**** BEGIN LOGGING AT Sun Mar 04 02:59:58 2007 Mar 04 03:20:06 can anyone tell me quickly the correct syntax for PREMIRROR? Mar 04 03:20:34 i know it needs some regular expression patterns, but not exactly which Mar 04 03:20:43 and there seems to be no single example on the web Mar 04 03:31:24 mh. time to go to bed Mar 04 03:58:23 hey Mar 04 08:11:21 hi all **** ENDING LOGGING AT Sun Mar 04 09:44:25 2007 **** BEGIN LOGGING AT Sun Mar 04 09:44:57 2007 Mar 04 10:18:50 03pH5 07org.oe.dev * rceff47ff... 10/ (1 packages/gnome/libbonobo_2.14.0.bb): libbonobo: QA issues, move debug files out of -dev package Mar 04 10:29:58 morning all Mar 04 10:50:09 hi RP Mar 04 10:50:11 hi zecke Mar 04 10:51:22 hey Mar 04 10:54:44 hi zecke Mar 04 11:00:32 RP: my final say: http://lists.linuxtogo.org/pipermail/openembedded-devel/2007-March/001598.html Mar 04 11:11:32 koen|away: I guess we need some opinions from the other core devs Mar 04 11:39:29 03pH5 07org.oe.dev * rafeec6e5... 10/ (1 packages/gnome/libbonoboui_2.14.0.bb): libbonoboui: QA issue, move debug files out of -dev package. Mar 04 11:52:15 OT: which RSS library is maemo using (if any) Mar 04 11:58:35 zecke: liferea iirc Mar 04 12:00:00 hmm Mar 04 12:00:10 1.3mb of sourcecode hmm Mar 04 12:02:38 anyway off to bed... Mar 04 12:02:57 zecke: wow. 1.30 pm ;) Mar 04 12:03:40 LaF0rge: yeah, I'm still sick and beside the wish to work, my body completely refuses to carry out any work Mar 04 12:03:58 zecke: oh, ok. I thought you hacked all night and half the day through ;) Mar 04 12:04:13 zecke: anyway, hope you get better soon. Mar 04 12:04:14 no, not this time :( Mar 04 12:06:31 gws zecke Mar 04 12:22:34 03pH5 07org.oe.dev * r1a452f14... 10/ (1 packages/ppp/ppp_2.4.3.bb): ppp: debug package fix, install all debug files. Mar 04 12:22:40 03pH5 07org.oe.dev * r01aabe23... 10/ (1 packages/tslib/tslib_1.0.bb): tslib: dev package fix, install all .la files. Mar 04 12:34:46 03rpurdie * r764 10bitbake/lib/bb/msg.py: msg.py: Add events for standard message types Mar 04 12:38:56 03rpurdie * r765 10bitbake/lib/bb/providers.py: providers.py: Remove now unused parameter from filterProviders Mar 04 12:43:12 03rpurdie * r766 10bitbake/lib/bb/shell.py: shell.py: Remove build_cache_fail references since this no longer works as intended Mar 04 12:44:15 RP: are there any news about bitbake epoch handling? Mar 04 12:44:49 pH5: Not yet. I have a big pile of pending changes. Once I wade through them, I'll look at it Mar 04 12:45:19 RP: ok, thanks. Mar 04 12:46:06 03rpurdie * r767 10bitbake/lib/bb/event.py: event.py: Remove data dependencies to resolve a dependency loop Mar 04 12:46:59 how do i set a svn patch in SRC_URI? Mar 04 12:47:26 svn://svn.foo.bar/trunk/my.patch;patch=1" doesnt work Mar 04 12:49:02 monsti: http://svn.o-hand.com/view/poky/trunk/meta/packages/qemu/qemu_cvs.bb?rev=1164&view=auto is an exmaple of what you want Mar 04 12:49:17 monsti: Basically a separete svn line and a file:// patch Mar 04 12:49:19 thx! Mar 04 12:49:45 what version of bitbake does this require? i am ob 1.6.6 Mar 04 12:50:51 monsti: Good question, I'm not 100% sure Mar 04 12:52:19 morning, koen Mar 04 12:52:39 hey NAiL Mar 04 12:56:16 RP: RP what BB version do you have? Mar 04 12:57:34 monsti: That file is in poky and poky uses bitbake trunk Mar 04 13:02:17 doesn't work Mar 04 13:12:02 03koen 07org.oe.dev * r78f0eb3e... 10/ (5 files in 3 dirs): tzcode, tzdata: update to something that actually is fetchable, remove hardcoded paths, install binaries in correct staging dir and make sure we use the correct 'zic' Mar 04 13:36:05 * Marex waves on koen :b Mar 04 13:43:44 03koen 07org.oe.dev * r5baf7c04... 10/ (3 files in 3 dirs): openmoko-session: autostart the gsm plugin now it has been converted to mb-panel 2 Mar 04 13:43:54 hi Mar 04 13:48:48 03rpurdie * r768 10bitbake/lib/bb/ (cooker.py runqueue.py): cooker.py, runqueue.py: Move statistics implementation from cooker to runqueue Mar 04 13:54:58 RP: I mentioned on here yesterday about a problem I was having with some of the openchrome packages being named differently to how they are specified in PACKAGES Mar 04 13:55:12 and you told me that debian.bbclass doing some kind of package renaming Mar 04 13:55:19 tkp: yes Mar 04 13:55:42 RP: do you know a way I can get round this, and get the packages installed on my root image? Mar 04 13:56:27 tkp: There should be no need to get around it Mar 04 13:56:49 tkp: The only known issue is if you're trying to specify them in PACKAGE_INSTALL Mar 04 13:57:01 If thats the problem, you should read the mailing list archives Mar 04 13:57:04 RP, ok, which I am Mar 04 13:59:11 tkp: Have a look at the mailing list archives since this was disucssed there recently Mar 04 13:59:22 RP: am searching now, thanks Mar 04 14:07:04 RP: is there a searchable archive of the mailing list somewhere? Mar 04 14:10:44 gmane or google site://lists.linuxtogo.org/pipermail/ Mar 04 14:11:16 koen: ah ok, thanks Mar 04 14:19:59 03rpurdie * r769 10bitbake/lib/bb/ (cooker.py runqueue.py shell.py): runqueue.py: Improve initialization instead of repeating parameters Mar 04 14:23:27 RP: http://rafb.net/p/SWHS1N31.html Mar 04 14:25:02 03rpurdie * r770 10bitbake/lib/bb/cooker.py: cooker.py: Remove old reference to BBStatistics Mar 04 14:25:06 koen: thanks Mar 04 14:26:50 it took a bit longer to spot since svn mangle the checkout with conflict markers *again* Mar 04 14:26:56 * koen stabs svn Mar 04 14:27:22 koen: you have a new fan: http://www.openprotium.org/tiki-index.php?page_ref_id=54 Mar 04 14:27:22 You have local changes to cooker? Mar 04 14:27:42 RP: no, runque debug output Mar 04 14:27:45 RP: http://rafb.net/p/wjr1Jf66.html Mar 04 14:27:59 I have no local diffs anymore in bitbake Mar 04 14:28:22 CoreDump|home: cool Mar 04 14:28:39 CoreDump|home: if all goes well, I'll be meeting Andy next month Mar 04 14:29:59 The iomega thingies are a bit on the pricy side for me Mar 04 14:30:13 "a bit"? Mar 04 14:30:29 they are way expensive over here Mar 04 14:30:46 right. ~300e or so Mar 04 14:31:23 * gremlin[it] want a Intel NAS SS4000 :) Mar 04 14:31:59 gremlin[it]: wouldn't a 4 drive thecus be cheaper? Mar 04 14:32:17 * CoreDump|home would settle for another slug or two =) Mar 04 14:32:24 gremlin[it]: http://www.thecus.com/products_over.php?cid=1&pid=2 Mar 04 14:32:31 they are sold under different names as well Mar 04 14:32:40 CoreDump|home: the slug doesn't have gbit :( Mar 04 14:33:13 but it's only ~80¤. Can't have everything :\ Mar 04 14:33:22 03rpurdie * r771 10bitbake/lib/bb/runqueue.py: runqueue.py: Remove extra now unneeded parameter Mar 04 14:33:25 koen yes i know but TheCus don't have the minipci and only one gigabit lan ... Mar 04 14:34:18 RP: thanks Mar 04 14:34:47 koen: I was testing things but you've highlighted I was testing the wrong copy of bitbake :/ Mar 04 14:34:51 * mwester has many bills and settles for the DSM-G600 (RevA). Mar 04 14:34:59 * RP has too many copies of bitbake Mar 04 14:36:47 RP: next one: http://rafb.net/p/THOBe417.html Mar 04 14:38:46 koen: That one is interesting. I'm not sure I've changed that code :/ Mar 04 14:40:30 03rpurdie * r772 10bitbake/lib/bb/runqueue.py: runqueue.py: Try to kill remaining processes but don't error if they no longer exist (reported by koen) Mar 04 14:46:28 03rpurdie * r773 10bitbake/ (bin/bitbake lib/bb/cooker.py): cooker.py: Move some functionality to the module init function Mar 04 14:47:08 RP: preparing to land the gui? Mar 04 14:48:20 * koen wtfs Mar 04 14:48:28 File "/data/build/oe/bitbake/lib/bb/runqueue.py", line 528, in execute_runqueue_internal Mar 04 14:48:28 os.kill(-k, signal.SIGINT) Mar 04 14:48:28 OSError: [Errno 3] No such process Mar 04 14:48:33 I did do svn up Mar 04 14:49:45 It should just ignore that now... Mar 04 14:50:29 indeed Mar 04 15:06:41 03rpurdie * r774 10bitbake/lib/bb/runqueue.py: runqueue.py: Refactor some functions/data Mar 04 15:07:07 RP: hey, should we tag 1.8.0 now? Mar 04 15:07:42 zecke: not yet. I want to rearrange the code first, then we branch. Only after we shake down test the branch do we tag Mar 04 15:08:38 * zecke hates to be obsessed... Mar 04 15:09:52 zecke: I just want any 1.8.x branch to be more easily diffed agaist 1.9.x if possible. I have a load of changes that should make 1.8.x nicer too so they're going in first Mar 04 15:10:37 RP: at the risk of breaking poky, and making mallum scream at you? Mar 04 15:10:45 RP: it is your call :) Mar 04 15:11:13 zecke: I'm being careful about the changes I'm choosing ;-) Mar 04 15:16:26 RP: http://rafb.net/p/NO5CcR44.html for some more context Mar 04 15:17:02 zecke: gws Mar 04 15:17:40 koen: Did you press Ctrl+C? Mar 04 15:17:51 RP: not that I know off Mar 04 15:18:01 koen: hehe, I'm in bed feel sick but I'm horrible obsessed... Mar 04 15:18:22 zecke: I got my dvb-t stick working today :) Mar 04 15:18:40 (they activated the new broadcast tower ~5km from here) Mar 04 15:19:29 koen: welcome to the DVB elite Mar 04 15:21:15 koen: what's the preferred way to have a certain set of packages built (like a task) but not installed in the image? Mar 04 15:21:16 zecke: it is kind of stupid that I can't view the 'encrypted' channels, even if I paid to view them via analog cable Mar 04 15:21:34 oxo: I think a task Mar 04 15:21:56 well, that has the effect of only having the meta-package built, without the dependencies Mar 04 15:22:04 hmmm Mar 04 15:22:15 there was a trick for that.... Mar 04 15:22:16 "Sending SIGINT to remaining 98 tasks" Mar 04 15:22:22 * koen tries to remember Mar 04 15:22:25 I've gone and broken something... Mar 04 15:22:28 koen: interesting, encrypted programs Mar 04 15:22:36 someone suggested that the meta-main-to-be-installed-task could RRECOMMEND the other tasks, but I'm sceptical Mar 04 15:22:36 ~emulate mallum Mar 04 15:22:54 zecke: I've not checked it into poky yet ;-) Mar 04 15:23:24 oxo: the behaviour is a bit different between different bitbake versions iirc Mar 04 15:24:03 koen: All supported versions of bitbake should work the same (ish) Mar 04 15:24:12 RP: what do we need to merge from poky to get package_write to get run for packages within a task? Mar 04 15:24:48 koen: I'm not sure what you mean? Mar 04 15:26:18 'bitbake task-foo' will build stuff, but not write out the actual packages to deploy Mar 04 15:27:02 koen: Hmm. That should work... Mar 04 15:27:25 koen: bitbake 1.6 or bitbake trunk (or both?) Mar 04 15:30:39 I had the problem with the 1.6.6-1 (debian) Mar 04 15:40:19 koen: I'll have to look at the bitbake trunk problem later, I'm out of time now. It might be a problem I've been seeing with the UI for a while so its perhaps good in that sense though :) Mar 04 15:40:29 * RP -> back later Mar 04 15:48:53 Does anyone know if it's possible to change the flash layout "on the fly"? I'm trying to work around a vendor firmware that has the partitioning scheme set in the kernel. Mar 04 15:49:24 NAiL: mtdparts on cmdline? Mar 04 15:49:35 NAiL: but the cmdline might be compiled into the kernel as well? Mar 04 15:50:00 I can change the kernel cmdline from the bootloader, fortunately Mar 04 15:50:58 is there a way to change the partitioning from userspace? Mar 04 15:55:20 NAiL: I don't know a straight forward way Mar 04 15:55:44 zecke: np, thanks ;) Mar 04 16:06:37 sup? Mar 04 16:07:02 hello? Mar 04 16:11:48 03freyther 07org.oe.dev * r854c81da... 10/ (5 files in 4 dirs): Mar 04 16:11:48 packages/{libmrss,libnxml}: Add two interesting libraries from Andrea Marchesini Mar 04 16:11:48 mrss is a simple RSS library which is suing curl and libnxml Mar 04 16:17:49 If my program needs libpng and zlib all I need to do is adding DEPENDS = "zlib libpng" in my bb file, right? Mar 04 16:19:16 sounds right Mar 04 16:20:12 hmm, but during linking it fails. Moment I pastebin it Mar 04 16:20:36 zecke: you have some 'foo//bar' in the SRC_URI's you just added Mar 04 16:21:12 koen: you are probably right, but this is the original link Mar 04 16:21:25 koen: http://www2.autistici.org/bakunin//codes.php Mar 04 16:21:35 zecke: bb file: http://rafb.net/p/W9p6eL70.html Error: http://rafb.net/p/Kr9Fvg91.html Mar 04 16:22:17 don't we have a fbgrabber in OE yet? Mar 04 16:22:39 stefan_schmidt: LDFLAGS += "-L${STAGING_LIBDIR}" ? Mar 04 16:23:30 zecke: This one has network support and viewer on client side. We already asked about upstream inclusion. Need to ping again. Mar 04 16:23:47 stefan_schmidt: hmm I have that http://pastebin.ca/381110 in my fbgrab for my EZX phone hmm Mar 04 16:24:31 stefan_schmidt: as koen said, LDFLAGS in the Makefile is missing Mar 04 16:25:02 koen, zecke: ah, nice. thanks. Mar 04 16:25:33 (and this patch is in OE) Mar 04 16:26:28 zecke: I take it and ping the author again to get it into upstream Mar 04 16:26:44 good! Mar 04 16:27:07 http://www.openembedded.org/bonsai/query/file/fbgrab/makefile.patch/ Mar 04 16:27:22 koen: you wrote OE in one day, you must be better than MVista? Mar 04 16:27:45 zecke: and cheaper Mar 04 16:29:08 zecke: how do we want to handle md5sums for stuff like openmoko-panel-demo-simple_svn.openmoko.org_.trunk.src.target.OM-2007.examples__20070304.tar.gz ? Mar 04 16:30:10 koen: as I'm a nihilist I would say md5sum = 1234 on SCM based systems. But I forsee people complaining about broken builds Mar 04 16:30:52 zecke: I'm putting backdoors in the svn tarballs on the angstrom mirror right now ;) Mar 04 16:31:45 koen: recipes pinned to a date/revision can/should have a md5sum Mar 04 16:31:52 koen: but how do you like my new Job Title? Mar 04 16:32:01 I like your new job title Mar 04 16:32:24 zecke: svn/cvs checkouts contain local info iirc Mar 04 16:33:17 koen: If you have followed GettingStarted you installed it into /stuff Mar 04 16:34:32 koen: we could start using svn export, similiar for cvs. Alternatively disable source mirror for svn/cvs, or stop using svn/cvs based recipes for distributions? Mar 04 16:34:39 zecke: do you have a semacode decoder for osx? Mar 04 16:34:57 zecke: all 3 options make sense Mar 04 16:35:10 koen: but this is tricky. People should switch to git/bzr/hg/mtn with builtin security Mar 04 16:35:13 zecke: could you respond to the list with that? Mar 04 16:36:24 koen: semacode: isn't the java version working? Mar 04 16:43:23 ah, it works now Mar 04 16:44:13 re Mar 04 16:44:33 zecke: http://www.flickr.com/photos/koenkooi/410125572/ :) Mar 04 16:45:15 I wonder how long it would take me to code a semacode parser in matlab Mar 04 16:45:23 koen: hehe, hehe Mar 04 16:45:26 koen: not php? Mar 04 16:45:35 peel. Mar 04 16:45:41 zecke: :) Mar 04 16:45:44 koen: will you redo the oe site with octave? Mar 04 16:46:21 zecke: only if you expend webcore to handle embedding of octave widgets :) Mar 04 16:46:40 koen: you have file output in octave? Mar 04 16:47:45 zecke: is the greenphone camera hooked up to the pxa27x camera IF? Mar 04 16:48:09 koen: I don't know this yet Mar 04 16:48:26 koen: I assume it is, it is pretty lame hardware (thanks to the highly integrated pxa27x) Mar 04 16:51:26 one less driver to write Mar 04 16:51:58 now you only need to flip a coin if you want to use justins or stanleys patches.... Mar 04 16:53:17 koen: I still need to find the power enable calls for the camera Mar 04 16:53:33 koen: I also found out that hh.org has a driver for the pxa27x keypad Mar 04 16:53:55 for i in `seq 1 255` ; do devmem2 0x3432 $i ; done ? Mar 04 16:53:59 something like that ? Mar 04 16:54:50 koen: more like objdump -D Mar 04 16:55:10 time to pick up the gf from the station Mar 04 16:55:12 later all Mar 04 17:01:09 zecke, some library is suing another? :) Mar 04 17:03:06 stefan_schmidt: https://svn.datenfreihafen.org/iec16022/ ? Mar 04 17:03:16 Crofton: I lost the context Mar 04 17:05:18 your last commit messgae :) Mar 04 17:12:19 morning Mar 04 17:19:36 zecke: yes, one of my research projects. Way to less time for it Mar 04 17:20:17 zecke: I adapted it from a company. Also have the standard here, just no time. Mar 04 17:20:48 stefan_schmidt: but the link is not working ;) Mar 04 17:21:10 stefan_schmidt: and you don't have a decoder Mar 04 17:21:46 zecke: right, only encoder. There are some C# decoder based on this. Mar 04 17:22:02 zecke: damn, svn repo is broken. Mar 04 17:22:37 zecke: I'll take a look after finding out where bitbake caches the SRC_URIS Mar 04 17:23:46 stefan_schmidt: it is not caching them at all Mar 04 17:24:07 zecke: hmm, so why I have this: Mar 04 17:24:15 NOTE: Update svn://svn.o-hand.com/repos/dates/branches;module=openmoko;proto=http Mar 04 17:24:15 Error validating server certificate for 'https://svn.o-hand.com:443': Mar 04 17:24:27 I changed from https -> http in the bb file Mar 04 17:25:41 stefan_schmidt: changed? Mar 04 17:25:59 stefan_schmidt: if you changed it you're using the old and broken moko svn overlay Mar 04 17:26:52 koen: Yes, I do. If I work for openmoko, I have to use their stuff. Mar 04 17:27:07 well, it's broken Mar 04 17:27:36 koen: yeah, but a changing url should still be possible Mar 04 17:27:41 I have fixes, but I'm waiting for a response to http://lists.openmoko.org/pipermail/distro-devel/2007-March/000017.html Mar 04 17:28:20 koen: I read and I really looking forward into merging the stuff. But it's more or less mickey|austria decision Mar 04 17:29:09 * Crofton wishes he was in Austria skiing :) Mar 04 17:29:41 koen, what needs work on uboot? Mar 04 17:30:14 Crofton: I live in Austria for 15 years or so, have not been skiing once:) Mar 04 17:31:17 koen: it's unlikely that somebody will respond now that the most qualified person for doing so is on holidays Mar 04 17:31:32 koen: I'm fine with that proposal, but I'm not an expert in that area ;) Mar 04 17:31:38 Crofton: I am a bit behind with merging the openmoko uboot stuff, as they are changing their patches to get them upstream Mar 04 17:31:45 ah Mar 04 17:31:53 so patches to existing uboot Mar 04 17:31:57 LaF0rge: I know, I'm just saying I'm waiting on feedback :) Mar 04 17:32:21 LaF0rge: I don't expect feedback till the p0 device is on my desk :) Mar 04 17:32:23 koen: as for phase0: there is no fixed version Mar 04 17:32:41 koen: thursday I guess. maybe wednesday Mar 04 17:33:10 koen: and what's broken about our overlay? Mar 04 17:33:11 LaF0rge: I kind of expect the courier to mess up and deliver it after I leave for brazil Mar 04 17:33:27 koen, http://www.omg.org/news/meetings/SBC2007/index.htm Mar 04 17:33:39 I will be amongst software types all week Mar 04 17:33:49 LaF0rge: all sorts of minor stuff, from packaging errors, to wrong patches to not using mb-panel 2 Mar 04 17:34:15 koen: but how does all that relate to a URL change not becoming visible after you change it? Mar 04 17:35:44 koen: but how does all that relate to a URL change not becoming visible after you change it? Mar 04 17:35:53 LaF0rge: it relates in a way that the url had been fixed in oe some $long time ago Mar 04 17:36:23 you're probably hitting some cache issue Mar 04 17:36:32 koen: don't complain to me, I'm not in charge of the packaging. honestly. I'm the happier the more stuff gets into mainline oe Mar 04 17:36:53 cache issue: yes, I can see that bitbake has many shortcomings in that area :( Mar 04 17:36:53 LaF0rge: openmoko userspace is already in OE ;0 Mar 04 17:36:56 :) Mar 04 17:37:22 koen: Yes, but I am not the one who will update to recent OE snapshot/mtn Mar 04 17:37:34 keon: I have not the slightest idea what kind of fallout that might produce Mar 04 17:37:36 LaF0rge: actually, oe .dev has more openmoko packages as the svn (openmoko-messages, openmoko-today, etc) Mar 04 17:37:54 and wrt. bitbake: henryk is now doing paid work on that Mar 04 17:38:14 LaF0rge: not much fallout, i'm building openmoko from .dev since svn went public Mar 04 17:38:20 ok, gotta leave Mar 04 17:38:21 talk to you later Mar 04 17:42:08 LaF0rge: fwiw, mickeyl and i sketched out the merge patch, we just need to pick some revision and do tests Mar 04 17:42:12 grrrr Mar 04 17:42:19 return != backspace Mar 04 17:42:28 merge path, not merge patch Mar 04 17:44:19 koen: hehe, relax and take care of your gal Mar 04 17:52:21 03zecke123 * r775 10bitbake-parser/standalone/bitbakescanner.l: Print the name of the broken file Mar 04 17:52:43 03rpurdie * r776 10bitbake/lib/bb/runqueue.py: runqueue.py: Split execute_runqueue into components Mar 04 17:54:30 03zecke123 * r777 10bitbake-parser/standalone/: Ignore more files Mar 04 17:55:36 !oebug 1939 Mar 04 17:55:38 * * Bug 1939, Status: NEW, Created: 2007-03-04 10:28 Mar 04 17:55:39 * * r.farina(AT)adelphia.net: Void11 Mar 04 17:55:40 * * http://bugs.openembedded.org/show_bug.cgi?id=1939 Mar 04 17:55:45 do I have a commit friend? :-) Mar 04 17:55:58 or even a testing friend (but it should work) Mar 04 17:57:00 03zecke123 * r778 10bitbake-parser/parse-all.py: Compile all anonfunctions in one go Mar 04 18:13:34 RP: your last commit seems to have fixed the sigint issue Mar 04 18:16:54 does anyone succesfully build with x86_64 als host? Mar 04 18:18:09 oxo: iirc it depends on the host distro you are using Mar 04 18:18:34 * koen <- uses 32 bit userspace with 64 bit kernel Mar 04 18:18:35 anyone with commit mind pushing 1939 please :-) Mar 04 18:20:12 later Mar 04 18:20:40 * oxo is having trouble with 64-bit userland (gcc-cross 3.4) Mar 04 18:21:06 heh Mar 04 18:21:15 gcc 3 has a nice crosscompile bug Mar 04 18:21:42 compiling on x86 for arm yields a different result as building on x86_64 for arm Mar 04 18:21:53 maybe sparc suffers from the same Mar 04 18:22:57 aka "why is my kernel too big for the flash partition" problem Mar 04 18:23:53 koen, strange i never notice till i was using gcc3 ... Mar 04 18:28:26 koen: the problem here is that the gcc-build just fails Mar 04 18:28:33 but I think this box is going to be 32bit Mar 04 18:50:35 hi, I'm trying to bitbake FreeCiv - and fail on diet-x11 with: "/bin/sh: line 1: 4370 Floating point exception../src/util/makekeys ks_tables_h" can anyone help me with this? Mar 04 19:05:40 hi all Mar 04 19:06:54 hi florian Mar 04 19:17:21 hi florian Mar 04 19:18:02 gremlin[it]: hi! Mar 04 19:18:53 Zero_Chaos: did you get feedback on your void11 recipe? Mar 04 19:19:03 Zero_Chaos: I have some Mar 04 19:19:14 koen: There was a typo which that commit fixed ;-) Mar 04 19:19:15 Zero_Chaos: FILES_${PN} += "${libdir}/libvoid11.so" NO! Mar 04 19:19:38 Zero_Chaos: oe_runmake CC="${CC}" CXX="${CXX}" CFLAGS="${CFLAGS}" CXXFLAGS="${CXXFLAGS}" LD="${LD}" LDFLAGS="${LDFLAGS}" isn't oe_runake setting those as well? Mar 04 19:19:51 Zero_Chaos: DEPENDS = "" leave it away, some for PR0 Mar 04 19:19:59 Zero_Chaos: S = ${WORKDIR}/void11-0.2.0 use quotes! Mar 04 19:20:23 Zero_Chaos: on do_install your indention seems weird Mar 04 19:20:31 ich: it works, or it doesn't? Mar 04 19:20:52 I don't care about my indentation Mar 04 19:21:24 Zero_Chaos: this recipe is not valid, it might be parsed by bitbake1.6, but it is not valid Mar 04 19:21:37 it was parsed, as it by bitbake 1.6.4 Mar 04 19:21:49 Zero_Chaos: and specially the FILES_${PN} += "${libdir}/libvoid11.so" is just a NONO Mar 04 19:22:38 zecke: forgive me, my bitbake foo is weak, the receipe works... if there are problems I welcome fixes attached to my bug Mar 04 19:23:18 Zero_Chaos: well, this is not how it works. If this has been comitted I'm going to disapprove it for the reasons above Mar 04 19:23:41 zecke: I don't think it has been commited yet. Mar 04 19:24:09 zecke: but like I said, if you have some fixes I can make please tell me in the ticket, I know it is a bit dirty, but I wanted it working more than I wanted it pretty Mar 04 19:24:46 Zero_Chaos: why do you think you need to install the .so file? Mar 04 19:24:56 Zero_Chaos: why do you disable the -dbg package? Mar 04 19:25:09 zecke: the .so file is required for operation of the program Mar 04 19:25:19 zecke: it's not a devel lib Mar 04 19:25:24 koen: did you ever manage to build a gcc-cross-sdk 4.x? Mar 04 19:25:33 Zero_Chaos: it loads plugins from $libdir? wow this app dev needs to be shot Mar 04 19:25:47 zecke: that's not me, I just made it work Mar 04 19:25:55 florian: no, I'm still waiting for the fixes nils made Mar 04 19:26:40 Zero_Chaos: if this app loads plugins from ${libdir}, you should not even dare to package this crap Mar 04 19:27:04 koen: ok, let me take a look Mar 04 19:27:27 zecke: dude, I'm not rewriting it, I'm just cross-compiling it, if you won't add a program that has been used in linux for years to bitbake than you are only limiting the community Mar 04 19:27:51 Zero_Chaos: I will certainly not add it to bitbake, but you have not asked for that Mar 04 19:28:14 * koen hands zecke a cookie Mar 04 19:28:27 zecke: I asked it to be added to openembedded, that is why I posted it Mar 04 19:28:55 zecke: like I said, I appreciate help in fixing my terrible receipe but if you won't add it because it loads a lib then I'm not fixing it Mar 04 19:30:08 Zero_Chaos: I just say it(loading this from libdir) is a crime against humanity, and just because it is legal in some countries to beat your wife, it should not encourage you to do so ;) Mar 04 19:30:15 Zero_Chaos: as to the recipe, I will add the coments to the bug Mar 04 19:31:54 zecke: If I make the reciepe less ugly will you commit it despite the horrible loading of lib? Mar 04 19:33:34 Zero_Chaos: can you remove oe_compile? why do you need that? Mar 04 19:33:58 Zero_Chaos: also the last line in do_install, how many tabs/spaces did you add to force the line wrap in my browser? Mar 04 19:34:11 Zero_Chaos: why do you want to disable the -dbg package of void? Mar 04 19:34:54 zecke: there is no -dbg package, it is just the runtime lib Mar 04 19:34:55 koen: Where is my cookie? Can you put some frozen frikandel's into the box and send it to me? Mar 04 19:35:18 Zero_Chaos: yes there is no -dbg package because disabled it with your PACKAGES line Mar 04 19:36:03 zecke: I was under the impression that there was nothing to go into the -dbg package, I am incorrect? Mar 04 19:36:34 Zero_Chaos: OE will split debug symbols into another directory (.debug) and package that as -dbg package Mar 04 19:37:37 zecke: if that is a desired feature I don't see why I have a problem with it. I was unaware Mar 04 19:37:55 Zero_Chaos: ah, your libvoid11.so is probably fine. It is just a badly named library Mar 04 19:38:06 Zero_Chaos: e.g. try to remove the PACKAGES line and see what happens Mar 04 19:38:59 zecke: I will make the suggested changes and test it. Mar 04 19:40:28 zecke: FYI, according to my vi, my tabbing is fine, not sure why you have two lines where one should be Mar 04 19:41:24 koen: http://www.mcdonalds.de/data/products/promotion/schmecktakel/img/side.jpg take a look at the dolphin. McDonald gurantees that your thuna contains dolphin Mar 04 19:43:49 cool Mar 04 19:44:40 Dolphin salad? ;) Mar 04 19:44:58 zecke: complete, if you would be kind enough to look it over again Mar 04 19:47:09 Zero_Chaos: a lot cleaner, we don't use the MAINTAINER field anymore, instead feel free to patch the toplevel mainatiner file Mar 04 19:47:39 zecke: very well, will you remove the maintainer field and commit it? Mar 04 19:48:04 Zero_Chaos: one more thing (sorry), can you tell me the md5sum/sha256sum of void11 and hostapd? Mar 04 19:48:17 zecke: of the downloaded source files? Mar 04 19:48:43 Zero_Chaos: yes Mar 04 19:49:05 zecke: if that is something that can go into the .bb file I'd prefer you tell me how I should be doing it so that I can do it myself next time. Mar 04 19:49:26 Zero_Chaos: ;md5sum=SUM at the end of each url Mar 04 19:49:41 Zero_Chaos: and for now (which is unchecked) the same for ;sha256sum=sum Mar 04 19:50:00 okay, shall I fix it myself of can I just tell you the sums? Mar 04 19:50:21 Zero_Chaos: whatever you prefer Mar 04 19:50:28 1c5b3e3e70916de74c2932c7f3e46d9e void11-0.2.0.tar.bz2 Mar 04 19:51:00 54563fb51f143c4bf26ddec2516e8f9f hostapd-0.1.3.tar.gz Mar 04 19:51:12 I will do that myself next time, but I appreciate the help Mar 04 19:51:19 didn't know I could do that Mar 04 19:52:37 zecke: remember the qtopia source changing md5sum? Mar 04 19:52:59 koen: QtE, but that was probably my error Mar 04 19:53:19 zecke: didn't we compare the md5sums when we added it? Mar 04 19:53:21 koen: (human error), I wanted to check if my md5 code worked and probably accidently comitted the 'wrong' sum Mar 04 19:53:51 koen: IIRC only one digit was different, and it was the first which indicates "Zecke Error" Mar 04 19:54:02 ah Mar 04 19:54:21 koen: I just wonder why it was unnoticed for some time Mar 04 19:54:37 Zero_Chaos: I will transparently change the tabs away ;) Mar 04 19:54:58 zecke: your jedi tricks appear to work on some people Mar 04 19:55:10 Hello Mar 04 19:55:14 zecke: my vim shows proper tabs, if you have an issue, you are welcome to play with it, I really don't care about tabs :-) Mar 04 19:55:25 spaces no tabs :) Mar 04 19:56:23 zecke: if the only thing between my reciepe and commit is a few tab/space, help yourself Mar 04 19:58:55 Hi Mar 04 19:59:26 can anyone tell me how can i compile a simple app and put it in the rootfs? Mar 04 19:59:34 i sent an email some hours ago Mar 04 19:59:39 i have it compiled Mar 04 19:59:50 and inserted in slugos-init dir Mar 04 19:59:55 as flash file Mar 04 20:00:33 but im not sure, i think it gets reduced from 40k to 21k, and it is no stripping, since stripping reduces that file to 20k exactly Mar 04 20:04:06 zecke: I have go to run, if there are any issues that prevent commit just leave me a note in the bug and I'll take a look when I'm done, sigh, cleaning my office Mar 04 20:04:58 Zero_Chaos: comiting and pushing atm Mar 04 20:05:12 zecke: thanks, Mar 04 20:05:46 /bin/sh: line 1: 4498 Floating point exception../src/util/makekeys ks_tables_h Mar 04 20:05:53 sorry Mar 04 20:05:59 hi, I'm trying to bitbake FreeCiv - and fail on diet-x11 with: "/bin/sh: line 1: 4370 Floating point exception../src/util/makekeys ks_tables_h" can anyone help me with this? Mar 04 20:07:07 03Zero_Chaos 07org.oe.dev * ra46fe2d2... 10/ (6 files in 4 dirs): Mar 04 20:07:07 packages/void11: Add void11 a hacking tool for wireless networks Mar 04 20:07:07 Patch from Rick Farina from bug #1939 Mar 04 20:13:17 YoG im having the same problem Mar 04 20:14:16 saladino, do you know what it means? how to solve it? anything? Mar 04 20:14:27 nope Mar 04 20:14:45 stefan_schmidt: hey, what is your job? *neugier* Mar 04 20:15:10 zecke: openmoko Mar 04 20:15:24 zecke: But only for one month Mar 04 20:15:51 I'm having some trouble with tslib + randr screen resolution changes. Mar 04 20:15:56 stefan_schmidt: What part on OpenMoko? Emulating a phone, being a 'dummy' unit? Mar 04 20:16:50 Should tslib be aware of resolution changes? Mar 04 20:16:59 stefan_schmidt: anyway, have fun and good luck Mar 04 20:17:23 zecke: Before the device arrives some bugquashing. Afterwards gsm passthrough mode and 'LCM switch to qvga mode' Mar 04 20:17:31 saladino, there are some bugs that maybe related in http://bugs.treke.net/, check #1543, #1411 and #1412.... Mar 04 20:17:34 zecke: thanks, I hope I'll have. :) Mar 04 20:23:10 03rpurdie * r779 10bitbake/lib/bb/cooker.py: cooker.py: Split cook into several functions Mar 04 20:23:40 Zero_Chaos: get rid of your hotmail address (oh well maybe I'm a chauvinist...) Mar 04 20:44:30 RP: Hello Mar 04 21:03:40 03zecke123 * r780 10bitbake-parser/ (ast.py data_visitor.py evaluator_data.py): Mar 04 21:03:40 Start converting the parser from mix-in to Visitor using Mar 04 21:03:40 Python2.4 decorator features to create my jump_table on Mar 04 21:03:40 the fly Mar 04 21:03:57 * LaF0rge returns Mar 04 21:05:37 hi LaF0rge Mar 04 21:08:19 hi florian Mar 04 21:10:43 hi sirfred Mar 04 21:11:46 03zecke123 * r781 10bitbake-parser/data_visitor.py: Rename these methods as well Mar 04 21:13:34 03zecke123 * r782 10bitbake-parser/data_visitor.py: Recursively call the Root visitor Mar 04 21:13:40 CIA-3: quit Mar 04 21:26:20 woglinde: hey, happy anniversary (nachtraeglich) Mar 04 21:26:58 woglinde: how far are you with building javaME? Mar 04 21:27:23 zecke sorry no time Mar 04 21:27:53 zecke: Would you say now is a good time for a bitbake-1.8 branch? Mar 04 21:28:12 RP: any time, so my new parser won't be in bitbake-1.8? Mar 04 21:28:54 zecke: That depends on how soon it will be ready I guess... Mar 04 21:29:17 RP: so now we will need to update the MANIFEST, run the imaginary test suite Mar 04 21:29:39 zecke: I'm planning to branch and request testing, not release yet Mar 04 21:30:05 The MANIFEST should be up to date? Mar 04 21:31:40 03nail 07org.oe.dev * rd48faafc... 10/ (4 files in 3 dirs): openprotium: Move TARGET_FPU to the individual machines instead of distro config Mar 04 21:31:43 RP: yes, but we should check it again Mar 04 21:35:58 RP: feel free to branch. But there is one thing to take care of Mar 04 21:36:20 RP: feature backports are forbidden, bugfixes must go to both branches Mar 04 21:36:32 RP: and we should try to release 1.8 quickly Mar 04 21:36:43 03zecke123 * r783 10bitbake-parser/ (TODO ast.py data_visitor.py): Mar 04 21:36:43 -Separate AST and Visitor (remove the root stuff from the AST) Mar 04 21:36:43 -Add a pseudo TODO with emphasis on how to test this parser Mar 04 21:36:43 in the absence of a test suite. Mar 04 21:40:28 zecke: no backports? Mar 04 21:40:58 RP: no feature backports. "Oh I have this neat feature and it would be a shame if Bitbake 1.x would lack it" Mar 04 21:41:05 RP: Just say no to this temptation Mar 04 21:41:46 wow directfb1.0 is out Mar 04 21:41:56 and it is not 1st of april Mar 04 21:42:01 zecke: we have done it in the past... Mar 04 21:42:10 wow Mar 04 21:42:43 RP: after a release, I see trunk changing quickly, so we need to get 1.8.0 released ASAP Mar 04 21:42:54 RP: otherwise we end up with stupid parallel development Mar 04 21:44:10 zecke: I basically agree although I think making trunk work well will take a while... Mar 04 21:44:25 zecke: parallel development is fun ... I have 9 concurrent branches here ... and I can't merge any (stupid paying customers and managers) Mar 04 21:44:39 RP: yes, but we need to get 1.8.0 out :) Mar 04 21:45:03 RP: it is a non-easy from the current point of view, but we need to make sure it will stay like this Mar 04 22:24:09 zecke: Ok, I agree, no backports :) Mar 04 22:24:41 cya tomorrow, I declare my self ready for/to work... Mar 04 22:24:57 I thought you meant no backports onto 1.8.x at all :) Mar 04 22:25:06 zecke: cya Mar 04 22:42:23 03rpurdie 07bitbake-1.8 * r784 10/: Branch for bitbake 1.8 Mar 04 22:44:20 03rpurdie 07bitbake-1.8 * r785 10/ (bin/bitbake lib/bb/__init__.py): bitbake-1.8: Prepare for release Mar 04 23:27:32 RP: still around? Mar 04 23:30:30 robtaylor: yes Mar 04 23:30:45 RP: cool :) you're quite experienced at LTTng, right? Mar 04 23:31:15 03rpurdie * r786 10bitbake/ (bin/bitbake lib/bb/__init__.py): Update versions Mar 04 23:31:21 robtaylor: I guess you could say that Mar 04 23:34:23 RP: i was ltting udev startup, and there was a strange bit where udevtrigger is just sitting doing irq_entry, irq_exit, soft_irq_entry, softirq, settimer, softirq_exit for about 1/10 of a second Mar 04 23:34:44 i was wondering if you might have an idea of what's happening there Mar 04 23:36:35 (cos i'm stumped ;) ) Mar 04 23:41:13 03pfalcon 07org.oe.dev * rb6170746... 10/ (1 conf/bitbake.conf): Mar 04 23:41:13 bitbake.conf: Set default value of ROOT_FLASH_SIZE to something big. Mar 04 23:41:13 * As common use for ROOT_FLASH_SIZE would be to not include some packages Mar 04 23:41:13 in the image if it is too small, the default value should be something big. Mar 04 23:41:13 Use 256Mb as we're on embedded stuff here. Mar 04 23:41:38 robtaylor: That sounds like normal background kernel timer noise? Mar 04 23:43:11 RP: i wonder what's taking the time then :/ Mar 04 23:44:12 robtaylor: Is it not waiting on something like a select? Mar 04 23:45:19 RP: not that i cna tell, its just after a close completes and an open is started Mar 04 23:46:19 robtaylor: Waiting on IO? Mar 04 23:46:56 hmm, could be. the close syscall returned Mar 04 23:49:57 RP: its all a bit strange, as this is just reading and writing sysfs Mar 04 23:51:08 from the udev code (udev 100), scan block completes, ui see the return from the clast close, then there's this block of inactivity just before the 1st open of exec_lists happens Mar 04 23:52:25 Is it in user or kernel space? Mar 04 23:55:47 RP: how do i tell? Mar 05 00:02:07 robtaylor: by the entry and exit times from any syscalls I guess Mar 05 00:02:31 RP: yeah, there's no syscall in process over that period Mar 05 00:05:17 robtaylor: It sounds like the process itself is chewing up cpu time then... Mar 05 00:05:40 RP: strange, as it s getting colourd blue rather than green in the display Mar 05 00:06:14 hmm, i guess i should add some reporting into udev and see whats takeing the time :) Mar 05 00:06:51 s/display/control flow viewer/ :) Mar 05 00:07:32 RP: in lttv-gui, that is Mar 05 00:30:24 'night all Mar 05 01:06:57 ugh....now qemu won't compile... Mar 05 01:07:02 * JustinP waves several fists Mar 05 02:12:19 Does anyone know what tasks are required for developers for openzaurus? Mar 05 02:17:42 hi guys Mar 05 02:18:17 could anyone give me some tips for image recipes? I'm building an image and it does root_fs but I don't seem to get a kernel or anything in the rootfs.. about 1.8mb total :( Mar 05 02:19:22 I'm beginning to wonder if openzaurus devel really is dead Mar 05 02:19:31 If i run "bitbake -v myimage" I see bitbake selecting all the correct packages but it just doesn't put the files in the image Mar 05 02:22:14 anyone with commit be a pal and fix the maintainer according to bug 1941? Mar 05 02:22:16 thanks Mar 05 02:28:17 the guys at #openzaurus pointed me to this channel, I am having a problem with automake on a spitz-like device Mar 05 02:28:34 spitz-like? Mar 05 02:28:53 an SL-C3200 in this case Mar 05 02:29:10 I am trying to build bzflag from source Mar 05 02:29:59 but for some reason libtool is looking in the wrong directory for my libraries, in a home subfolder that doesn't exist /home/hrw... Mar 05 02:30:14 why is that? Mar 05 02:30:42 you have BBPATH set in your environment? Mar 05 02:30:58 if you do and it is accurate maybe http://pastebin.ca your conf/local.conf Mar 05 02:31:11 I didn't mess with any of the system variables, if it isn't set in the install it's not set now Mar 05 02:33:35 I checked, bbpath is empty. what should it be set to? Mar 05 02:33:43 well if you read Getting Started on www.openembedded.org it might clear somethings up Mar 05 02:33:49 ok Mar 05 02:35:15 hmm... I think I was sent to the wrong place Mar 05 02:35:54 I am talking about ./configure; make; make install, that's not bitbake afaik Mar 05 02:36:48 yeah, you are in the wrong place then, this is a cross-compiling environment to help you Mar 05 02:38:19 somehitng like gentoo's portage, I've heard of it Mar 05 02:38:50 it's rather nice in my opinion, but then again, I use it :-) Mar 05 02:39:37 is it in ipk form or do I have to do a lot of work by hand to install it? I might try it. Mar 05 02:40:28 it makes ipk, it runs on your desktop Mar 05 02:40:36 read the website, you may like it Mar 05 02:41:09 ok **** ENDING LOGGING AT Mon Mar 05 02:59:57 2007