01:20:42 khem: eastern Pennsylvania Oct 27 01:20:58 ok Oct 27 01:21:35 khem: I've been fortunate to stay out of the workforce for the past 15 years and doing consulting. Oct 27 01:22:12 though, last 2 years have been brutal Oct 27 01:22:38 yeah I imagine Oct 27 01:22:45 but thats long enough time Oct 27 01:23:15 khem, finally getting some decent work in again. Oct 27 01:23:35 thats encouraging. I hope it goes well all the way Oct 27 01:23:35 khem, almost went looking for a day-job! Oct 27 01:23:57 do u do hardware design mainly ? Oct 27 01:24:31 03Khem Raj  07master * r97222a4ffc 10openembedded.git/recipes/scratchbox/ (sbrsh-7.1/add-limits.h.patch sbrsh_7.1.bb): Oct 27 01:24:31 sbrsh: Fix complilation with gcc 4.5 Oct 27 01:24:31 Signed-off-by: Khem Raj Oct 27 01:24:32 khem, both. I love doing the software, but need to design the hardware it runs on. Also like doing that too. Oct 27 01:24:34 03Khem Raj  07master * r9feccf3a73 10openembedded.git/recipes/kstars/ (2 files in 2 dirs): Oct 27 01:24:34 kstars-embedded_0.8.5.bb: Fix complilation with gcc 4.5 Oct 27 01:24:34 Signed-off-by: Khem Raj Oct 27 01:24:34 03Khem Raj  07master * rb3125ddae3 10openembedded.git/recipes/linux-ha/ (cluster-glue/fix-const-cast.patch cluster-glue_1.0.5.bb): Oct 27 01:24:35 cluster-glue_1.0.5.bb: Fix compilation w.r.t. using constant cast using gcc 4.5 Oct 27 01:24:35 Signed-off-by: Khem Raj Oct 27 01:24:37 03Khem Raj  07master * ra578c12219 10openembedded.git/recipes/supertux/ (files/compilation-fixes.patch supertux_0.1.3.bb): Oct 27 01:24:37 supertux_0.1.3: Fix complilation with gcc 4.5 Oct 27 01:24:37 Signed-off-by: Khem Raj Oct 27 01:24:38 03Khem Raj  07master * rd9f64014e9 10openembedded.git/recipes/ubahnnav/ (ubahnnav/gcc45-compilation-fixes.patch ubahnnav_0.4.1.bb): Oct 27 01:24:38 ubahnnav_0.4.1.bb: Fix complilation with gcc 4.5 Oct 27 01:24:38 Signed-off-by: Khem Raj Oct 27 01:24:40 03Khem Raj  07master * r0058d3f85f 10openembedded.git/recipes/aspell/ (aspell-native_0.60.5.bb aspell_0.60.6.bb): Oct 27 01:24:40 aspell: Use BBCLASSEXTEND for native recipe Oct 27 01:24:40 Signed-off-by: Khem Raj Oct 27 01:24:41 03Khem Raj  07master * r1b857681e6 10openembedded.git/recipes/lvm2/ (7 files in 2 dirs): Oct 27 01:24:41 lvm2: Add recipe for 2.02.75 version Oct 27 01:24:55 stable is a mixed blessing Oct 27 01:25:01 IMHO Oct 27 01:25:03 * jconnolly chimes in Oct 27 01:25:15 jconnolly: well we clearly need a release. Oct 27 01:25:28 the thing is to plan it Oct 27 01:25:29 jconnolly: yeah, you can't bleed while working with it. Oct 27 01:25:46 jconnolly: we have to get sctricter with our APIs Oct 27 01:26:01 khem: bitbake APIs? Oct 27 01:26:03 or say predictable Oct 27 01:26:19 jconnolly: incompatible changes Oct 27 01:26:22 that we do Oct 27 01:26:40 right now its not that easy to backport fixes into stable e.g. Oct 27 01:26:50 the staging move was great but we're so far back now at my company using stable-2009. lots of shiny delicious things in dev that are very painful to merge into stable Oct 27 01:26:57 but this will not be that big issue if we released often Oct 27 01:27:03 right Oct 27 01:27:13 I want to help more Oct 27 01:27:26 jconnolly: ok then get into Testing mode Oct 27 01:27:34 and do continuous testing Oct 27 01:27:42 I'll read up on it. that's a good start Oct 27 01:28:41 jconnolly: I might step back and propose about making a release Oct 27 01:28:56 khem: you'd have stefan_schmidt agreeing I think Oct 27 01:28:57 say in Jan Oct 27 01:29:27 jconnolly: yes, I think testing is an effort that we should ramp up Oct 27 01:29:43 and that will eventually lead to a good release Oct 27 01:30:11 problem is that distro's release on their own schedule Oct 27 01:30:21 like angstrom etc. so it has to be coordinated Oct 27 01:30:28 I wonder what the workflow for the stable 2009 release was. Oct 27 01:30:43 angstrom has a release schedule? Oct 27 01:31:05 jconnolly: I would like to identify a set of architectures that we would support as release criteriea Oct 27 01:31:21 limiting scope sounds like a good first step too Oct 27 01:31:23 and then add continous testing of those into weekly testing Oct 27 01:31:36 we can not qualify all combinations and permutations Oct 27 01:31:41 indeed Oct 27 01:32:10 hrw was our advocate and guide for the 2009 stable release based on angstrom Oct 27 01:32:24 if we had a forum I could put up a voting meter to identify most used machines Oct 27 01:32:38 mostly what I know is armv6/armv7a, angstrom (and before that poky) Oct 27 01:33:02 yeah we have to also cover other arches like mips ppc and sh Oct 27 01:33:06 and x86 Oct 27 01:33:12 may be a machine of earch Oct 27 01:33:25 so people could use stable to get going on those Oct 27 01:33:39 I'd be happy to test armv6/armv7a perhaps even x86. Oct 27 01:33:45 brb Oct 27 01:33:45 cool Oct 27 01:33:57 lets see what the outcome of OEDEM is Oct 27 01:34:08 then we can work towards a release Oct 27 01:34:41 may be Jan and July would be a good bi annual release schedule Oct 27 01:46:32 hi , is alsamixer having microphone control inside ? i vent seen any analog microphone controls there , does someone know any solution ? Oct 27 01:54:26 biannual releases are great motivators... Oct 27 01:56:07 we strive for even more frequently at BUG but being stuck on stable makes me worried. we wouldn't be able to get anything out the door release-wise until Jan 2011 anyway though. We'd like linaro-based toolchain/kernel too but it looks like those upstream goodies are just trickling into dev Oct 27 01:56:39 I think stefan_schmidt at OEDEM will voice some of this though Oct 27 01:56:46 I hope so anyway. ;D Oct 27 02:28:31 hi all Oct 27 02:29:22 anyone tried to compile the ti-dsplink recipe against the 2.6.36 kernel? Oct 27 02:33:00 (it needs a small patch, which i've figured out and implemented, but i'm not sure exactly how to create the patch and submit it to be included in OE) Oct 27 02:33:56 mrj10, your best bet is to look at other recipes as an example to see how they do it. Oct 27 02:35:15 yeah, good call, it looks like people use git to create the patch but i'm not too handy with it yet Oct 27 02:35:29 mrc3, and there is this: http://wiki.openembedded.org/index.php/How_to_submit_a_patch_to_OpenEmbedded Oct 27 02:35:37 oops, i mean mrj10 Oct 27 02:35:39 i can probably figure it out. should i just post to the developer mailing list once it's done? Oct 27 02:35:55 even better, thanks grg Oct 27 02:36:02 should've looked for that before bugging people Oct 27 02:36:16 Yes, the -devel list is the correct one for pretty much everything. Oct 27 02:36:32 was also curious if anyone else had run into this issue, but probably most aren't dumb enough to use the 2.6.36-rc7 recipe Oct 27 02:36:38 :P Oct 27 02:37:29 can't help with that, i dont build for arm Oct 27 02:37:43 fair enough Oct 27 02:37:59 you can check http://tinderbox.openembedded.net/ though, it might show something useful Oct 27 02:38:04 once i figure this out i'll whip up a patch for another annoying problem with this stupid beagleboard Oct 27 02:38:40 (the board is great, just a couple annoying problems like always booting with a random MAC address) Oct 27 02:38:55 that does sound annoying Oct 27 02:39:34 yeah, it looks like you and i make up most of the first page on tinderbox :P Oct 27 02:40:11 it wouldn't be a problem except the network admins at the office do MAC filtering; at home it DHCPs itself an IP address no problem **** ENDING LOGGING AT Wed Oct 27 02:59:57 2010