**** BEGIN LOGGING AT Fri Nov 27 02:59:58 2009 Nov 27 03:48:55 I have not been able to build an image for a few daays due to pixman - http://pastebin.com/ma342e4e Nov 27 03:49:04 any clues to fix? Nov 27 04:53:01 BillK_: http://cgit.openembedded.net/cgit.cgi/openembedded/commit/?h=shr/merge&id=9add11d42376c9f42981bad537f6dbb85625693a Nov 27 05:41:55 JaMa, why not disable tests in pixman? Nov 27 05:42:09 you cannot run them anyways Nov 27 06:05:59 playya_: because then I had the same problem with missing libpixman all over the tree Nov 27 06:06:11 playya_: so its better to fix it while installing pixman Nov 27 06:27:24 Hi, all! I have problems again :) Nov 27 06:27:25 I'm running perl program yarxi.pl - which is russian - japan dictianry. Nov 27 06:27:25 On my gentoo desktop I have http://pastebin.ca/1689161 but on neo I have http://pastebin.ca/1689160. Advice smth if you can :) Nov 27 06:35:53 mornin' Nov 27 06:47:17 blindcoder: morning... just added your cocktailbar to the feed :) Nov 27 06:54:30 \o/ Nov 27 06:54:41 moin Nov 27 06:55:00 mrmoku|away, today or yesterday? Nov 27 07:15:20 morning Nov 27 07:21:13 moin spaetz Nov 27 07:24:37 moin Nov 27 07:25:04 Hi, all! I have problems again :) Nov 27 07:25:04 I'm running perl program yarxi.pl - which is russian - japan dictianry. Nov 27 07:25:04 On my gentoo desktop I have http://pastebin.ca/1689161 but on neo I have http://pastebin.ca/1689160. Advice smth if you can :) Nov 27 07:27:19 parilo: no clue... Nov 27 07:37:38 spaetz: maybe some key feature is disabled in openembadded's perl Nov 27 07:41:51 mrmoku|away: is download dir on buildhost set to directory published here http://build.shr-project.org/sources/ ? Seems not and I think it should be.. as its one of shr.conf premirrors Nov 27 07:48:25 that would be a nifty idea actually... if buildhost can take the load Nov 27 07:48:36 it seems to be quite slow Nov 27 07:49:06 parilo: check "opkg list|grep perl" Nov 27 07:49:14 perl is divided in many subpackages Nov 27 07:49:29 I guess you need to install some of them. No clue which ones Nov 27 07:51:11 JaMa: I think we should download the def-configs for linux-openmoko-shr-devel and store them in OE Nov 27 07:51:19 (mickey said it's policy anyway) Nov 27 07:51:35 this way we can easily tweak our kernel config without patching all the time Nov 27 07:51:57 I think we might be able to speed up the boot process by getting rid of some uneeded modules Nov 27 07:52:14 and compiling those in that we need (eg. sound) Nov 27 07:54:19 spaetz: with that sources directory pupulated by buildhost you don't need bzr on your host.. and if its too much load for buildhost we should remove it from our PREMIRRORs conf Nov 27 07:54:41 playyaWorking_: today morning Nov 27 07:55:31 JaMa: let me check Nov 27 07:55:55 spaetz: if its policy then, its easy to do it.. but don't expect tweaking a lot easier with defconfig in OE :) Nov 27 07:56:19 JaMa: hmm... where should it point to? Nov 27 07:56:27 spaetz: at least I'm always trying to stay as close as possible to defconfig in kernel tree Nov 27 07:56:55 JaMa: to downloads? Nov 27 07:57:15 spaetz: which .config is not after -c menuconfig because some stuff is changed during do_configure based on machine setting etc Nov 27 07:57:18 mrmoku: I guess so Nov 27 07:57:35 yeah sounds sane Nov 27 07:57:37 doing that now Nov 27 07:57:46 mrmoku: ok thanks Nov 27 07:58:04 JaMa: not easier when it's in OE, that might be true Nov 27 07:58:05 * JaMa installed bzr anyway.. babiloo could be usefull Nov 27 07:58:08 ok... we all should have less 404 now :) Nov 27 07:58:18 ohh Nov 27 07:58:19 mrmoku: 3 times less :) Nov 27 07:58:20 hmm Nov 27 07:58:25 now I remember Nov 27 07:58:31 buildhost has no bzr either :| Nov 27 07:58:32 but as we have a per machine defconfig, we should be much freeer Nov 27 07:58:45 to modify it. Of course we can do it with patches as well Nov 27 07:59:12 I just think we should e.g. get rid of OSS (if it is able to break our sound) and compile sound//Wifi in, in gta02 Nov 27 07:59:19 and get rid of some modules Nov 27 07:59:23 now we have a problem... Nov 27 07:59:35 IIRC I already asked Ainulindale if it is possible to install bzr on the buildhost Nov 27 07:59:52 which (IIRC) would need some updating that is not easily possil be Nov 27 07:59:56 mrmoku: I'll send you tarball Nov 27 07:59:59 mrmoku: :0 Nov 27 08:00:03 hehe Nov 27 08:00:15 send ma bazaar-native.bb instead ;) Nov 27 08:00:28 that is missing anyway... Nov 27 08:01:02 hmm... subversion-native is missing too Nov 27 08:01:20 Ask the guys to host a tarball of the latest snapshot on launchpad Nov 27 08:01:20 mrmoku: cvs-navite too :) Nov 27 08:01:28 just git-native is there Nov 27 08:01:32 is always... git rules :P Nov 27 08:01:33 no need to fetch everything via source code repo Nov 27 08:01:38 mrmoku: because those are "bitbake soft" dependencies Nov 27 08:01:39 s/is/as/ Nov 27 08:01:40 mrmoku meant: as always... git rules :P Nov 27 08:02:07 should I send them an email, asking? Nov 27 08:02:08 JaMa: yeah, but git could be too... Nov 27 08:02:18 spaetz: bearstech or oe? Nov 27 08:02:26 mrmoku: I'm not sure if bb/fetch/git.py is using git-native from cross/stagging area anyways.. Nov 27 08:02:38 JaMa: I think so yes Nov 27 08:02:41 mrmoku: I meant sending a mail to bailoo. They can host snapshot tarballs Nov 27 08:02:50 ahh Nov 27 08:02:51 no need to fetch it via bzr Nov 27 08:02:52 :) Nov 27 08:03:02 mrmoku: do you want those tarballs? Nov 27 08:03:13 JaMa: tarballs of bhr? Nov 27 08:03:16 s/bhr/bzr/ Nov 27 08:03:16 mrmoku meant: JaMa: tarballs of bzr? Nov 27 08:03:51 mrmoku: no babiloo Nov 27 08:03:52 http://launchpad.net/babiloo/trunk/2.0.9/+download/babiloo-2.0.9.tar.gz Nov 27 08:03:57 Is that the efl version? Nov 27 08:04:04 mrmoku: http://jama.homelinux.org/org.openembedded.shr/bzr_bazaar.launchpad.net_.~vaudano.babiloo.efl_288.tar.gz Nov 27 08:04:19 JaMa: ahh... no I get you... stuff it into downloads :) Nov 27 08:04:24 and .md5 in the end.. wget them to downloads and you can build babiloo now to feed Nov 27 08:04:28 nahh, the release tarball is qt Nov 27 08:04:28 JaMa: ok Nov 27 08:04:30 not efl Nov 27 08:05:29 what is babiloo btw... ? :) Nov 27 08:05:39 some dict? Nov 27 08:05:43 dict Nov 27 08:05:49 ahh... good name then :) Nov 27 08:05:51 never used it Nov 27 08:06:01 (and explains why the tarball is sooo big :) Nov 27 08:06:35 qt frontend by default with an experimental efl frontend as I understand it Nov 27 08:06:42 but quite active dev, it seems Nov 27 08:10:49 send the babiloo-efl guy a mail asking for a src tarball Nov 27 08:10:57 s/send/sent/ Nov 27 08:10:57 spaetz meant: sent the babiloo-efl guy a mail asking for a src tarball Nov 27 08:11:41 * spaetz wonders if mickey found sth with the call while resumed issue. Nov 27 08:11:52 which reminds me that we're ignoring max_posedon's request for release tarballs for quite some time now :| Nov 27 08:11:58 we really should not Nov 27 08:12:07 release tarballs of what? Nov 27 08:12:07 max_posedon: sorry :) Nov 27 08:12:07 shr apps? Nov 27 08:12:10 spaetz: yup Nov 27 08:12:18 phonefsod, phoneuid, libphone-ui, libphone-ui-shr, ... Nov 27 08:12:21 mmh, makes sense. Distro like release tarballs Nov 27 08:12:35 Ah, this time the channel is full ;) Nov 27 08:12:59 :) Nov 27 08:13:06 Can't we just tag it as 0.0.1 and refer to http://git.shr-project.org/git/?p=phonefsod.git;a=snapshot;h=beb2fe989bfbf6b180a58d247db302082504fa01;sf=tgz Nov 27 08:13:10 Heinervdm: because we all want to have fun instead of working ;) Nov 27 08:13:15 or something like that for a source tarball Nov 27 08:13:17 mrmoku: hehe Nov 27 08:13:56 spaetz: yeah... Nov 27 08:14:04 mrmoku: no, i meant that with my first connect there were only 10 people in the channel because of netsplit :) Nov 27 08:14:19 spaetz: TAsn wanted to fix the buildsystem for phoneuid... after that Nov 27 08:14:22 Heinervdm: we are busy because we were all discussing your babiloo recipe :) Nov 27 08:14:22 Heinervdm: ahh Nov 27 08:14:32 :D Nov 27 08:14:39 man... that tarball is biiiig Nov 27 08:15:03 i think it's because there is the .bzr dir in it Nov 27 08:15:20 Heinervdm: which means the complete src history? Nov 27 08:15:26 fuck Nov 27 08:15:39 that would be stupid Nov 27 08:15:44 perhaps, i have no idea about bzr ;) Nov 27 08:18:56 Heinervdm: babiloo is imho only bzr:// recipe :) at least grep -R says so :) Nov 27 08:19:18 JaMa: java-gnome should be one too Nov 27 08:20:22 JaMa: tkx for pixman fix, running now Nov 27 08:22:15 Heinervdm: find recipes/ -name \*java-gnome\* -> *nothing* :) Nov 27 08:22:52 JaMa: yes, i don't know where it is, but i have sources from it in my downloads dir ;) Nov 27 08:22:59 :) Nov 27 08:23:02 oki Nov 27 08:23:22 or it was a recipe i tried to make Nov 27 08:25:12 mrmoku: is that the downloads dir? http://build.shr-project.org/shr-unstable/sources/ Nov 27 08:25:24 nahh, can't be Nov 27 08:28:02 without shr-unstable Nov 27 08:31:42 ahh Nov 27 08:32:27 cool. is that already used as a mirror URL? Should it? If yes, we probably need to be prepared to take quite some bandwidth hit Nov 27 08:33:25 afaik yes Nov 27 08:33:39 and do bzr recipes like babiloo need a MD5 checksum in order to download a tarball from the mirror? Nov 27 08:33:49 playyaWorking__: ahh, k Nov 27 08:34:14 spaetz, http://cgit.openembedded.org/cgit.cgi/openembedded/tree/conf/distro/shr.conf?h=shr/merge#n50 Nov 27 08:35:05 good Nov 27 08:35:45 yes: http://build.shr-project.org/sources/bzr_bazaar.launchpad.net_.~vaudano.babiloo.efl_288.tar.gz Nov 27 08:35:56 downloads now rather than complaining about bzr Nov 27 08:36:34 158MB tarball? For a dictionary app without dictionaries= Nov 27 08:36:34 ? Nov 27 08:36:34 OMG Nov 27 08:37:50 ohh... it is without dicts? puuuhh Nov 27 08:38:07 * mrmoku assumed there what be at least 10 languages in it ;) Nov 27 08:38:10 I *think* so Nov 27 08:38:28 at least the /dict folder only contained a README when I glances at it Nov 27 08:38:34 glanced Nov 27 08:39:24 dammit, nobody seems to be able to debug the incoming-call-while-suspended issue Nov 27 08:39:28 I WANT THAT FIXED Nov 27 08:40:18 and it's weird. ogsmd hasn't changed. fsodeviced doesn't seem guilty (it happens withodeviced too) Nov 27 08:41:14 kernel? Nov 27 08:41:45 ok babiloo compiles. enough to push the recipe I'd think Nov 27 08:41:51 spaetz: ??? Nov 27 08:41:58 babiloo is already in the feed :P Nov 27 08:42:02 spaetz: yesterday mickey said that he will look into it today Nov 27 08:42:10 ooh, this old man here is too slow for mrmoku Nov 27 08:42:14 spaetz, did you change apm? Nov 27 08:42:31 spaetz: the recipe has no do_compile, there is just a do_install Nov 27 08:42:54 playyaWorking__: well, as we migrated to OE.dev versions there might have been some changes. Nov 27 08:43:13 but then I uninstalled apm/apmd and reinstalls fso-apm Nov 27 08:43:38 kernel had hardly changed since sep08 image, jama told me Nov 27 08:43:51 leviathan_, had the problem that fso can't wake up the g1. it had been normal apm Nov 27 08:43:55 hmm. yes Nov 27 08:44:08 we need a big meta diff :P Nov 27 08:44:21 have fun, the overall changes are huge :) Nov 27 08:44:46 althogh looking at the kernel log, there have been quite some changes too Nov 27 08:45:51 * mrmoku is downloading some cocktails :P Nov 27 08:46:30 it might make sense to build and use the 08/08 kernel and see if it works Nov 27 08:46:39 and bisect the changes Nov 27 08:47:06 mrmoku: no drinking in the mornings! ;) Nov 27 08:47:12 :P Nov 27 08:47:31 blindcoder: isn't B52 missing the funny part to set it on fire? :P Nov 27 08:47:34 do we have a tarball of the uImage and modules from 08/08? Nov 27 08:48:09 mmh, yes Nov 27 08:48:32 spaetz: shr-obsolete Nov 27 08:48:43 yep Nov 27 08:48:46 downloading Nov 27 08:49:11 yes, that is an interesting test Nov 27 08:49:18 brb Nov 27 08:54:45 mrmoku: true, I should implement editing on the website :) Nov 27 08:54:57 the database supports history already, only the frontend doesn't Nov 27 08:58:18 mmh, damn now I've overwritten my old /lib/modules/2.6.29-rc3 Nov 27 09:02:40 has sysfs changed since Sep02? I don't get charging LEDs on the old kernel Nov 27 09:03:02 but the old kernel is fast too! Nov 27 09:03:03 hmm, don't think so Nov 27 09:03:59 spaetz: if you unplug/replug? Nov 27 09:04:03 nope Nov 27 09:04:06 tried already Nov 27 09:04:17 ahh now it works Nov 27 09:04:25 just needed some more time to settle :) Nov 27 09:04:59 damn, old kernel from Sep02 nearly feels faster than what we have now.. Nov 27 09:05:38 let me try to call myself Nov 27 09:09:37 Heinervdm: libnl 1/2 collision should be fixed now? Nov 27 09:09:51 Heinervdm: seems like not completelly Nov 27 09:09:52 JaMa: mickey fixed that Nov 27 09:10:27 he changed includedir of libnl2 to ${includedir}/netlink2 Nov 27 09:10:35 interesting Nov 27 09:10:38 if there are other issues tell him :) Nov 27 09:10:47 Heinervdm: if I build libnl2 first and then libnl1 then libnl1 tries to link against something from libnl2 i guess Nov 27 09:11:03 1) while the thing is on, I can call myself just fine. Nov 27 09:11:26 Heinervdm: libnl-1.1/src/nl-addr-add.c:33: undefined reference to `rtnl_addr_alloc' Nov 27 09:11:27 2) When suspended I get a long ringtone which doesn't stop after a second. BUT... Nov 27 09:11:30 mickey|zzZZzz: ^^^ Nov 27 09:11:38 3) I get no Active call screen Nov 27 09:11:51 there is just no way to accept the call Nov 27 09:12:42 mrmoku: can it be that libphone* phonefsod do weird stuff? Nov 27 09:12:53 or that a signal is missing that should be send? Nov 27 09:13:07 spaetz: hmm Nov 27 09:13:11 logs? Nov 27 09:15:11 spaetz, could you write a log like: mdbus -s -l 2>&1 > ~/foobar.log & ? Nov 27 09:16:48 no logs :) Nov 27 09:16:59 will take an mdbus log Nov 27 09:17:12 no logs? Nov 27 09:17:14 anything else? phoneuid phonefsod logs? Nov 27 09:17:18 /var/log/phoneuid? Nov 27 09:17:20 .log Nov 27 09:17:21 no logs taken so far. Nov 27 09:17:26 ahh, ok. hold on Nov 27 09:19:43 http://sspaeth.de/uploads/tmp/logs.tar Nov 27 09:19:58 they contain my home phone number. please don't start calling me :) Nov 27 09:20:12 *sigh* Nov 27 09:20:14 ERROR: '/home/w/OM/SHR/build/shr-unstable/openembedded/recipes/bluez/bluez4_4.56.bb' failed Nov 27 09:20:14 actually they might contain my whole address book. Nov 27 09:20:24 please download and I'll kill the file off again Nov 27 09:20:40 plugins/netlink.c:91: error: too few arguments to function 'genl_ctrl_alloc_cache' Nov 27 09:20:45 heh :) Nov 27 09:20:45 Wonka: -c clean both libnl recipes Nov 27 09:21:00 spaetz: got it Nov 27 09:21:11 ok, deleted it from host Nov 27 09:21:23 Wonka: then -c build libnl_*bb then -c build libnl2_ and then -c build bluez4 should work.. Nov 27 09:21:38 Wonka: at least it worked for me before 5 mins Nov 27 09:21:46 JaMa: sorry? "bitbake -c clean" I guess, but where Nov 27 09:22:01 Wonka: yes Nov 27 09:22:02 let me check when I suspended mrmoku Nov 27 09:22:34 JaMa: in which directories? or, alternatively, with what other parameters? Nov 27 09:22:34 spaetz: (r.e. SHR-Devel just now) - don't we already know what made it crawl? Nov 27 09:22:54 JaMa: or just clean everything? Nov 27 09:22:56 Weiss: what made it crawl? Nov 27 09:23:29 we have the glamo tweak and the accelerator interrupts patch that made it faster Nov 27 09:23:31 Wonka: http://pastebin.ca/1689292 Nov 27 09:23:41 but those things had not changed since sep 02, right? Nov 27 09:23:48 ah.. hmm.. Nov 27 09:23:52 JaMa: thx Nov 27 09:23:52 spaetz: 2009.11.27 10:07:23.347207 [phoneuid] org.shr.phoneuid.CallManagement.DisplayIncoming Nov 27 09:24:01 2009.11.27 10:07:28.259362 [phoneuid] org.shr.phoneuid.CallManagement.HideIncoming Nov 27 09:24:11 2009.11.27 10:07:21.966 frameworkd.resource INFO setting resource status for GSM from suspended to resuming Nov 27 09:24:20 spaetz: so you think there might be more speed to get still? :) Nov 27 09:24:20 this is when it woke up as I called myself Nov 27 09:24:22 spaetz: you can try to build the same revision from current shr-u to see if the difference is not because of our newer toolchain Nov 27 09:24:34 Weiss: this is what I am wondering about Nov 27 09:24:43 JaMa: sorry not today. Nov 27 09:24:58 or a change of config perhaps? Nov 27 09:25:13 Weiss: it might be a regression in the kernel, or it might be due to our updated toolchain Nov 27 09:25:24 (as JaMa pointed out. I totally forgot about that) Nov 27 09:25:38 forgot to turn on -O2 or something? :) Nov 27 09:25:54 hehe Nov 27 09:26:26 mrmoku: interesting. the log might contain more calls Nov 27 09:26:35 spaetz: yeah, it has more calls Nov 27 09:26:44 I tested about 4 times. some while online, some while suspended Nov 27 09:26:58 3 calls I see Nov 27 09:27:11 and when online I was able to press the "Accept" button, so that one was ok Nov 27 09:27:12 spaetz: yeah.. me too not today.. I just wanted to make point that bisect should start with same revision :) Nov 27 09:27:30 both times when it was resumed I never saw the Active call win Nov 27 09:27:52 hmm Nov 27 09:27:56 although I had to fiddle with shr-today to make it go away once Nov 27 09:28:07 * JaMa|Wrk needs to notify himself that there is still that daywork waiting :/ Nov 27 09:28:18 JaMa|Wrk: agreeed Nov 27 09:28:29 bisect should start with that revision Nov 27 09:29:09 I am not a kernel guy at all though... Nov 27 09:29:28 * spaetz thinks we need to hire someone for the SHR kernel department :) Nov 27 09:29:46 you definitely don't need to be a kernel person to bisect :) Nov 27 09:29:59 but lots of time on your hands :) Nov 27 09:30:10 * Weiss finds bisecting can be strangely relaxing if I have nothing else to do Nov 27 09:30:16 which is a rather scarce resource. Nov 27 09:30:54 I have a 2nd kid coming in about 15 days or so, so I have to do lots of extra-stuff now Nov 27 09:30:54 ah there is only 14 commits after that revision Nov 27 09:31:59 JaMa|Wrk: actually compiling the exact same rev withour current toolchain would be an interesting test Nov 27 09:33:45 spaetz: I see one call where you pressed release Nov 27 09:34:51 spaetz: I just checked commit messages.. and if someone tests speed without touchscreen then there is only very few commits worth trying to bisect Nov 27 09:35:14 spaetz: jo first new toolchain, then disable touchscreen, then maybe 3 steps Nov 27 09:35:52 but still lots of time to rebuild it everytime and test it on neo :/ Nov 27 09:36:13 hi guys! again a new image yesterday! :p cool! what is new (since last we)? Nov 27 09:37:06 GarthPS: check git commits in shr/merge :) Nov 27 09:37:19 GarthPS: I guess nobody is keeping track in summary changelog Nov 27 09:37:29 hooh we have that? where? Nov 27 09:37:54 GarthPS: git log or changelog? :) Nov 27 09:38:08 JaMa|Wrk: yep Nov 27 09:38:08 GarthPS: opkg upgrade is good log imho :) Nov 27 09:38:12 GarthPS: http://git.shr-project.org/git/ and http://cgit.openembedded.org/cgit.cgi/openembedded/log/?h=shr/merge Nov 27 09:38:18 THX Nov 27 09:38:19 possibly relevant: I don't think I have the slowdown with my kernel builds.. (I use an old toolchain, and drm-tracking is currently up to date with andy-tracking) Nov 27 09:38:42 toolchain might be a suspect then... Nov 27 09:39:14 we need some sort of benchmark Nov 27 09:39:23 and try it all Nov 27 09:39:50 for me its still the "same" Nov 27 09:40:11 JaMa|Wrk: well bootchart-lite time till enlightenment start could be a possibility. or the x11perf tool Nov 27 09:40:15 sometimes its fast and nice.. and sometimes on same config it feels slow as hell Nov 27 09:42:42 spaetz: ok, I'll try bootchart here.. Nov 27 09:43:01 just to see if its so "slow" as in your image Nov 27 09:43:27 just compile bootchart-lite (it's a recipe) Nov 27 09:43:54 and install on the FR. It wants "install" on the FR. I just made a dummy binary. All it does is create "/etc/bootchart-lite" Nov 27 09:43:57 yup.. Nov 27 09:44:32 append-GTA02:init=/usr/bin/bootchart-lite logleve=.... Nov 27 09:45:23 tar all logs in /etc/bootchart-lite and feed that tar to "bootchart" on your host Nov 27 09:45:37 bootchart -f png logs.tar (or similar) Nov 27 09:45:39 that's it Nov 27 09:45:59 oki.. will do.. now compiling and back to work, thanks Nov 27 09:46:06 postinst for the bootchart-lite should probably be using "mkdir" rather than "install" Nov 27 09:46:16 not sure if all devices will have that installed Nov 27 09:46:30 i have coreutils so i have install Nov 27 09:46:37 ahh, k Nov 27 09:46:41 but you're right.. Nov 27 09:46:51 then it should either RDEPENDS on coreutils Nov 27 09:46:55 or use mkdir Nov 27 09:47:27 ah no :) mkdir would be better Nov 27 09:47:41 not all likes coreutils as me and docs :) Nov 27 09:49:56 I just send a mail to openembedded-dev to make all happy :) Nov 27 09:50:06 but I tend to use mkdir Nov 27 09:50:22 especially for a "-lite" package :) Nov 27 09:51:05 mrmoku, here? Nov 27 09:51:08 yup Nov 27 09:52:51 saw my commits? :P Nov 27 09:53:02 (from last night) Nov 27 09:53:13 many bug fixes (segs) Nov 27 09:53:21 but more importantly Nov 27 09:53:26 sliders now work completely Nov 27 09:53:35 so if you can, please drop the speaker/handset button Nov 27 09:55:19 mrmoku, anyhow, I'm fixing phoneuid, so be ready to commit the recipe. Nov 27 09:55:29 TAsn: go ahead Nov 27 09:55:32 (I don't have time for phonefsod though) Nov 27 09:55:39 but it should be easy to reproduce. Nov 27 09:55:43 mrmoku: babiloo says: I will post the source tarball on Lauchpad. Nov 27 09:55:45 TAsn: I will do phonefsod then Nov 27 09:55:53 and tag 0.0.1 after that :) Nov 27 09:56:02 spaetz: good :) Nov 27 09:56:14 mrmoku, with speaker button and toggle that do the same on the same screen? :( Nov 27 09:56:25 :P Nov 27 09:56:26 TAsn: first phonefsod and phoneuid Nov 27 09:57:20 so that max_posedon can build ebuilds for those :) (well at lest for phonefsod... as phoneuid needs the other ones) Nov 27 09:57:54 spaetz: I did not quite get the result of you call while resumed with old kernel test? Nov 27 09:57:58 did it work? Nov 27 09:58:22 mrmoku: To cut a long story short: It failed differently :) Nov 27 09:58:29 hehe Nov 27 09:58:34 The ringing never stopped, the full ringtone continued playing Nov 27 09:58:43 spaetz: rebooting :) lets see if it boots (as neo is at home again..) Nov 27 09:58:49 but I never got the Active Call dialog where I could accept the phone Nov 27 09:59:08 but just one incoming call? Nov 27 09:59:32 Same results both times on incoming call when suspended Nov 27 09:59:38 it worked when it was awake Nov 27 09:59:57 let me know if I should retest something. I still have it booted with the old kernel Nov 27 10:00:04 hmm... hopefully mickey|zzZZzz will find something :| Nov 27 10:00:16 spaetz: well... no idea Nov 27 10:00:25 neither have I.... Nov 27 10:00:36 not kernel... not ogsmd, not fsodeviced... Nov 27 10:00:42 I am running out of ideas... Nov 27 10:00:58 HAH, phone* libphone* are the only suspects left! Nov 27 10:01:07 new toolchain is too fast for calypso :P Nov 27 10:01:32 maybe TAsn broke it ;) Nov 27 10:01:37 what? Nov 27 10:01:38 * mrmoku hides and whistles Nov 27 10:01:47 broke what? Nov 27 10:01:54 incoming call while resumed Nov 27 10:01:56 :P Nov 27 10:01:57 no Nov 27 10:01:59 :) Nov 27 10:02:05 not me. Nov 27 10:02:13 mickey|zzZZzz, already said it's something in modem or something Nov 27 10:03:32 TAsn: old Sep02 kernel is fast on FR too Nov 27 10:03:46 what's the diff? Nov 27 10:03:50 kind of fast. Nov 27 10:03:59 TAsn: 14 commits and a different toolchain Nov 27 10:04:24 I think someone has forgotten -Obloat somewhere Nov 27 10:05:12 did it start before or after the switch to shr/merge Nov 27 10:05:59 maybe it's a broken build with some rev not staged correctly... and nobody noticing Nov 27 10:05:59 no clue Nov 27 10:06:16 might well be. Nov 27 10:06:23 spaetz: first kernel http://jama.homelinux.org/org.openembedded.shr/bootchart-2.6.31-drm.png Nov 27 10:07:07 jama cheater! there is no udev delay Nov 27 10:07:23 actually, that might be because there are hardly any modules to go through Nov 27 10:08:47 I don't see no udev at all... Nov 27 10:08:50 which confirms my suspivion that many modules might delay boot time significantly Nov 27 10:09:04 mrmoku: right :) Nov 27 10:09:45 spaetz, duh :P Nov 27 10:09:51 we should compile all the musts into the kernel... Nov 27 10:10:05 (for gta02 we know the musts) Nov 27 10:10:10 we should do it for every device Nov 27 10:11:46 TAsn: I know and I agree Nov 27 10:11:52 cool ;) Nov 27 10:12:08 no need to keep eg sound modules separate as we know they will be needed Nov 27 10:12:19 and the config files are machine specific... Nov 27 10:13:03 But I am hesitant to touch linux recipes and change stuff Nov 27 10:13:29 I will get shouted at from Koen about breaking every distro but SHR again Nov 27 10:15:59 Kernel command line: loglevel=4 console=tty0 console=ttySAC2,115200 init=/sbin/init ro mtdparts=physmap-flash:-(nor);neo1973-nand:0x00040000(qi),0x00040000(depr-ub-env),0x00800000(kernel),0x000a0000(depr),0x00040000(identity-ext2),0x0f6a0000(rootfs) root=/dev/mmcblk0p1 rootdelay=1 init=/usr/bin/bootchart-lite loglevel=1 quiet splash Nov 27 10:16:09 did something changed in the trac? today the font size is bigger then yesterday... Nov 27 10:16:29 Heinervdm: menu->view->zoom ;-P ? Nov 27 10:16:50 spaetz: i already did CTRL+0 Nov 27 10:17:07 looks normal to me Nov 27 10:17:19 hmm Nov 27 10:17:40 actually it might be bigger you could be right Nov 27 10:18:31 can be an illusion too Nov 27 10:18:45 so back to openttd Nov 27 10:22:23 spaetz: are you sure that bootchart is used as init? files are generated only if i call bootchart after boot :/ Nov 27 10:24:29 mrmoku, sent shr-devel an email and committed phoneuid Nov 27 10:24:30 SHR: 03tom 07phoneuid * re4d827de1851 10/data/ (16 files in 4 dirs): Fixed the autotools configuration to include all the files Nov 27 10:25:09 TAsn: ok Nov 27 10:25:56 cool, thanks :) Nov 27 10:26:02 * TAsn is going to get food. Nov 27 10:26:05 JaMa|Wrk: I do think writing is delayed due to some fs caching Nov 27 10:26:23 TAsn: did you test it? Nov 27 10:26:25 but I am not sure how it works. But it does seem to have produced sth sensible... Nov 27 10:28:45 mrmoku, yes. Nov 27 10:28:57 (only tseted he package is built and stuff go the right way0 Nov 27 10:29:23 mrmoku, btw, our gsmhandset is really crappy Nov 27 10:29:34 even gsmspeaker has less noise Nov 27 10:29:38 and I'm not joking. Nov 27 10:30:10 where are our FULL_OPTIMIZATIONS set? Nov 27 10:30:38 bitbake.conf uses -O2 and sane-toolchain-eglibc.inc uses -Os Nov 27 10:30:58 TAsn: heh, I know :| Nov 27 10:31:11 but Nov 27 10:31:40 dropping Mono Sidetone Volume (or something like that) Nov 27 10:31:41 to 0 Nov 27 10:31:43 instead of 100 Nov 27 10:31:47 fixed it for me. Nov 27 10:31:54 (from what I was able to test) Nov 27 10:31:57 please give it a try. Nov 27 10:32:50 numid=12,iface=MIXER,name='Mono Sidetone Playback Volume' Nov 27 10:32:54 is what I meant. Nov 27 10:33:21 12:'Mono Sidetone Playback Volume':1:7 -> 12:'Mono Sidetone Playback Volume':1:0 Nov 27 10:33:52 and in speaker (where I hear better) Nov 27 10:33:56 it's 6 Nov 27 10:33:57 so maybe 6 is enough Nov 27 10:34:08 DocScrutinizer-8, what does this mixer represent? Nov 27 10:35:06 mmh, we use: FULL_OPTIMIZATION="-fexpensive-optimizations -fomit-frame-pointer -frename-registers -Os" Nov 27 10:35:40 anyhow, I'm off Nov 27 10:35:50 though this is probably the main culprit. :) Nov 27 10:36:33 mrmoku, if you can, please remove the speaker/handset button from active call (as we already have the toggle) and maybe make some space between the mute and the speaker toggle button. Nov 27 10:36:47 ciao. Nov 27 10:37:26 btw MicSidetone playback volume seems to be sound from other side of gsm :P Nov 27 10:37:31 so don't ruin it just yet. Nov 27 10:37:41 hehe Nov 27 10:38:26 no Nov 27 10:38:28 still works Nov 27 10:38:32 even when turning that to 0 Nov 27 10:38:37 just confirm with DocScrutinizer Nov 27 10:38:41 but it's what broke it here Nov 27 10:38:43 and phone still worksl Nov 27 10:39:09 mrmoku: b52 is now on fire :-) Nov 27 10:39:25 :) Nov 27 10:41:00 Heinervdm: what is with your apm patch in trackwork? to apply? Nov 27 10:42:18 with that we will build image with fso-apm and mickey said that it won't conflict with others Nov 27 10:43:27 openttd: http://patchwork.dev.bearstech.com/patch/545/ Nov 27 10:45:50 applied Nov 27 10:48:19 next one: scummvm Nov 27 10:49:43 * spaetz suspects heiner wants to turn the FR into a gaming machine Nov 27 10:50:13 http://jama.homelinux.org/org.openembedded.shr/bootchart-2.6.29-oe11+gitr4+8c65792a5c83c76d662a617a7c4e1ae8104bb6a5-r5-om-gta026.31-drm.png Nov 27 10:50:17 spaetz: no that was TAsn with that ticket: http://www.shr-project.org/trac/ticket/720 ;) Nov 27 10:51:51 spaetz: I think its not used as init.. thats why it shows load and disk so late for me.. Nov 27 10:52:03 spaetz: after I run bootchart from cmdline Nov 27 10:52:19 spaetz: and then it counts time for stuff from ps maybe.. Nov 27 10:53:29 possible Nov 27 10:53:51 your 2.6.31 has batget starting 60 sec earlier than the 2.6.29 one :) Nov 27 11:12:03 TAsn: is that really needed? +FILES_${PN} += "${datadir}" Nov 27 11:12:10 is datadir not included by default? Nov 27 11:14:39 spaetz, yeah, for some reason it's not ;( Nov 27 11:14:46 no idea why. Nov 27 11:16:25 ${datadir}/${PN} is included Nov 27 11:16:35 i c. Nov 27 11:16:38 so this explains it. ;) Nov 27 11:16:53 so yeah, we should add $(datadir) Nov 27 11:17:57 ticket 720 looks quite reasonable :) Nov 27 11:18:05 ahh, +FILES_${PN} += "${datadir}" Nov 27 11:18:22 is in the default Nov 27 11:18:24 or should be Nov 27 11:18:27 anyway. lunch Nov 27 11:18:38 TAsn: don't you think you should be able to find out yourself, instead of everybody testing random mods in mixer without any idea or plan? :-((( Nov 27 11:18:56 DocScrutinizer-8, I read the wiki page. Nov 27 11:19:07 it seems to be like a volume mixer between gsm and speaker Nov 27 11:19:10 I'm not sure though Nov 27 11:19:24 and I don't have time to verify. Nov 27 11:19:36 I'm not letting anyone doing crazy random things Nov 27 11:20:01 With knowing this control gave me trouble in the past Nov 27 11:20:10 I suggested it'll be turned down a bit. Nov 27 11:20:15 DocScrutinizer-8, ATM the states suck Nov 27 11:20:19 there's no doubt about that. Nov 27 11:20:27 In my pov we should just switch back to the old ones Nov 27 11:20:37 but, someone said the new ones are "more correct" Nov 27 11:20:48 so I'm willing to give them a chance by doing stuff I know that work. Nov 27 11:20:55 but still, that's not random. Nov 27 11:21:10 if you want, you can help by suggesting non random improvements Nov 27 11:21:20 that will maybe, just maybe make the sound a bit more like before. Nov 27 11:24:01 sorry, I'm not interested in this senseless new alsa foo. won't test it, won't debug it, won't try to understand what went wrong, but obviously sth did Nov 27 11:24:41 DocScrutinizer-8, if you can point me to correct settings Nov 27 11:24:43 that'd be great Nov 27 11:24:51 I'll just write a script to convert. Nov 27 11:25:04 see wiki Nov 27 11:25:25 thanks. Nov 27 11:25:28 they are up to date? Nov 27 11:28:04 maybe one advice missing in wiki (as it's so obvious): all non mentioned controls *need* to be off/0 Nov 27 11:37:00 playyaWorking__: _now_ is kindergarten time :P Nov 27 11:37:04 bbiab Nov 27 11:53:13 dos1: hi Nov 27 11:54:06 dos1, I'm just reminding, if you can, add country to shr-wizard. We'll probably make in image soon (at the moment answer on resume is fixed) and I really don't want to forget having this in image. :) Nov 27 11:55:58 dos1: and another reminder... before doing that... add some signals :P Nov 27 11:56:02 :) Nov 27 11:56:15 does any of you know how I can contact "Fate", the maintainer/author of mokonnect? Nov 27 11:56:33 sybren: ping F4t :) Nov 27 11:56:39 I tried to add a reply to http://www.assembla.com/flows/show/diGhnW2oir3R8teJe5aVNr, but I get a "Not Permitted" message Nov 27 11:56:43 great Nov 27 11:56:50 F4t: ping ;-) Nov 27 11:57:22 mickey|zzZZzz: ping, could you look at patch on http://trac.freesmartphone.org/ticket/483 ? Nov 27 11:58:49 TAsn, mrmoku: sorry, i didn't have any boring lession in school, as i returned to home earlier - we had only 4 persons in class, rest is ill :P Nov 27 11:59:00 ohh Nov 27 11:59:10 but then you had lots of time anyway :P Nov 27 11:59:19 dos1: ping Nov 27 11:59:26 Abaddon_: pong Nov 27 11:59:30 mrmoku: yup ;) Nov 27 11:59:35 mrmoku, so before freeze (0.1) we need dos1 to finish signals ;P Nov 27 11:59:48 dos1: I've ben trying out the latest images Nov 27 12:00:17 dos1: and I found you had now migrated contacts and messages to local storage Nov 27 12:00:22 SQL, right ? Nov 27 12:00:25 Abaddon_: yup, to opimd Nov 27 12:00:33 TAsn: well, we should just tag 0.0.1 what we have more or less now Nov 27 12:00:33 ok Nov 27 12:00:38 with SQLite-Messages and SQLite-Contacts backends by default Nov 27 12:00:39 then here comes my question Nov 27 12:00:49 TAsn: we won't loose anything... but it's a big win for max_posedon :) Nov 27 12:00:53 mrmoku, after you fix phonefsod autotools Nov 27 12:00:56 yup Nov 27 12:00:58 :) Nov 27 12:01:00 how can I access those dbs ? Nov 27 12:01:01 okie. ;) Nov 27 12:01:05 will do that now :) Nov 27 12:01:08 mrmoku, cool. :) Nov 27 12:01:11 btw, I emailed you. Nov 27 12:01:15 Abaddon_: with opimd dbus interface Nov 27 12:01:27 Abaddon_: http://docs.freesmartphone.org/ Nov 27 12:01:39 thx, I'll have a look at that Nov 27 12:02:11 Abaddon_: you can also look at opimd-cli from opimd-utils-cli package Nov 27 12:02:41 thx again Nov 27 12:03:18 TAsn: hmm Nov 27 12:03:33 TAsn: looking at your phoneuid commit I'm not sure ... Nov 27 12:03:56 TAsn: will give your state file a try Nov 27 12:04:12 TAsn: I have a problem with phoneui.conf + phoneuid.conf Nov 27 12:04:20 mrmoku, what problem? Nov 27 12:04:29 I just renamed it Nov 27 12:04:33 phoneuid.conf is (or better was :P) the dbus config Nov 27 12:04:38 mrmoku, no Nov 27 12:04:40 look in Nov 27 12:04:45 dbus-1/system.d/ Nov 27 12:04:46 ... Nov 27 12:04:49 (in the data dir) Nov 27 12:05:15 wait a sec Nov 27 12:05:15 ahh Nov 27 12:05:19 ok Nov 27 12:05:19 I'll send you a patch for phonefsod as well. Nov 27 12:05:22 I'm half way there anyway :P Nov 27 12:05:32 (started doing it earlier and figured I didn't have time) Nov 27 12:05:33 just got confused by the git diff... did not see the rename part :) Nov 27 12:05:44 TAsn: ok :) Nov 27 12:05:56 but Nov 27 12:06:00 please remove the Nov 27 12:06:02 "speaker" Nov 27 12:06:04 button Nov 27 12:06:05 from active call Nov 27 12:06:14 and add spacing between the two toggles :) Nov 27 12:06:36 alternatively just scale those 2 toggles by 1.5 :-) Nov 27 12:06:38 TAsn: extend the two other buttons then? Nov 27 12:06:41 this is finger friendly Nov 27 12:06:50 mrmoku, probably. Nov 27 12:07:06 and maybe even replace the "Release" with a nice icon (but that's asking too much ;p ) Nov 27 12:07:25 TAsn: our marketing dep. says this is 0.0.2 material :) Nov 27 12:07:32 TAsn: two steps... remove the button and spacing now... and redesign later :P Nov 27 12:07:36 :P Nov 27 12:07:37 mrmoku, cool. Nov 27 12:07:51 BTW, dialer on some contacts shows weird info, got from PIM. dos1's opimd utils shows everything ok on the same contacts. Nov 27 12:08:06 Q-Master: i have opposite situation here Nov 27 12:08:07 Q-Master: like what? Nov 27 12:08:24 libphone-ui-shr shows everything correctly, and opimd-utils fail to resolve names Nov 27 12:08:26 :D Nov 27 12:08:45 dos1, i can confirm this regardles of used backend Nov 27 12:08:58 JesusMcCloud: backend doesn't matter Nov 27 12:09:07 mrmoku: istead of phone numbers it shows some weird symbols... but I have a complex field names like home cell phone etc... Nov 27 12:09:47 mrmoku: also it doesn't show all the fields. sometimes loses some phone numbers Nov 27 12:10:57 dos1: and name resolving doesn't work for me. I suppose that that was an old non-fixed problem with russian letters. Nov 27 12:11:19 dos1: like in searching in opimd-contacts Nov 27 12:11:21 soemting completely ot: where can i get an omap3 toucbook? Nov 27 12:11:21 Q-Master: i suppose i just fscked resolving up in opimd-utils ;P Nov 27 12:12:01 dos1: not sure. but resolving works with phonelog perfect. Nov 27 12:12:12 Q-Master: so? phonelog isn't part of opimd-utils ;P Nov 27 12:12:29 and i'm not using russian letters and it also doesn't work to me ;D Nov 27 12:13:05 dos1: you should fix it. 8) Nov 27 12:13:15 yup. i want to fix it today ;P Nov 27 12:13:30 also make that stupid things for mrmoku and TAsn, as they are getting annoying! Nov 27 12:13:32 ;D Nov 27 12:13:40 :) Nov 27 12:13:41 8) Nov 27 12:13:56 dos1: did you start the signals? :P Nov 27 12:14:18 and also i want to do little fix to geekogotchi, but that's after everything else :P Nov 27 12:14:21 mrmoku: yup Nov 27 12:14:24 :D Nov 27 12:15:54 <|Fate|> hey all, how do i attach to a previous SSH session? Nov 27 12:16:25 mrmoku: what's wrong is now with make in repo? Nov 27 12:16:26 |Fate|: usr screen Nov 27 12:16:38 |Fate|: *use Nov 27 12:16:59 <|Fate|> ricky-ticky, how? Nov 27 12:18:47 Q-Master: what do you mean? Nov 27 12:19:10 mrmoku: repository doesn't have it Nov 27 12:19:34 Q-Master: which one? Nov 27 12:19:46 unstable Nov 27 12:19:53 |Fate|: ssh to remote pc, launch screen, working, When done type ctrl-a-d and logout. nex time use screen -x and you will attach to previos session.And man screen for other possibilities Nov 27 12:20:15 <|Fate|> oh, ok Nov 27 12:20:25 Q-Master: you mean the OE branch shr-unstable? Nov 27 12:20:59 mrmoku: also we need to patch the fbreader. it is almost useless. listing of pages by the finger is a pain in the ass. Nov 27 12:21:38 mrmoku: this one http://build.shr-project.org/shr-unstable/ Nov 27 12:22:10 Q-Master: did it ever have a Makefile? our Makefile is at http://build.shr-project.org/Makefile Nov 27 12:22:28 Q-Master: ahh, think I understand Nov 27 12:22:39 before it was linked to the build area... and thus you saw the Makefile Nov 27 12:22:42 mrmoku: i meant the MAKE command, not Make file. 8) Nov 27 12:22:43 now we rsync Nov 27 12:22:48 oh Nov 27 12:23:12 Q-Master: you mean make package for FR? Nov 27 12:23:17 yep Nov 27 12:23:20 ok Nov 27 12:23:40 mrmoku, sent email and committed. Nov 27 12:23:46 SHR: 03tom 07phonefsod * rcecbf058232f 10/ (8 files in 4 dirs): Fixed the autotools configuration to ship all files Nov 27 12:23:52 SHR: 03tom 07phonefsod * r84470e8b5a08 10/data/etc-phonefsod.conf: Fixed a typo in config file Nov 27 12:24:08 TAsn: thanks Nov 27 12:24:25 I just wanted to rebuild fbreader with patch to make it useable and also to check it works or not Nov 27 12:24:49 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * r09aaa3685059 10framework/framework/subsystems/opimd/pimd_generic.py: opimd: fix error when using Python 2.4 Nov 27 12:24:49 np. Nov 27 12:24:50 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * r9806711154ee 10framework/framework/subsystems/opimd/pimd_contacts.py: opimd: Contacts: implement DeletedContact and UpdatedContact signals Nov 27 12:25:00 dos1, YAY :P Nov 27 12:25:03 mrmoku, ^ :) Nov 27 12:25:04 mrmoku: ^^^, now copy'n'pasting to other domains :P Nov 27 12:25:06 TAsn: btw. could you please prefix OE patches with the package? :) Nov 27 12:25:14 like 'phonefsod:' Nov 27 12:25:16 mrmoku, ffs yeah :) Nov 27 12:25:33 and if possible... produces patches against org.oe.dev in the future :) Nov 27 12:25:43 s/produces/produce/ Nov 27 12:25:43 mrmoku meant: and if possible... produce patches against org.oe.dev in the future :) Nov 27 12:26:17 mrmoku, bah, no idea how :P Nov 27 12:26:23 but ok, next time I'll just ask. Nov 27 12:26:39 git is still a mystery to me :P I'm a long time svn user Nov 27 12:27:58 TAsn: not important... but patches with SRCPV will conflict in oe.dev... and git am does not apply them then Nov 27 12:28:03 have to do it manually Nov 27 12:28:08 no problem though Nov 27 12:29:15 * mrmoku lunch first Nov 27 12:29:19 bbiab Nov 27 12:29:52 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * r4cb08815f4bb 10framework/framework/subsystems/opimd/pimd_contacts.py: opimd: Contacts: fix dbus interfaces in new signals Nov 27 12:29:54 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * rc2e27dc638cf 10framework/framework/subsystems/opimd/pimd_messages.py: opimd: Messages: implement DeletedMessage and UpdatedMessage signals Nov 27 12:32:53 mrmoku, soory :) Nov 27 12:33:47 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * ra1859d741d03 10framework/framework/subsystems/opimd/pimd_calls.py: opimd: Calls: implement DeletedCall and UpdatedCall signals Nov 27 12:35:03 :) Nov 27 12:35:07 dos1, thanks a lot. Nov 27 12:35:53 mrmoku, don't forget to remove manual deletion Nov 27 12:36:01 from messages and contacts Nov 27 12:37:26 if exist. Nov 27 12:37:29 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * r2e70bcbc73df 10framework/framework/subsystems/opimd/pimd_dates.py: opimd: Dates: implement UpdatedDate and DeletedDate Nov 27 12:37:56 do we want 9 seconds ogsmd startup improvement? Nov 27 12:38:33 spaetz: i can apply it to FSO git if only someone acks it Nov 27 12:38:44 spaetz, with that patch? Nov 27 12:38:58 probably Nov 27 12:39:06 I don't have a patch Nov 27 12:39:10 it's a oneliner Nov 27 12:39:17 J Nov 27 12:39:19 oops Nov 27 12:39:21 http://trac.freesmartphone.org/ticket/483#comment:2 Nov 27 12:40:17 yeah, saw it. (the ticket) Nov 27 12:40:19 It's stupid to explicitely decode a file to UTF-8 that we have put there ourselves Nov 27 12:40:35 spaetz, why? Nov 27 12:40:37 just removing decode("UTF-8") saves us 9 seconds Nov 27 12:40:47 Because it saves us 9 seconds. Nov 27 12:40:49 what about utf8 Nov 27 12:41:02 I mean Nov 27 12:41:02 We don't check the correctness of our .config files either Nov 27 12:41:17 we need utf8 support in config as well :P Nov 27 12:41:19 anyhow Nov 27 12:41:20 ciao. Nov 27 12:41:22 or we don't check that nobody has fudged around in shared libs we load Nov 27 12:42:09 playyaWorking__: hi Nov 27 12:42:11 gnutoo: hi Nov 27 12:42:26 leviathan_, hi Nov 27 12:42:39 the following thing: I'll post some patches of what I've finally done to get it work Nov 27 12:42:58 would be nice if you could apply it and make some patches out of it :D Nov 27 12:42:59 leviathan_, did you see my yesterday patch? Nov 27 12:43:04 yes Nov 27 12:43:29 ok,what do you think of it? Nov 27 12:44:35 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * r81b3d2c33986 10framework/framework/subsystems/opimd/pimd_notes.py: opimd: Notes: implement DeletedNote and UpdatedNote signals Nov 27 12:47:09 leviathan_, gnutoo you could fork OE somewhere, add your stuff and send a pull request Nov 27 12:47:52 spaetz: I'd happily discard networks.tab completely Nov 27 12:48:02 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * r95c6e7afcb9a 10framework/framework/subsystems/opimd/pimd_tasks.py: opimd: Tasks: fix references to Notes domain remaining there after copy'n'paste Nov 27 12:50:15 spaetz: eats up huge amounts of flash space, obviously is a cpu hog (even with patch), and for me it's worth nothing as all networks I ever tested supply their own clear text name info anyway Nov 27 12:51:00 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * r9e8410dd71db 10framework/framework/subsystems/opimd/pimd_tasks.py: opimd: Tasks: implement DeletedTask and UpgradedTask signals Nov 27 12:51:07 gnutoo: I sent you an email with my rules.yaml Nov 27 12:51:18 is that all it does? Provide a clear text name? Nov 27 12:52:01 ok Nov 27 12:52:26 why can't we do that in real-time, just looking it up when we need one, rather than parsing 2100 network names at startup? Anyway I don't care. But 9 seconds earlier SIM PIN is nice Nov 27 12:52:59 * spaetz keeps that as local patch Nov 27 12:52:59 leviathan_, so feel free to improve my patch because many things are disabled but: Nov 27 12:53:08 gnutoo, have you tested phone calls? Nov 27 12:53:12 *xorg shows up,and isn't black Nov 27 12:53:20 *phone register Nov 27 12:53:34 spaetz: I don't know of any other use of that huge tab Nov 27 12:53:46 I always get registration denied Nov 27 12:53:47 playyaWorking__, phonecalls work but maybe that's one of the few things that work....no alsa for instance Nov 27 12:53:55 but I also got that error under android Nov 27 12:53:59 leviathan_, ah? Nov 27 12:54:10 so I believe its more a problem with the SIM card Nov 27 12:54:11 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * rcbe4fb2a5e0a 10framework/framework/subsystems/opimd/pimd_tasks.py: opimd: Tasks: few references to Notes remained, kill'em! Nov 27 12:54:15 then with the OS Nov 27 12:54:21 leviathan_, do you have another phone? Nov 27 12:54:28 yes Nov 27 12:54:33 there the simcard works Nov 27 12:54:38 ok Nov 27 12:54:42 even with the openmoko? Nov 27 12:55:17 both openmoko freerunner do perfectly acceppt my SIM-Card Nov 27 12:55:27 and it works under SHR and ASU Nov 27 12:55:30 ok Nov 27 12:55:38 strange...did you trace the AT commands Nov 27 12:55:42 like for instance: Nov 27 12:55:49 AT+CREG if I remember well Nov 27 12:55:57 I think its a deeper layer problem Nov 27 12:55:58 I don't have 0707 here Nov 27 12:55:58 on this computer Nov 27 12:56:04 I cant register either under Android Nov 27 12:56:19 but what android? Nov 27 12:56:21 replicant or another? Nov 27 12:56:29 cupcake Nov 27 12:56:47 what I meant is what ril did you use? Nov 27 12:56:49 and then in the beginning the initial android Nov 27 12:56:51 mmm I've an idea Nov 27 12:56:57 which was on it, when I bought it Nov 27 12:56:57 is your phone simlocked? Nov 27 12:57:08 spaetz: mickey|zzZZzz: seems no other phone needs such a networks.tab. It's no static data and needs to be naintained and updated frequently (networks change their names every now and then), and no other phone breaks when not updated regularly Nov 27 12:57:10 ohhh Nov 27 12:57:20 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * r2dcd511ccdd1 10framework/framework/subsystems/opimd/pimd_notes.py: opimd: Notes: fix copy'n'paste error in DeletedNote Nov 27 12:57:33 leviathan_, that would be a bad news Nov 27 12:57:37 you mean the gsm-part? Nov 27 12:57:56 because software side there cant be locked anything Nov 27 12:58:04 I mean in the firmware of the gsm Nov 27 12:58:08 ok Nov 27 12:58:15 could be possible Nov 27 12:58:20 huh? Nov 27 12:58:22 its from t-mobile Nov 27 12:58:32 but a friend of mine used his sim Nov 27 12:58:36 and with it it worked Nov 27 12:58:44 he was able to do phonecalls Nov 27 12:59:00 its really strange Nov 27 12:59:09 the phone seems to hate my SIM-Card Nov 27 12:59:12 >_< Nov 27 13:00:26 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * reba9b439f8ef 10framework/framework/subsystems/opimd/pimd_dates.py: opimd: Dates: fix copy'n'paste errors in new signals Nov 27 13:00:52 TAsn, mrmoku: merged into master Nov 27 13:01:11 DocScrutinizer-8: my old Motorola C651 had something like networks.tab Nov 27 13:01:28 so what? Nov 27 13:01:29 DocScrutinizer-8: i "hacked" it and replaced "Plus GSM" text into "dos GSM" :P Nov 27 13:01:49 DocScrutinizer-8: just clarryfing after you said "seems no other phone needs such a networks.tab" :P Nov 27 13:01:59 did it break? doesn't it work today because this table is out of date? Nov 27 13:02:13 it works, but with old names Nov 27 13:03:58 so is an old name worse than e.g. "207"? As mentioned before, I see "wrong" name here too, on FR, just for 20sec. then network manages to send clear text name info and display changes to correct name. Nov 27 13:04:34 I really don't care to have "207" instead of a wrong cleartext name for 20sec Nov 27 13:05:35 so I suggest for 99% of users it would be a great bargain to completely discard that foo Nov 27 13:09:33 dos1: great, thanks Nov 27 13:09:36 instead of wasting storage, mem, and cpu on a >100k useless table Nov 27 13:13:18 DocScrutinizer-8: I don't care about making big changes in a maintenance-only python prototype blob. Whether new fso* needs it can (and should) be debated. I just want to get the most out of the python blob we have Nov 27 13:13:41 and if I can speed it up by 9 seconds without visible drawbacks (works fine here), I'll do it :) Nov 27 13:14:15 spaetz: of course. no criticism on that Nov 27 13:14:30 spaetz: great finding indeed Nov 27 13:14:34 hehe, k. Nov 27 13:14:45 not mine though, the trac ticket had already existed Nov 27 13:15:21 uhh... ouch Nov 27 13:15:28 Actually I could see what happens when I just pull networks.tab away under frameworkd's ass Nov 27 13:15:30 spaetz: now I did the same mistake you did :| Nov 27 13:15:42 * mrmoku added a srcpv branch to OE :( Nov 27 13:17:19 spaetz: the problem is, coders sometimes are with their head in the clouds, especially when using OO. "create object tablecontent" ouuuuuch Nov 27 13:17:23 hehe Nov 27 13:17:49 mrmoku: who noticed? Nov 27 13:18:54 I noticed it immediately Nov 27 13:19:40 spaetz: I bet a grep would be faster than what network.py does there Nov 27 13:20:49 mrmoku: lucky you Nov 27 13:21:21 why? Nov 27 13:21:30 before koen attacked you :) Nov 27 13:21:40 well... he can still do that :( Nov 27 13:21:44 as I can't remove it Nov 27 13:21:52 mrmoku: jaj :/ Nov 27 13:22:01 actually just replacing networks.tab with an empty file doesn't seem to have any drawbacks. It still shows "Swisscom" in my status top bar thingie Nov 27 13:22:13 ahh, you added a NEW BRANCH 8-) Nov 27 13:22:14 git push origin srcpv:martin_jansa/srcpv is what it should have been... Nov 27 13:22:23 spaetz: what I told ya Nov 27 13:22:27 I mistook it as in you added SRCPV to a recipe :) Nov 27 13:22:31 mrmoku: me too from e-mail Nov 27 13:22:53 ping mickey for remove :) Nov 27 13:22:58 mrmoku: mickey will help you. It just costs 3 bars of swiss chocolade :) Nov 27 13:23:21 hmm... wondering if 3 Weisswurst will be fine as well :P Nov 27 13:24:13 mrmoku: if you manage to bring them to FRA before 12:00 ;-P Nov 27 13:24:33 DocScrutinizer-8: sure thing :P Nov 27 13:24:57 JaMa|Wrk: at least _your_ srcpv branch is update now :) Nov 27 13:25:15 s/update/uptodate/ Nov 27 13:25:15 mrmoku meant: JaMa|Wrk: at least _your_ srcpv branch is uptodate now :) Nov 27 13:25:22 DocScrutinizer-8: Interestingly enough. shr-today shows "M-Budget" (which is my provider). Illume top bar shows "Swisscom" which is the network they use as backend. Nov 27 13:25:34 so there seem to be 2 sources of information anyway Nov 27 13:25:46 and that is with networks.tab already removed Nov 27 13:26:18 spaetz: actually there are. name and name-long iirc Nov 27 13:26:27 ah Nov 27 13:27:41 SHR: 03seba.dos1 07opimd-utils * r78b7f09723bb 10/ (opimd-messages opimd-notifier opimd-resolve): opimd-messages, opimd-notifier, opimd-resolve: fix contact name resolving Nov 27 13:27:52 mickey|zzZZzz: do you like Weisswurst? Nov 27 13:28:59 * DocScrutinizer-8 decides it's time for a coffee finally Nov 27 13:30:41 is the babiloo thingie in already? Nov 27 13:30:45 I got a response: Nov 27 13:30:46 http://bazaar.launchpad.net/%7Evaudano/babiloo/efl/download/head%3A/babiloo_2.0.93.tar.g-20091127111926-nbrakkua6acwrlmg-3/babiloo_2.0.9-3.tar.gz Nov 27 13:31:12 this would be the SRCURI for the babiloo-efl tarball Nov 27 13:32:45 hehe: * [new branch] srcpv -> origin/srcpv Nov 27 13:33:04 it propagates already Nov 27 13:34:44 spaetz: would be "interesting" to find out about the mem footprint penalty for loading that >100k table foo Nov 27 13:36:17 * DocScrutinizer-8 wouldn't surprised of a factor 10 even Nov 27 13:38:24 probably python converts each char to a UTF-16 object ;-) Nov 27 13:38:38 bah ;-P Nov 27 13:38:51 utf-8 by default. Python is quite sensible in that regard, actually Nov 27 13:39:31 aah, no fonttype, properties like bold, cursive etc? XP Nov 27 13:39:57 no, that would be WinWord :) Nov 27 13:40:49 dos1, thanks. Nov 27 13:41:13 or Qt in twinklephone. Open a large log with tinkles logviewer to get really scared ;-P Nov 27 13:42:25 IIRC it's factor 30, and twice that on "close" - muhahaha Nov 27 13:47:39 mrmoku: http://tinderbox.openembedded.net/public/logs/task/3798043.txt Nov 27 13:47:42 bluez failed to build Nov 27 13:48:28 strange this time no libnl is installed... Nov 27 13:50:13 dos1: I had to remove libnl2 and libnl1, then build libnl1 againd and then libnl2 and then bluez4 Nov 27 13:50:14 * mrmoku moving downstairs Nov 27 13:50:15 brb Nov 27 13:53:44 spaetz: what's the runtime savings with an empty networks.tab? Nov 27 13:56:39 /usr/bin/env bzr pull -r 294 --overwrite failed with signal 3, output: Nov 27 13:56:39 bzr: ERROR: No pull location known or specified. Nov 27 13:56:51 bitbake bug when updating brz repository? Nov 27 13:58:08 bazaar? Nov 27 14:01:54 DocScrutinizer-8: just running the test app: full networks.tab:17 sec, no decode("UTF-8") 9 sec, empty networks.tab (around 5 sec) Nov 27 14:02:22 so parsing the whole shebang isn't that expensive. decoding to utf8 is Nov 27 14:02:42 but it is saved in utf8, so that is effectively a noop IMHO Nov 27 14:02:55 still expensive enough to better get rid of it Nov 27 14:08:43 spaetz: "time grep 901 networks.tab #> real:0.786s, user:0.155s, sys:0.080s" Nov 27 14:10:05 TAsn: ^so much for the "optimized python code, competing to a standard regex call Nov 27 14:11:27 to it's defence, the file is doing much more than just grepping through. Nov 27 14:11:33 defense Nov 27 14:11:55 how can I tell bitbake that a released version should be preferrable over a _bzr version? Nov 27 14:12:35 spaetz: I bet it's doing a grep for MCC first Nov 27 14:13:52 spaetz: put DEFAULT_PREFERRENCE = "-1" to _bzr.bb Nov 27 14:13:54 spaetz: and if I got it right, the diff between empty tab and 100k tab still is some 4sec, no? Nov 27 14:15:39 yep. parsing and building a hash and what-not Nov 27 14:16:04 JaMa|Wrk: thanks Nov 27 14:16:04 yep, total nonsense Nov 27 14:16:15 oh, split Nov 27 14:16:19 spaetz: I have patch for bitbake with bzr :) Nov 27 14:16:27 Now on revision 294. Nov 27 14:17:46 spaetz: eating 12 seconds for a function we rarely *ever* need Nov 27 14:18:31 and which could basically be done by a syscall to grep taking 0.7 seconds Nov 27 14:19:01 maximum Nov 27 14:20:17 but obviously all "real programmers" think system() is a ugly uncool thing to do :-/ Nov 27 14:20:28 JaMa|Wrk: I just pushed my stuff Nov 27 14:20:58 so babiloo should be build from a downloaded tar ball now by default :) Nov 27 14:21:08 "nah, we want to *optimize* it" ( TAsn !!) Nov 27 14:21:32 it might need cleaning out though. and I don't know if the existing bzr doesn't count as superior now.. Nov 27 14:21:40 DocScrutinizer, huh? Nov 27 14:21:49 I don't see anything about regex in all you have said. Nov 27 14:22:07 furthermore, grep = just reading lines of text and comparing. Nov 27 14:22:13 there's no way to make it faster Nov 27 14:22:28 TAsn: grep is regex basically Nov 27 14:22:28 and making it *in app* will be faster than using "system()" Nov 27 14:23:05 dos1: ping Nov 27 14:23:14 can you clean and build babiloo-efl? Nov 27 14:23:23 DocScrutinizer-8, so it means you can write an app to do it faster than grep. Nov 27 14:23:31 TAsn: or, for your better understanding, s/grep/sed/ Nov 27 14:23:31 spaetz: ok, after fixing that libnl* issue Nov 27 14:23:38 thanks Nov 27 14:23:47 spaetz: do you don't like locales in babiloo? Nov 27 14:24:02 Heinervdm: I do. But they are not included in Luca's tarball :) Nov 27 14:24:12 ah :) Nov 27 14:24:16 so the installation crapped out Nov 27 14:24:23 same with the .doc pdf Nov 27 14:24:48 Actually, I have never used babiloo, so I don't even know how it looks like Nov 27 14:25:25 i startet it once after making the package Nov 27 14:25:48 and i thought the pdf can be usefull for beginners :) Nov 27 14:26:26 Luca asked me if we need more. So we can tell him that it would be nice to include the pdf in the next release Nov 27 14:27:03 spaetz: i'm writing an answer to him, that i want to have distutils ;) Nov 27 14:28:28 to summarize: mv network.tab network.tab-away && touch network.tab -> boot spped up of 12 seconds Nov 27 14:28:42 Heinervdm: also write him that putting src tar balls in bazaar directly might not be the best solution Nov 27 14:29:02 weird URLs. launchpad provides a possibility to host src tarballs Nov 27 14:29:21 also bloating the bzr uneccessarily. Nov 27 14:30:42 ecore failed to build.. ecoreecore_evas_private.h:6 errorerror: Evas_Engine_FB.h: No such file or directory Nov 27 14:31:13 cleaning out... Nov 27 14:38:52 Heinervdm: I wouldn't buy new desktop graphics card yesterday, if I knew that you'll make all games available on neo :) Nov 27 14:39:07 hehe Nov 27 14:39:15 :D Nov 27 14:39:24 * JaMa|Wrk and TAsn have COD on wishlisht :) Nov 27 14:40:24 :P Nov 27 14:41:12 dos1: very cool, can you also rebuild package index? Nov 27 14:41:16 and thanks Nov 27 14:42:13 I'm just happy I have xboard :P Nov 27 14:42:25 and syncing :) Nov 27 14:44:54 spaetz: done Nov 27 14:45:00 thanks Nov 27 14:45:23 mmh, it still shows 2.0.9-bzrr288-r0.4 in package list Nov 27 14:45:38 darn. I think it did not pick up my DEFAULT_PREFERENCE=-1 Nov 27 14:45:53 well anyway, it will sort itself out over time. I hope Nov 27 14:45:55 i havn't added these packages to the feed, so we should do that too Nov 27 14:46:15 Heinervdm: right Nov 27 14:46:35 let me do that right away Nov 27 14:46:38 before I forget it Nov 27 14:46:48 dos1: have you tried duke3d with xorg Nov 27 14:46:54 spaetz: oh, right Nov 27 14:46:57 it landed in morgue Nov 27 14:46:59 all/morgue/babiloo-efl_2.0.9-3-r0.4_all.ipk Nov 27 14:47:15 because it is considered older= Nov 27 14:47:16 ? Nov 27 14:47:18 yup Nov 27 14:47:19 why does a package go to morgue? Nov 27 14:47:34 Heinervdm: because they are considered older ;P Nov 27 14:47:39 ah Nov 27 14:47:48 mmh, can you manually remove the babiloo-efl_bzr.bb then? Nov 27 14:47:55 cleaning should be enough, should it? Nov 27 14:47:59 sorry for the trouble Nov 27 14:48:18 spaetz: well, shouldn't PE be bumped? or something different? Nov 27 14:48:47 perhaps yes, but we just added it today, so there are no real users besides us yet Nov 27 14:49:28 ok then Nov 27 14:51:17 no removing .ipk is enough.. Nov 27 14:51:38 -c clean won't work irc.. you need to remove that file.. Nov 27 14:51:45 JaMa|Wrk, spaetz: done ::P Nov 27 14:51:56 (removed ipk, not recipe) Nov 27 14:52:08 or maybe -c clean doesn't work only if you bump revision or something in between Nov 27 14:52:35 btw Nov 27 14:52:39 am I the only one that Nov 27 14:52:42 -c rebuild Nov 27 14:52:45 doesn't work for him anymore? Nov 27 14:54:21 TAsn: no.. for me too Nov 27 14:54:35 :( Nov 27 14:54:38 sucks. Nov 27 14:54:43 I liked that. Nov 27 14:54:50 it rebuilds Nov 27 14:55:08 but doesn't call clean for unpack and patch reapply etc Nov 27 14:55:29 exactly. Nov 27 14:55:34 so it's essentially like "build" Nov 27 14:55:47 I just pushed babiloo-efl into the feed Nov 27 14:56:05 TAsn: not like build -f -c compile imho then -b build Nov 27 14:56:07 and removed hermining, where even the original requester said that it's not working Nov 27 14:56:26 TAsn: because -b build says.. no need to run any task Nov 27 14:56:30 i c. Nov 27 14:56:38 okie :) Nov 27 14:56:48 and we actually have mysql in our task-shr-feed ! It's not just concidentally pulled in Nov 27 14:57:01 Which joker added that? Nov 27 14:57:09 what's wrong with that? Nov 27 14:57:20 TAsn: are you running mysql on your FR? Nov 27 14:57:24 I mean, it's nice to have on the feed. Nov 27 14:57:33 do you think the FR is suitable for running mysql? Nov 27 14:57:33 spaetz, no, but I have seen people who do. Nov 27 14:57:37 spaetz: sorted task-shr-feed? Nov 27 14:57:38 really? Nov 27 14:57:50 JaMa|Wrk: just some random bits Nov 27 14:58:03 not exhaustively. (yes, I know that confuses git blame) Nov 27 14:58:05 spaetz, yes. Nov 27 14:58:05 furthermore Nov 27 14:58:05 SHR is not fr only Nov 27 14:58:27 bad joke would be having it included in image :P Nov 27 14:58:30 but for feeds it's ok Nov 27 14:58:34 dos1, :P Nov 27 14:58:40 i would even like firefox and gimp in feeds ;P Nov 27 14:58:49 it didn't build anyway IIRC Nov 27 14:58:52 dos1, probably yeah. Nov 27 14:58:59 * JaMa|Wrk agree.. mysql in feed is better then users building it or using version built against om2008 Nov 27 14:59:30 exactly ;) Nov 27 15:00:26 * spaetz shakes head. waste of cpu cycles, but doesn't hurt anybody, so well Nov 27 15:02:13 :P Nov 27 15:08:49 dos1: I've ran OpenOffice 2 on my Zaurus, worked just fine once you waited > 5 min. for it to start :) Nov 27 15:09:25 :P Nov 27 15:09:31 what about abiword? should probably work. Nov 27 15:09:36 though anyhow, it doesn't matter. Nov 27 15:09:47 those are inferior ways to pass data :P Nov 27 15:09:54 office suites suck :P Nov 27 15:10:05 hehe Nov 27 15:10:10 SHR: 03seba.dos1 07opimd-utils * r9c1283ec0374 10/opimd-messages: opimd-messages: fix loading more messages on reaching bottom edge of scroller Nov 27 15:10:17 yeah Nov 27 15:10:30 dos1: almost forgot. there's some stranges in blanking. default blank time is set to 10sec and setting it to more always get lost after reboot Nov 27 15:10:33 I'd love a SIP/IAX client in SHR though Nov 27 15:10:44 Q-Master: set it in /etc/frameworkd.conf Nov 27 15:11:00 Q-Master: fsodeviced isn't storing changes done by dbus interface yet :x Nov 27 15:11:05 dos1: what fields? Nov 27 15:11:18 Q-Master: in [fsodevice.kernel_idle] Nov 27 15:13:09 dos1: ok Nov 27 15:13:12 thnx Nov 27 15:13:27 sybren: then do some research for one that suites the FR ;) Nov 27 15:21:47 http://trac.shr-project.org/trac/ticket/720#comment:3