**** BEGIN LOGGING AT Sat Feb 03 02:59:59 2007 Feb 03 04:09:24 http://bugs.openembedded.org/show_bug.cgi?id=1842 <-- can anyone help me with this? Feb 03 04:13:44 JustinP: Ping Feb 03 05:38:58 cmd = "PATH=\"%s\" I18NPATH=\"%s\" %s -L %s/%s %s/bin/localedef %s" % (path, i18npath, qemu, bb.data.getVar("STAGING_DIR", d, 1), bb.data.getVar("TARGET_SYS", d, 1), treedir, localedef_opts) Feb 03 05:39:09 I'm almost certain there is an error on that line, with i18npath Feb 03 05:39:36 seems to refer to my local system instead of the oe build environment Feb 03 05:39:39 anyone? Feb 03 09:30:06 !oebug 172 Feb 03 09:30:08 * * Bug 172, Status: RESOLVED, Created: 2005-07-27 12:58 Feb 03 09:30:09 * * papercrane(AT)reversefold.com: ppp-2.4.3-r0 fails when linking libradiusclient.a Feb 03 09:30:10 * * http://bugs.openembedded.org/show_bug.cgi?id=172 Feb 03 11:26:38 hi Feb 03 11:27:52 I set up OE and bitbake as told in the getting started wiki page. machine is spitz, without telling a machine, it works... but with, i only get errors like "iterable argument missing" or "information not available for target" Feb 03 11:29:44 what's wrong? any missing information in the local.conf? doesn't look like... entered everything what's necessary :( Feb 03 11:31:19 and i followed the instructions in the wiki. Feb 03 11:35:26 03ifaistos 07unknown * rffc9da89... 10/ (1 packages/uboot/u-boot-1.1.4/u-boot-dht-walnut-df2.patch): Add missing uboot patch for dht-walnut Feb 03 11:35:30 03ifaistos 07org.oe.dev * r1fc9a835... 10/ (3 files in 3 dirs): Feb 03 11:35:30 Fix getline error when compiling for uclibc Feb 03 11:35:30 closes bug #1335 Feb 03 11:40:01 is it speaking to me? Feb 03 11:40:17 or ain't you a bot? :} Feb 03 11:42:04 CIA is a bot Feb 03 11:42:22 mickeyl : Hi ! Feb 03 11:42:29 cheers gu Feb 03 11:42:29 s Feb 03 11:42:30 guys Feb 03 11:45:10 mickeyl : i think i made a typo in the branch when commiting ia missing patch for uboot Feb 03 11:45:38 mickeyl : i missed a 'd' in openembedded Feb 03 11:46:16 uh oh Feb 03 11:47:14 mickeyl : does this create a new branch or another head ? Feb 03 11:47:21 koen: should we stop servers to prevent that branch being distributed? Feb 03 11:47:45 we sort of do Feb 03 11:47:57 sorry guys :( Feb 03 11:47:57 Ifaistos: it creates a new branch Feb 03 11:48:02 no problem, can happen Feb 03 11:48:06 koen: can you sort it out somehow? Feb 03 11:48:11 * koen wonders how it can happen Feb 03 11:48:14 * mickeyl having problems with his new mouse Feb 03 11:48:23 there is absoultely no need to use -b when committing Feb 03 11:49:17 especially since you can't commit to another branch from a working copy without messing up the ancestry Feb 03 11:49:52 i think i saw that in the wiki and been following it from start... Feb 03 11:50:28 the problem seems to be while parsing... (see above) Feb 03 11:50:48 any ideas? Feb 03 11:52:27 btw, any idea how the bugdays are coming along? Feb 03 11:54:28 not sure. i just closed one :) Feb 03 11:55:31 Ifaistos: if you do all mtn ops from your OE working directory, there is no need to use the --branch op -- wonder if we should remove that from the wiki:MonotonePhraseBook? Feb 03 11:56:39 cbrake : will do from now on.... Feb 03 11:59:37 I'm also not sure why the --db argument is also specified for the commit. Seems like if it is needed, it would also be required for the pull and update commands. Feb 03 12:03:09 * cbrake adds a notes to MonotonePhraseBook about commits ... Feb 03 12:13:49 can the problem by fixed on my local db or do i have to get a new copy ? Feb 03 12:26:10 mickeyl, would you mind being on a discussion panel on Sunday (lunch time) at Fosdem about Power and device management? Feb 03 12:28:02 philippe: not at all. device management is really interesting to me and I'm working on a general solution. having experience with opie HAL and stuff it might be a good opportunity for me. Feb 03 12:31:22 mickeyl, great! btw this also has to do with the Power stuff I am trying to get you involved in Feb 03 12:31:41 ok, cool Feb 03 12:31:50 we have a long way to go re. power Feb 03 12:31:59 especially on smartphones Feb 03 12:32:07 so i'm all ears to get involved Feb 03 12:32:36 hi, all! Feb 03 12:32:38 mickeyl, ok. Consider yourself official member of the discussion panel then :-) Feb 03 12:32:43 lrg, ping Feb 03 12:33:00 *nod* :) Feb 03 12:35:58 slapin_nb: pong Feb 03 12:36:16 g'day mickeyl Feb 03 12:36:20 hi lrg Feb 03 12:36:30 hey pb_ Feb 03 12:37:02 lrg, I have sent you email :) Feb 03 12:37:13 lrg: did you receive that second sample unit that we sent to you the other day? Feb 03 12:37:15 cheers pb_ Feb 03 12:37:18 lrg, could you please check it? Feb 03 12:37:27 philippe-budapes: is there any more spaces on panel ? I'm doing a lot of work on phone pm atm and would be happy to fill a chair Feb 03 12:37:45 pb_: yes thanks, brian has it set up in the lab now Feb 03 12:37:49 great Feb 03 12:38:00 lrg, Why not? Feb 03 12:38:01 hopefully with two of them he will be able to make it go wrong twice as often Feb 03 12:38:07 slapin_nb: checking mail. Feb 03 12:38:29 lrg, you will be at Fosdem then (Sunday 13-14?) Feb 03 12:38:36 philippe-budapes: yes Feb 03 12:38:49 ok. I' ll add you too. Feb 03 12:38:54 thanks. Feb 03 12:39:10 lrg, thank you... The more people we have the better Feb 03 12:40:08 cool Feb 03 12:40:58 philippe-budapes: fwiw, I should be staying at the astrid Feb 03 12:41:48 pb_: you're familiar with XComposite? Feb 03 12:42:13 lrg, I should be at my parents place. Makes them happy to see me again. Feb 03 12:42:25 heh, btw did you see my pm ? Feb 03 12:42:49 mickeyl: roughly speaking, yeah. I wouldn't describe myself as an expert. Feb 03 12:43:15 pb: cool. my question: does the Xcomposite extension affect performance even if you're not using fancy compositing but just accessing the backing store to display thumbnails in a taskmanager? Feb 03 12:44:02 yes, though not necessarily so severely as to cause a problem. Feb 03 12:44:29 whenever COMPOSITE is enabled, you have an extra level of indirection for all the drawing: it needs to go first into the backing buffer, and then be copied (under control of the compositor) onto the actual screen. Feb 03 12:44:44 ok, that's what i was afraid of Feb 03 12:44:47 it's a no go for us then Feb 03 12:44:51 if the compositor is basically a no-op and can use video card accel for the copy then it may not be too bad. Feb 03 12:45:10 we have severe problems on s3c2410 with a fancy pixmap theme Feb 03 12:45:21 speaking about pixmap themes... *hint* ;) Feb 03 12:45:26 oh right, s3c2410 would be no fun Feb 03 12:45:42 heh, right, I should get my finger out Feb 03 12:45:49 I might be able to look at that later this afternoon actually Feb 03 12:46:00 thanks, that'd be much appreciated! Feb 03 12:46:10 btw., did you get your lottery win email? Feb 03 12:46:13 *g* Feb 03 12:46:40 hm, don't think so. should I have? Feb 03 12:46:50 yes. you should have got one Feb 03 12:47:06 it may gone to your spam filter :/ Feb 03 12:47:14 oh dear Feb 03 12:47:15 since it had "lottery win" in the subject and came from a chinese address Feb 03 12:47:17 :/ Feb 03 12:47:29 heh, that's unfortunate Feb 03 12:47:43 doubly unfortunate because I just emptied my spam folder yesterday Feb 03 12:47:48 maybe I can find it in a backup Feb 03 12:48:25 do you happen to know which address it would have gone to? if they would have used my old nexus.co.uk address it will be languishing in another mailbox at the office. Feb 03 12:48:40 reciva.com and gnu.org both go to my main inbox though Feb 03 12:48:45 hold on Feb 03 12:49:00 11. Phil Blundell (Reciva CTO, glibc master, Gtk+ hacker, GPE hacker) Feb 03 12:49:06 oh dear Feb 03 12:49:10 pb@gnu.org is someone different Feb 03 12:49:14 heh Feb 03 12:49:15 philb@gnu.org is my address Feb 03 12:49:22 oops, sorry about that Feb 03 12:49:26 i'll fix it Feb 03 12:49:28 heh Feb 03 12:49:29 thanks Feb 03 12:49:47 which one should i use? Feb 03 12:49:50 easy mistake to make though, all my other emails are pb@ Feb 03 12:49:59 either philb@gnu.org or pb@reciva.com would be fine Feb 03 12:50:03 ok Feb 03 12:51:57 hrm, I wonder why stupid evolution can't talk to my mail server anymore Feb 03 12:51:57 ok, you should get it within two days. If not, please ping me Feb 03 12:52:02 great, thanks Feb 03 12:52:07 np Feb 03 12:52:10 I'll keep an eye on the spam box this time Feb 03 12:52:30 heh, probably not necessary. we learned from that mistake Feb 03 12:52:35 :-) Feb 03 12:52:36 new mails will come from xxx@openmoko.org Feb 03 12:52:41 and won't contain lottery win in subject Feb 03 12:52:46 ;) Feb 03 12:52:48 excellent Feb 03 12:52:56 bbiab, gotta go saw up some wood in the garage Feb 03 12:53:05 cu Feb 03 13:15:37 cheers do13 Feb 03 13:15:41 how are things? Feb 03 13:17:20 10 bugs down and counting :) Feb 03 13:17:25 http://bugs.openembedded.org/buglist.cgi?query_format=advanced&short_desc_type=allwordssubstr&short_desc=&product=Openembedded&long_desc_type=substring&long_desc=&bug_file_loc_type=allwordssubstr&bug_file_loc=&bug_status=RESOLVED&resolution=FIXED&emailassigned_to1=1&emailtype1=substring&email1=&emailassigned_to2=1&emailreporter2=1&emailqa_contact2=1&emailcc2=1&emailtype2=substring&email2=&bugidtype=include&bug_id=&chfieldfrom=2007-02-02&chfieldto=Now Feb 03 13:17:25 &chfieldvalue=&cmdtype=doit&order=Reuse+same+sort+as+last+time&field0-0-0=noop&type0-0-0=noop&value0-0-0= Feb 03 13:30:24 Ifaistos, FYI, there're recommendations how commit messages should be formatted, http://www.openembedded.org/monotonephrasebook Feb 03 13:36:10 psokolovsky : Was there a problem with a commit message ? Feb 03 13:36:56 yep, you miss to start it with package name, and that makes review harder, and search almost impossible Feb 03 13:37:46 psokolovsky: could be worse: http://www.openembedded.org/bonsai/view/scm/6/ Feb 03 13:43:53 mickeyl: http://linuxdevices.com/news/NS2797442785.html Feb 03 13:45:10 psokolovsky : Was aware of that... when this format started ? Feb 03 13:47:05 lrg, have you got that mail? Feb 03 13:48:43 koen: ah, once again a survey. let's encourage our guys to vote ;) Feb 03 13:48:51 * mickeyl checks whether OE is in from the start this time Feb 03 13:49:54 Ifaistos, was there when I got commit at Autumn ;-) Feb 03 13:50:40 indeed. it's there Feb 03 13:51:55 learn something new each day ;) Feb 03 14:28:37 hah Feb 03 14:28:39 "Would you consider paying per-unit royalties to your embedded Linux OS vendor?" Feb 03 14:29:50 is msoft considering a new product for embedded ? :) Feb 03 14:58:21 slapin_nb: email answered Feb 03 15:00:30 lrg, thanks a lot! Feb 03 16:24:44 morning all Feb 03 16:25:43 morning RP Feb 03 16:29:35 * RP has had enough of smashing holes in concrete for today... Feb 03 16:30:04 hey Richard Feb 03 16:32:30 re Feb 03 16:37:28 hey RP Feb 03 16:37:31 Jin^eLD: wb Feb 03 16:37:51 hey koen Feb 03 16:38:16 I guess I will have another go at the mipsel stuff today Feb 03 16:38:58 RP: is the sato engine from o-hand, or is there an 'upstream' somewhere? Feb 03 17:01:27 was there any further discussion on epoch handling in bitbake? Feb 03 17:01:45 not that I know off Feb 03 17:02:20 I suspect package_{rpm,ipk,deb,tar}.bbclass needs to be updated as well Feb 03 17:18:23 koen: Its from o-hand Feb 03 17:19:02 pH5: No further discussion yet, I need to try it and I have some UI stuff locally I need to try and get working first ;-) Feb 03 17:20:51 RP: sure, just didn't want to miss anything :) Feb 03 17:21:12 pH5: Did you do any further testing with it? Feb 03 17:31:33 RP: I tried those degenerate border cases (with both 1.0 and 1:1.0) with an empty test recipe, but not extensively. So far it seems to work. Also, like koen said, I have to add PE support to ipk/deb/rpm package classes. Feb 03 17:32:42 I'll post a patch that adds PE support to package_ipk.bbclass and maybe add an 1.0.1 version of libx11 with PE set to 1 as an example. Feb 03 17:32:46 pH5: and consider what to do with STAMP... Feb 03 17:33:02 mickey|walking: around? Feb 03 17:33:42 Offhand does anyone know how to signal python to wake up from a select? Feb 03 17:34:08 hm, right. your idea was to just prefix the version with epoch+underscore, right Feb 03 17:34:39 pH5: Yes, although it might be nice to drop the prefix in the PE=0 case and retain our current builds... Feb 03 17:34:48 RP: of course Feb 03 17:54:41 RP: currently the STAMP filename prefix is set to ${PF}, which is ${PN}-${PV}-${PR}, so I'd replace that ${PF} with some ${PF_WITH_EPOCH}, which will have to contain a python expression. is there a better way? Feb 03 17:58:17 pH5: I'd avoid _ in the name since it gives the code headaches with overrides Feb 03 18:00:04 RP: hmh, ${PNPEPVPR} :) is there any var besides STAMP and WORKDIR that should include the epoch information? Feb 03 18:01:49 pH5: It will need a python expression but I'd have something like PF = "${PN}${EXTENDPE}-${PV}-${PR}", EXTENDPE = "{@['','-${PE}'][bb.data.getVar('PE',d,1) == 0]}" Feb 03 18:02:25 pH5: Anything which needs it should hopefully use PF :} Feb 03 18:03:36 RP: but PF is used in many places, do we really want all places that use PF to include PE information? Feb 03 18:03:40 RP: is there any README on what sato tries to be/accomplish? Feb 03 18:03:54 it is used for FILESPATH for example, and I don't see how epoch is relevant there Feb 03 18:04:50 pH5: Sure its relavent since two packages with the same PR/PV might have two different files directories... Feb 03 18:05:28 Note I'm not convinced about using '-${PE}'. A different character might be better, I'm not sure... Feb 03 18:06:18 RP: how about ${PN}-${PE}_${PV}-${PR} Feb 03 18:07:08 koen: Not yet. Its actively being worked by o-hand people though Feb 03 18:07:14 pH5: Yes, that could work Feb 03 18:29:46 RP: EXTENDPE = "${@['','${PE}_'][bb.data.getVar('PE',d,1) > 0]}" fails if PE is not defined: Feb 03 18:30:30 I am trying to move my distro to task-base but i am getting the following error -> http://pastebin.ca/338839 Feb 03 18:30:31 SyntaxError: EOL while scanning single-quoted string (after the first PE) Feb 03 18:30:37 any idea what i am doing wrong ? Feb 03 18:30:40 RP: you can (mis)use signal(), however I suggest adding a magic Unix domain socket to the read-set in select. Feb 03 18:30:49 RP: EXTENDPE = "${@['','PE_'][bb.data.getVar('PE',d,1) > 0]}" works though Feb 03 18:32:42 seems I can't use the '}' character in embedded python code? Feb 03 18:34:19 heh, this works: Feb 03 18:34:19 EXTENDPE = "${@['','${PE\x7d_'][bb.data.getVar('PE',d,1) > 0]}" Feb 03 18:35:20 mickeyl: The specific problem is I need to wake up the XMLRPC listener to make it check self.quit and exit. Currently I've settled on overloading get_request but then I have to overload some other bits and its ugly :/ Feb 03 18:36:31 pH5: Does that not print PE_ ? Feb 03 18:36:55 * RP suspects his logic was backwards in that expression or something Feb 03 18:37:03 yuck, I pasted the wrong snippet Feb 03 18:37:09 no, wait, I didn't Feb 03 18:37:17 ${PE\x7d does the trick Feb 03 18:42:31 mickeyl: The bottom part of http://www.rpsys.net/openzaurus/temp/uievent.py is what I have at the moment, neater solutions welcome :) Feb 03 18:47:23 RP: looks good to me :) Feb 03 20:02:51 Zero_Chaos: PONG!! Feb 03 20:24:39 does anyone knows why the forums are down or when will they be back up? Feb 03 20:27:45 YoG: which forums? Feb 03 20:29:11 oesp I assume Feb 03 20:29:50 calmofthestorm7: oesp? Feb 03 20:30:13 I mean oesf Feb 03 20:30:32 yeah, that Feb 03 20:30:33 sorry;( Feb 03 20:31:11 so does anyone knows what's the problem with them? Feb 03 20:31:13 YoG: ah, that. no idea, sorry. you know that oesf and open embedded are not related, right? Feb 03 20:32:45 not related... no i didn't, I thought that oe stands for open embedded Feb 03 20:34:08 maybe I'll try my luck with openzaurus... Feb 03 20:34:19 YoG: ok, now you do :) there are zaurus people here and I bet some of them read this forum, but oesf.org and openembedded.org only share initials. Feb 03 20:34:34 well openzaurus is build with oe Feb 03 20:34:40 I'm a Z suer Feb 03 20:34:55 building lyx with oe at the moment;) Feb 03 20:35:09 rehi, all! Feb 03 20:35:50 what the purpose of module_autoload_xxx lines and how to use them? Feb 03 20:36:23 calmofthestorm7: what openzaurus flavor are you using? (GPE or OPIE) Feb 03 20:36:31 GPE Feb 03 20:36:37 OPIE's too buggy on the C1000 Feb 03 20:36:40 and I need X Feb 03 20:36:54 'cause I wnat to be able to build gaim encryption plugins Feb 03 20:36:58 which is honestlyb eing a bitch Feb 03 20:38:14 calmofthestorm7: I'm on 860 with opie but maybe I'll put on GPE when 3.5.5 will come out Feb 03 20:38:43 well opie is nice but Feb 03 20:38:48 gpe is just better imho Feb 03 20:40:03 slapin_nb: module_autoload_xxx will make the xxx module package have a pkg_postinst script that adds xxx to the /etc/modules file (list of modules to load during boot). Feb 03 20:40:29 anyway, thanx guys... I'll try asking in #openzaurus Feb 03 20:41:13 pH5, are all kernel modules packages automatic? Feb 03 20:42:15 they are created together with the kernel build. classes/kernel.bbclass takes care of that Feb 03 22:35:42 JustinP: Ping? Feb 03 22:36:29 cannot open locale definition file `/usr/share/i18n/locales/byn_ER': No such file or directory Feb 03 22:36:33 NOTE: Task failed: localedef returned an error. Feb 03 22:36:35 NOTE: package glibc-2.3.5+cvs20050627-r3: task do_package: failed Feb 03 22:36:39 sorry, didn't mean to do that Feb 03 22:38:25 but that error is annoying Feb 03 23:04:12 how are the linux libc headers related to the actual kernel I am running? should I try to match the kernel version, or are those headers only needed by glibc and not really related to the current kernel? Feb 04 00:19:47 hmm.. any ideas what could cause "NOTE: preferred version 2.5 of glibc not available" Feb 04 00:19:55 ? Feb 04 00:20:25 the .bb file is there... Feb 04 01:12:19 doh.. I can compile glibc-2.5 with bitbake -b, but if I try to do bitbake task-base then I always get that preferred version of glibc 2.5 is not available Feb 04 01:12:27 I can't figure this one out.. Feb 04 01:14:52 there's special notes abuout building task-base here Feb 04 01:14:53 http://www.openembedded.org/wiki/ZaurusKernels Feb 04 01:14:55 dunno if it helps Feb 04 01:15:02 thanks, let me see... Feb 04 01:16:00 uhm.. not really.. Feb 04 01:16:05 lol sorry Feb 04 01:16:09 :) Feb 04 01:16:12 I just remember it saying something about special changes Feb 04 01:16:16 at least you tried ) Feb 04 01:16:18 yeah Feb 04 01:17:01 I'm actually fighting with a mipsel target Feb 04 01:17:52 what's that? Feb 04 01:18:37 a mips little endian based board Feb 04 01:19:23 aah Feb 04 01:25:45 time to sleep I guess.. let's hope I solve this tomorrow :> Feb 04 01:25:50 bye Feb 04 01:25:53 lol bya **** ENDING LOGGING AT Sun Feb 04 02:59:58 2007