**** BEGIN LOGGING AT Thu Jan 14 02:59:57 2010 Jan 14 06:07:18 gena2x: pong Jan 14 06:11:56 Q-Master: do you have this image? d45d7200db6fcd364a6e13d91bc00418 tmp/deploy/images/om-gta02/shr-full-eglibc-ipk--20100113-om-gta02.rootfs.tar.gz Jan 14 06:12:22 Q-Master: please check md5sum, because 20100113 was rebuilt during the day.. Jan 14 06:22:06 seems that busybox is all messed up... Jan 14 06:22:13 in SHR-u Jan 14 06:23:22 ThibG: was opkg postinst (Configuring opkg..) run before busybox posinst for you? Jan 14 06:23:38 huh, don't know Jan 14 06:24:00 anyway, it won't shut down Jan 14 06:24:30 when I try manually, busybox complains Jan 14 06:25:30 http://pastebin.com/de335eb4 Jan 14 06:25:32 there should be enough info to fix it in HeadsUp thread.. I'm leaving now to work sorry... Jan 14 06:26:15 hm ok Jan 14 06:26:26 this should never happen if you had upgraded update-alternatives-cworth or opkg before busybox.. please check.. Jan 14 06:27:22 well, I've updated daily since saturday Jan 14 06:27:41 (with opkg upgrade) Jan 14 06:28:02 which -cworth version do you have? Jan 14 06:28:44 update-alternatives-cworth - 0.99.154-r4.0.4 Jan 14 06:29:12 max_posedon: (fso2 gentoo builds) wow :-) Jan 14 06:32:45 mickeyl: I really, really would like to stop using libframeworkd-glib... tried it twice and spent quite some time on trying libfso-glib... with no success :| Jan 14 06:33:01 (and I've seen the "lengthy output" before yesterday's upgrade) Jan 14 06:33:39 ThibG: what does it do if you reinstall busybox? Jan 14 06:34:07 and I'm really of for 40mins or so.. Jan 14 06:34:17 is it safe to do that? Jan 14 06:35:01 I mean, I would prefer not to screw up with my installation, since I've no µSD yet :p Jan 14 06:36:37 well, doing it Jan 14 06:36:54 I can always reflash afterwards Jan 14 06:44:32 mickeyl: I will give it another try... and see if I can come up with an example that shows how it does not work Jan 14 06:47:52 JaMa, I've reinstalled busybox, it doesn't change anything Jan 14 07:02:25 JaMa: 20100111 also doesn't boot. It hangs after filesystems mounting. 8( Jan 14 07:09:43 ps, and other things, are also messed up by busybox... Jan 14 07:24:04 ThibG: then show me those alternatives files Jan 14 07:24:28 well, I don't know how the alternatives work Jan 14 07:24:30 ThibG: and yes we can fix everything here as long you don't shutdown that device Jan 14 07:24:40 ThibG: then please read that thread.. Jan 14 07:25:15 Q-Master: what md5 you have of your downloaded 20100113 image? Jan 14 07:26:05 I don't have /usr/lib/ipkg/alternatives Jan 14 07:26:13 that's good Jan 14 07:26:30 as it should be removed in last -cworth and opkg upgrade Jan 14 07:26:44 and what do you have in /usr/lib/opkg/alternatives/init Jan 14 07:26:47 yeah, but still, the alternatives are wrong... Jan 14 07:26:50 and also s/init/ps/ Jan 14 07:27:04 /sbin/init Jan 14 07:27:04 ../bin/busybox 50 Jan 14 07:27:16 same with ps Jan 14 07:27:32 ThibG: yeah that's definitelly wrong.. Jan 14 07:27:57 ThibG: can you please check diff between those files in /usr/lib/opkg/alternatives/ /usr/lib/opkg/alternatives.backup and /usr/lib/ipkg/alternatives.backup? Jan 14 07:29:13 I don't have /usr/lib/opkg/alternatives.backup Jan 14 07:29:38 ThibG: sorry its with '-' instead '.' Jan 14 07:29:39 ipkg/alternatives-bakup's one have '/sbin/init.sysvinit 60' too Jan 14 07:30:14 so ipkg/alternatives-bakup's was longer? I mean from wc -l POV? Jan 14 07:30:15 I don't have opkg/alternatives-backup either Jan 14 07:30:39 JaMa: it's wrong. but 20100111 doesn't boot also. And I mean tar.gz image not jffs2 Jan 14 07:31:06 JaMa, yes Jan 14 07:31:17 Q-Master: I understand that 20100111 doesn't boot that's why I'm asking if you tried NEW 20100113 (where it should be fixed) Jan 14 07:31:35 ThibG: now tricky part.. Jan 14 07:31:58 mv /usr/lib/ipkg/alternatives-backup /usr/lib/ipkg/alternatives/ Jan 14 07:32:06 JaMa: I haven't yet dloaded it. Jut gimme 10 minutes Jan 14 07:32:17 ThibG: reinstall opkg and send me the output Jan 14 07:32:56 Q-Master: 20100114 is already building but it will take some more time.. Jan 14 07:33:27 JaMa: I'm downloading newer 20100113 Jan 14 07:36:30 JaMa, http://pastebin.com/m19694097 Jan 14 07:38:27 ThibG: Replacing '/usr/lib/opkg/alternatives/ps' with '/usr/lib/ipkg/alternatives/ps' Jan 14 07:38:40 ThibG: is it right in /usr/lib/opkg/alternatives/ps now? Jan 14 07:38:59 yeah Jan 14 07:39:39 ThibG: I cannot understand how it is working now for you and wasn't before.. Jan 14 07:39:52 It isn't just for me Jan 14 07:39:56 ThibG: now reinstall busybox and you should be safe again.. Jan 14 07:40:34 ThibG: do you remember your upgrade steps a bit? what was upgraded first etc? Jan 14 07:40:46 ThibG: just in case you did something unusuall.. Jan 14 07:40:54 http://shr-project.org/trac/ticket/865 Jan 14 07:41:16 ThibG: ie very first fix (I gave) was to remove -cworth Jan 14 07:41:34 hm, I've never removed -cworth Jan 14 07:41:58 in fact, since I've reflashed SHR-u from 20100110, I've just opkg update && opkg upgrade Jan 14 07:41:59 and do you have /usr/lib/opkg/alternatives-backup now? Jan 14 07:42:14 yeah Jan 14 07:42:16 Aaah you reflashed.. that could be it.. Jan 14 07:42:43 not sure.. but just yesterday I've found that those images are still a bit broken.. Jan 14 07:42:53 ok Jan 14 07:44:02 I'll remove all images between 20100110 and 20100114 after 20100114 finish.. Jan 14 07:50:36 update-alternatives: Linking //bin/ps to ps.procps Jan 14 07:50:39 yeah, thanks :) Jan 14 07:52:31 * JaMa still curious how that happen even with reflashed 20100110.. Jan 14 07:53:03 * JaMa knows :) Jan 14 07:53:16 oh? Jan 14 07:53:26 you didn't have /usr/lib/opkg/ in that image right? Jan 14 07:53:58 that's why /usr/lib/opkg/alternatives-backup could not be created (because I didn't have -p before..) Jan 14 07:54:08 huh? Dunno Jan 14 07:54:17 and that's why that lenghty output was all about errors i expect Jan 14 07:54:23 I'll mount the jffs2 image and see Jan 14 07:54:31 > ah, maybe Jan 14 07:56:17 http://git.openembedded.org/cgit.cgi/openembedded/commit/?id=d6eb5340067ae02d3289ce6311d80a9752dbb8bc that first if wasn't there before Jan 14 07:57:05 I was thinking about alternatives where /usr/lib/opkg/alternatives doesn't exist and then added it just in case.. not knowing that we already have that case in our images.. Jan 14 07:58:38 and that was becase there was still -cworth on shr buildhost not upgraded because bitbake "didn't need it" so didn't upgrade it.. Jan 14 08:02:07 argh Jan 14 08:02:11 something went wrong Jan 14 08:02:36 the device doesn't boot anymore... Jan 14 08:02:52 JaMa: hi, can you share config for 32.3 kernel, i din't found one in /proc/config.gz Jan 14 08:03:16 oh, nevermind Jan 14 08:03:27 just no splashscreen displayed Jan 14 08:05:41 gena2x: sure mmt Jan 14 08:06:00 gena2x: btw did you "modprobe configs"? Jan 14 08:06:03 JaMa: same. It boots, mounts the fat partition and hang. 8( Jan 14 08:07:12 usb network is up and it pings. but that's all Jan 14 08:08:06 JaMa: configs? ok, no need in config :) Jan 14 08:08:10 gena2x: http://jama.homelinux.org/org.openembedded.shr.images/config.gz Jan 14 08:08:28 gena2x: yeah IIRC I told you its in module :) Jan 14 08:08:40 JaMa: somehow i didn't found that module yesterday :) Jan 14 08:08:48 and because of no debug - nobody knows wha't happening 8( Jan 14 08:09:11 * JaMa thinks everytime that the module is named config Jan 14 08:09:25 because its just ONE! :) .. i know about Jan 14 08:09:38 Q-Master: ? Jan 14 08:09:42 Q-Master: do you use Qi? Jan 14 08:09:47 JaMa: yep Jan 14 08:09:57 Q-Master: boot it with power button pressed or did you already? Jan 14 08:10:11 JaMa: not yet. Jan 14 08:10:16 JaMa: let me try Jan 14 08:10:18 why its mouting fat partition at all? Jan 14 08:10:41 you have vfat somewhere else then rootfs on uSD? Jan 14 08:11:04 JaMa: Ops. It started to do something. Jan 14 08:11:31 JaMa: illume booted Jan 14 08:11:54 JaMa: 5 mins for first start IMO too long. 8( Jan 14 08:12:06 shr wizard started Jan 14 08:19:47 this morning I have discovered an interesting bug. Jan 14 08:19:47 the dialler disappears after the screen is locked then unlocked, meaning that I cannot hang-up nor access the phone keypad to send DTMF(?) codes. Jan 14 08:19:47 Anyone have seen this behaviour? Jan 14 08:19:47 it runs unstable SHR with latest updates Jan 14 08:20:40 ah, yeah, I have the same issue Jan 14 08:20:58 is there a bug? Jan 14 08:20:58 also with messages, contacts or anything alike Jan 14 08:21:14 JaMa: seems all that debug options are turned off in config 32.3 config Jan 14 08:21:31 gena2x: yes.. as I said in e-mail on shr-devel.. Jan 14 08:21:37 gena2x: and they always were.. Jan 14 08:22:04 JaMa: oh, i'm not on shr-devel, i'll read this now. Jan 14 08:22:22 JaMa: I've finished testing, without .31. Jan 14 08:22:23 gena2x: well just UBI had lots of debug.. because it didn't work for me.. and after first few ubi success stories I removed that too Jan 14 08:22:33 JaMa, by the way, have shr-devel moved somewhere? I've registered to shr-devel and shr-user, but I've only got mails from shr-user Jan 14 08:22:41 JaMa: So, i'll publish results in few minutes Jan 14 08:22:49 gena2x: yeah.. .31 is least significant from all 3.. so you can ignore it.. Jan 14 08:23:05 ThibG: I don't think so.. Jan 14 08:23:15 ThibG: And I'm still reciving e-mails from both Jan 14 08:23:24 JaMa: had to add some glamo's parameters to have working sd on 32.3. Jan 14 08:23:42 gena2x: I'm using it on uSD.. here Jan 14 08:24:19 JaMa: i have same card working for .29 and don't for .32 without adding paramenter to uboot Jan 14 08:25:13 gena2x: see [Shr-Devel] community updates and fast kernel, Pavel Vanek tested boot time with all kernels.. including .3* Jan 14 08:25:29 gena2x: ah.. that rootdelay parameter? Jan 14 08:25:36 JaMa: no Jan 14 08:26:07 gena2x: well I never tried it with u-boot, for Qi was rootdelay enough.. Jan 14 08:26:14 gena2x: so which parametr? Jan 14 08:26:38 JaMa: glamo_mci.sd_drive=2 Jan 14 08:27:13 JaMa: i did't really understand that it do. Jan 14 08:27:22 JaMa, odd. I've sent a mail to the list, and it says I'm not registered. I tried to register again, it says that I've already registered... Jan 14 08:27:36 gena2x: maybe uSD is initiallized faster with it and then you don't need rootdelay? Jan 14 08:27:58 gena2x: this will help if you boot from uSD and use GPS. Otherway it might hang. Jan 14 08:28:16 gena2x: if you have time can you please try it with rootdelay=10 and without that glamo_mci? I'll try vice versa with normal rootdelay and with glamo_mci param Jan 14 08:28:24 JaMa: without this parameter i had error soon in boot sequence Jan 14 08:28:39 gena2x: about unknown root device? Jan 14 08:28:41 JaMa: no, i set rootdelay to 10 in all cases Jan 14 08:28:49 ah, ok Jan 14 08:29:08 JaMa: on, it was panicking on remount after checking root ext2 Jan 14 08:29:34 JaMa: to have partitions detected i haf to add glamo_mci.sd_max_clk=10000000 Jan 14 08:29:57 hmm strange :/ Jan 14 08:33:19 JaMa: yes, but it didn't work without that for sure, in reproducible way. Jan 14 08:35:12 gena2x: glamo_mci.sd_drive=2 means that Glamo will drive SD lines "harder", (to drive line high one connects it to Vbus with some resistance, the less the resistance, the more current can flow, the faster the transition from 0->Vbus; the more interference might be caused to GPS) Jan 14 08:35:59 PaulFertser: thanks, clean Jan 14 08:39:26 about µSD cards, I've read that some may not run properly. Is it a common problem? Or is it unlikely to happen? Jan 14 08:40:23 ThibG: well, it's not very likely that you get a card that doesn't work even when you lower the frequency and set a higher sd_drive value. Jan 14 08:40:49 ok Jan 14 08:45:11 ThibG: there were some weird interference issues when some cards stopped working in the presence of GSM traffic but those are fairly rare. Jan 14 08:46:01 ok, but those issue are mainly hardware issues? Jan 14 08:46:28 ThibG: yes Jan 14 08:46:47 erf ok Jan 14 08:50:33 JaMa: (read mailings) so, .32 have no problems except resume? Jan 14 08:51:32 gena2x: yes and suspend too Jan 14 08:51:42 JaMa: i've published test results Jan 14 08:51:45 gena2x: and some parts of fso needs update Jan 14 08:52:19 JaMa: should telephony work out of box? Jan 14 08:52:23 gena2x: but mickey is waiting for switch as it's difficult for supporting more kernels at once (different sysfs nodes etc) Jan 14 08:52:44 gena2x: see http://build.shr-project.org/tests/mrmoku/switch-for-kernel Jan 14 08:52:51 JaMa: ok Jan 14 08:53:28 JaMa: i just think which kernel to use to continue my investigation for optimization. Jan 14 08:53:58 JaMa: if anything is working, i'll continue with .32. Jan 14 08:54:53 gena2x: I don't see results in :/ http://www.bsdmn.com/openmoko/debugoptstesting/ Jan 14 08:55:03 and no e-mail arrived yet.. Jan 14 08:55:07 gena2x: mainly sysfs nodes and alsa statefiles are different Jan 14 08:55:17 but I should work a bit now.. so I'll check later Jan 14 08:55:46 and you can enable rfkill plugin with .32 :) Jan 14 08:55:58 JaMa: ops Jan 14 08:56:16 JaMa: send without cc to the list. Jan 14 08:56:40 mrmoku: killed few broken images (with -cworth in staging) and added md5 generation to sync_ aliases Jan 14 08:56:55 mrmoku: before someone sends a patch to bitbake to handle it properly Jan 14 08:57:28 JaMa: fixed. http://lists.openmoko.org/pipermail/openmoko-kernel/2010-January/010858.html Jan 14 08:57:48 JaMa: md5 generation? Jan 14 08:58:32 mrmoku: md5sum called and stored for all images in image*.md5 Jan 14 08:59:39 JaMa: ahh ok, good Jan 14 09:01:16 wow, this is interesting.. Jan 14 09:01:23 dot32 355.4 91.0 946.1 275.4 2426.0 1.430 63.0 Jan 14 09:01:26 nodebug 365.0 139.9 4237.3 1043.8 4813.0 1.636 70.0 Jan 14 09:01:29 nodebug 366.0 139.9 4830.9 1082.3 4802.0 1.620 68.0 Jan 14 09:01:32 nodebug 359.5 139.7 1239.2 5681.8 4801.0 1.657 68.0 Jan 14 09:01:36 what happen in last nodebug test? :) Jan 14 09:02:24 and really nice improvement in fs create/delete (maybe because ext3 default mount options changed in .31 or .32 IIRC ? ) Jan 14 09:03:14 JaMa: i told, that fs test should not be trusted Jan 14 09:03:37 JaMa: especially as i modified glamo parameters Jan 14 09:04:55 JaMa: it need major investigation how much they may be trusted as it's unclean that lmbench is doin on that tests, any kind of caches may influence this Jan 14 09:05:57 JaMa: and it's clear that they are varying too much in similar runs Jan 14 09:06:37 JaMa: also, for .32 and .29-drm i did run without X runnning Jan 14 09:08:59 JaMa: but number diffence is so big, that it really can't be ignored Jan 14 09:10:55 gena2x: thanks for testing.. I think it shows that future is bright and promissing :) Jan 14 09:11:37 JaMa: I always thought so. And any slowdown is just reason to do something challenging :) Jan 14 09:11:46 yep, let's switch to .32. who needs suspend or a phone .... when you can have fast :P Jan 14 09:12:12 yeah.. would be nice if someone invest more time in suspend/resume issue and help larsc a bit Jan 14 09:12:15 mrmoku, spaetz TAsn apply patches, apply patches, apply patches plz) Jan 14 09:12:19 Good morning) Jan 14 09:12:20 spaetz: btw, as cpu load is decreased almost x2, it should eat less power :) Jan 14 09:12:50 spaetz: yes apply patches, apply patches :) you really want that cherry-pick from opkg.. :/ Jan 14 09:12:56 hi Jan 14 09:12:57 shr-devel has 2 patches for libframeworkd-glib and libphone-ui, really important, whithout them sources not compiles) Jan 14 09:13:16 spaetz: because 1st version of merger doesn't work if there is no /usr/lib/opkg directory Jan 14 09:13:59 JaMa: and as i've told, i see many more simple thing in kernel to have speedups :) Jan 14 09:14:01 does anybody know about intone 'unuseability' in the current shr-u image? Jan 14 09:14:04 freesmartphone.org: 03maxposedon 07libframeworkd-glib * rf68fe525c20a 10/src/frameworkd-glib-dbus.c: Jan 14 09:14:04 freesmartphone.org: Fix gcc-4.4 warning(error), about doesn't specified format string Jan 14 09:14:04 freesmartphone.org: Signed-off-by: Klaus Kurzmann Jan 14 09:15:59 SHR: 03maxposedon 07libphone-ui * rfe0cd1df51fa 10/src/phoneui-utils-device.c: Fix gcc-4.4 warning(error) check write/system calls Jan 14 09:16:07 max_posedon: done Jan 14 09:16:16 thanks!) Jan 14 09:16:22 thank you :) Jan 14 09:16:47 and I think I'll finish shr ebuilds next days Jan 14 09:16:58 and will start fight you all for snapshots) Jan 14 09:17:33 I want smth stable, on that versions, that mickey will snapshot soon. Jan 14 09:17:33 max_posedon: why not use live ebuilds instead :) Jan 14 09:17:50 JaMa, I use them now, but its a pita for users Jan 14 09:18:20 max_posedon: yeah, there is no easy "locking" for SRCREV in gentoo, IIRC Jan 14 09:18:40 JaMa: sorry day work is prio 1 at the moment :-( Jan 14 09:18:41 are you kidding?) you always can do same as in shr Jan 14 09:18:43 max_posedon: so only HEAD or you had to update ebuild with every new revision right? Jan 14 09:18:58 just setup veriable in make.conf, and ask ebuild use it Jan 14 09:19:17 max_posedon: which is pita for users as you already said :) Jan 14 09:19:41 pita is try to use upstream soft, not prepare ebuilds) Jan 14 09:19:57 main problem you just can't create smth stable Jan 14 09:20:05 max_posedon: I know you can do everything (I'm using gentoo too) but its not so common to have locked SRCREV or AUTOREV just by some config file somewhere as in bitbake Jan 14 09:20:55 people just don't use it, but its easy to write ebuild, which will do it Jan 14 09:21:14 also kde team have did some good stuff if you want use paludis Jan 14 09:21:22 I just don't like this way Jan 14 09:21:49 you know better me, how often shr-U have smth broken in fso/fso2/shr parts) Jan 14 09:22:53 anyway, one more question Jan 14 09:22:57 moin Jan 14 09:23:18 How usable will be shr(phonefsod/etc) without opimd? Jan 14 09:26:40 not usable IMO Jan 14 09:26:56 not usable for users, or just doesn't work?) Jan 14 09:27:09 main question, will sim be enough or not Jan 14 09:27:21 what I can do with it without opimd? Jan 14 09:28:33 max_posedon: no opimd, no sms Jan 14 09:28:34 max_posedon: it won't work Jan 14 09:28:42 no contacts neither Jan 14 09:29:08 dialer _might_ work though :P Jan 14 09:29:10 max_posedon: no opimd, no timers/alarms even (eventuall) Jan 14 09:29:55 max_posedon: you don't like opimd? Jan 14 09:30:05 so, my python-less image is just a dream Jan 14 09:30:12 I don't like python) Jan 14 09:30:13 ahh Jan 14 09:30:31 gena2x: (speed of uSD access) on recent kernels you'll get 30% more bandwidth when the screen is blanked. Jan 14 09:30:40 * mrmoku agrees with max_posedon and wishes someone would rewrite it :) Jan 14 09:30:47 btw, very crazy question Jan 14 09:30:57 does you heard about akonadi? Jan 14 09:30:59 I do like python but not on the Freerunner :) Jan 14 09:31:18 its semi-standard for pim in fdo terms Jan 14 09:32:12 may be, may be, its better to write backend for it, instead of "our own api" Jan 14 09:32:45 mrmoku, and one more questions, am I correct understand that qt-4 will be soon in feeds? Jan 14 09:32:48 max_posedon: ack Jan 14 09:33:43 max_posedon: probably JaMa knows what's the state of qt in our feed... Jan 14 09:34:43 max_posedon: you don't like python? wanna talk 'bout that? 8) Jan 14 09:35:02 Q-Master, M-x doctor was enough, thanks) Jan 14 09:35:10 8))) Jan 14 09:35:41 mrmoku, max_posedon: last time I checked we don't have any application building qt to our feeds Jan 14 09:35:58 I read emails about wpa-qt4, will it be first?) Jan 14 09:36:03 max_posedon: python is a good language. not extremely fast, but good. Jan 14 09:36:31 Q-Master, sure, I just don't like it on mobile Jan 14 09:36:56 and dbus opim written in python, to answer very important pim calls Jan 14 09:37:14 python can be in mobile, but not in this place... Jan 14 09:38:16 max_posedon: yes Jan 14 09:38:52 I just thinking about try akonadi for this goals some day Jan 14 09:42:38 mrmoku, sorry!, but I did typo mistake in patch Jan 14 09:43:04 in src/phoneui-utils-device.c on line 85 Jan 14 09:43:13 g_assert instead assert of course Jan 14 09:43:25 looks like I sent wrong patch to ML...) Jan 14 09:44:24 ~spank max_posedon Jan 14 09:44:25 * infobot bends max_posedon over his knee and tatoos 'ibot' on max_posedon's pasty white buttocks. Jan 14 09:45:12 max_posedon: akonaid is the kde pim server. it requires a complete mysql server on the device... Jan 14 09:45:30 max_posedon: i think that's a bit to much for the small FR Jan 14 09:45:30 Heinervdm, it isn't "KDE", it can work with sqlite backend Jan 14 09:47:09 yes? Last time i checked i worked only with mysql, but that's half an year ago or so Jan 14 09:47:50 mrmoku, typo fix in ML now, sorry) Jan 14 09:48:13 I see postgresql backend, also I saw sqlite project about this Jan 14 09:48:46 and in long terms, it isn't so big, if we want some email and other support in opim I think Jan 14 09:49:41 PaulFertser: hm... why such difference? Jan 14 09:49:48 but i think, for our needs fsopimd will suite better Jan 14 09:51:22 wtf is fsopimd? Jan 14 09:51:31 just a dream) Jan 14 09:51:38 eheh Jan 14 09:52:11 joerg_42: it's the name of the vala rewirte of opimd that we will have in the future ;) Jan 14 09:52:12 gena2x: all SD traffic goes through glamo and that beast spends some of its internal bandwidth on screen refresh. When blanked, its pixelclock is stopped and all the bandwidth is now free to be used by SD. Jan 14 09:52:33 Heinervdm: aaah vaporware Jan 14 09:52:41 PaulFertser: nice. Jan 14 09:52:57 joerg_42: i hope not Jan 14 09:53:48 PaulFertser: I'm curious: are there any measurements of this effect before and after the FIFO patch? Jan 14 09:54:22 Weiss: i'm afraid not, larsc should know better, it's his invention. Jan 14 09:54:27 PaulFertser: but we are far from bandwidth limit of glamo with our 2Mb/s (according to dd)? Jan 14 09:55:13 SHR: 03maxposedon 07libphone-ui * r2a14ce504d23 10/src/phoneui-utils-device.c: fix typo in gcc-4.4 patch Jan 14 09:55:13 max_posedon: ok, applied the fix of the fix ;) Jan 14 09:55:22 there is a big elementary bug in current shr-u Jan 14 09:55:22 http://trac.enlightenment.org/e/ticket/36 Jan 14 09:55:28 I mean: edje bug Jan 14 09:55:37 gena2x: i'm afraid i do not know this topic that deep to say for sure why is that. But larsc certainly has a clue. Jan 14 09:55:53 any edje based application is involved (so anything, because elementary theme are edje-based) Jan 14 09:56:05 PaulFertser: ok :) Jan 14 09:56:21 gena2x: there seems to be some kind of memory access conflict which hits us quite badly. enabling the LCD engine's FIFO helps a lot, and is done in recent-ish kernels (possibly not in 2.6.31/32 though..?) Jan 14 09:56:48 i think it has already been fixed, i'm trying a little patch from a future revision (in relation to the current used in shr-u) Jan 14 09:56:59 if it will works I will post it... or either shr could update EFL :) Jan 14 09:57:12 spaetz: ok.. I cherry-picked that.. building now.. Jan 14 09:58:55 Weiss: lets check... Jan 14 09:59:39 you can easily check by pastebinning the output of "cat /sys/bus/platform/devices//glamo/regs" :) Jan 14 09:59:43 JaMa: this will fix intone and the silder in idle screen? Jan 14 10:00:10 carcinoma: no Jan 14 10:01:06 carcinoma: its about shr-t and right now intone is pretty down on my list, sorry :/ Jan 14 10:01:49 JaMa: ok, i'l take the image from 06.01. Jan 14 10:02:20 Weiss: not so easy :))) i got several kernel kernel oopses and my cat segfaulted :) Jan 14 10:02:22 carcinoma: intone was better in image from 06.01.? Jan 14 10:02:59 gena2x: ah :( Jan 14 10:03:01 Weiss: so it's better to use plan b and check sources :) Jan 14 10:03:07 carcinoma: then it can be some problem with upgraded mplayer.. but I won't have time to check this now.. would be great if you can provide more info in list or trac Jan 14 10:03:31 JaMa: jepp, i think it could be the slider problem Jan 14 10:03:51 JaMa: i'll test is Jan 14 10:04:13 ah then nothing with mplayer probably.. -> then I'm not involved much.. Jan 14 10:04:53 JaMa: maybe, when its mplayer i create a ticket Jan 14 10:05:11 ok, thanks Jan 14 10:08:47 thanks JaMa Jan 14 10:14:44 JaMa: which kernel branch are you using for build? Jan 14 10:15:58 gena2x: http://git.openembedded.org/cgit.cgi/openembedded/tree/recipes/linux/linux-openmoko-2.6.32_git.bb Jan 14 10:16:19 JaMa: thanks Jan 14 10:17:12 gena2x: those patches used in recipe you can find there too Jan 14 10:38:00 Weiss: well, seem one of drm patches adds FIFO change, in a in bit unclean way. Jan 14 10:39:40 Weiss: addr_high = ((addr >> 16) & 0x7f) | 0x4000; Jan 14 10:40:02 Weiss: reg_write_lcd(gdrm, GLAMO_REG_LCD_A_BASE2, addr_high); brrr... Jan 14 10:41:15 what's wrong with that? :) Jan 14 10:42:27 i guess it will work :) Jan 14 10:45:10 fifo patch is '| 0x4000' Jan 14 10:46:03 oh, it's ok :) Jan 14 10:56:59 error: evas not found in specified Vala API directories or GObject-Introspection GIR directories Jan 14 10:57:15 ooh fuck, I give up on Vala and Openembedded, it's just never working right Jan 14 10:57:35 spaetz, I actually develop with them :D Jan 14 10:58:04 is there an autotools skeleton that I can use? I just can't get it to work right Jan 14 10:58:08 anyway bbl Jan 14 10:58:30 spaetzLunch, for cross compile I use shr bitbake system Jan 14 10:58:44 and then I get manual gcc calls and use them in a custom shell script Jan 14 10:58:47 it's easier... Jan 14 11:03:47 gena2x: yep, it's a very simple thing :).. so you have it in your kernel? Jan 14 12:07:32 evening all Jan 14 12:09:44 Just wondering if many of the openmoko dev's are going to linux.conf.au in NZ this year? Jan 14 12:41:20 mrmoku: spaetz: all shr-t *ubi* images nuked, right ubi params cherry-picked.. build running :) Jan 14 13:04:36 JaMa: thanks. although I doubt that shr-t users want to use ubi. Jan 14 13:04:55 we are a conservative bunch and let you burn your fingers in shr-u before we take something new :) Jan 14 13:05:00 anyway bbl Jan 14 13:07:09 spaetz: I agree, but then you shoudn't build it at all :) so as long as its building to deploy directory its better with right params and at least possibly working ;) Jan 14 13:08:56 JaMa: right. It should probably be disabled on shr-t Jan 14 13:09:08 I never changed anything from the defaults there, I think Jan 14 13:09:44 how do I turn off ubi? Jan 14 13:09:53 in local.conf there is just jffs2 and tar.gz Jan 14 13:09:56 anyway bbl Jan 14 13:10:54 * daniele_athome is recompiling edje rev44950 (shr) with a recent revision of edje_calc.c, for dragables not working issue Jan 14 13:14:25 spaetz: in conf/machine/om-gta0*, but as we have the same kernels.. i think you can let it there now Jan 14 13:16:02 guys toggle issue fixed Jan 14 13:16:08 just 4 lines of code patched in edje Jan 14 13:16:25 daniele_athome: have you read that thread about xserver? :) Jan 14 13:16:37 JaMa, no sorry I forgot it :( Jan 14 13:17:33 ok.. I just wanted to show you, that it's possible and right way to do shr/neo specific changes directly in oe.dev branch Jan 14 13:18:12 daniele_athome: and with koen reply about hal, we talked here about having 3-way switch which would be optimal for all Jan 14 13:20:44 JaMa, ehm can you give me the url again? ;) Jan 14 13:20:52 Weiss: [i'm back from ski tour] really only way is to check regs, and this fails. btw, in regs_read()+0x0/0x134. btw, should i report this fault somewhere?\ Jan 14 13:21:06 (anyway I posted the edje patch on shr-devel, you should trigger a rebuild :) Jan 14 13:21:45 gena2x: oops when catting the Glamo registers on 2.6.31 (possibly .32?) with DRM? Jan 14 13:22:59 or even without DRM? Jan 14 13:23:24 Weiss: with drm, .32.3 Jan 14 13:23:36 ticket #36 closed :) Jan 14 13:24:06 gena2x: yep, that's on my list.. Jan 14 13:24:48 Weiss: should i report to some track / share backtraces? Jan 14 13:27:46 Weiss: I mean will it help somebody somehow? Jan 14 13:28:24 gena2x: if it happens without DRM, it's worth telling larsc as well. but I should be able to fix it either way (and it's on my list already, I just haven't had much time lately) Jan 14 13:28:47 if it's just with DRM, then it's definitely my fault Jan 14 13:30:59 daniele_athome: you lazy daniele! http://www.mail-archive.com/openembedded-devel@lists.openembedded.org/msg02597.html Jan 14 13:31:12 JaMa, ;) Jan 14 13:31:18 daniele_athome: good that you have woman name in nick :) Jan 14 13:31:31 JaMa, is the italian for daniel :S Jan 14 13:31:37 lol Jan 14 13:32:51 daniele_athome: yeah.. I guessed that long ago :/ :P Jan 14 13:33:01 ok... Jan 14 13:33:06 but still better than nothing :) Jan 14 13:33:22 ehm.. i mean that nick :) Jan 14 13:33:30 JaMa, i don't get you Jan 14 13:33:54 nvm.. Jan 14 13:35:48 ... Jan 14 13:36:39 JaMa, I already read that thread Jan 14 13:36:44 I thought there were some new posts Jan 14 13:41:48 someone applies this patch please :) Jan 14 13:41:49 http://lists.shr-project.org/pipermail/shr-devel/2010-January/001900.html Jan 14 13:42:38 * von_fritz prefers to bump EFL :) Jan 14 13:43:29 von_fritz, it's very tiny little patch, 4 lines of edje_calc.c code Jan 14 13:43:44 I mean von_fritz how much time will get passed until you bump EFL? Jan 14 13:44:10 sasasd~. Jan 14 13:44:11 in the meantine, 40% of the user interface is not working Jan 14 13:44:44 daniele_athome NOTE: Running task 8032 of 8062 well 5 minutes :P Jan 14 13:44:59 von_fritz, so you are doing it now? Jan 14 13:45:42 daniele_athome i have bumped EFL at 45111 localy ;) Jan 14 13:46:02 daniele_athome: seen that.. but trying to get back to work I wont apply it soon Jan 14 13:46:15 von_fritz, but what about shr buildhost? Jan 14 13:46:45 sorry i am not a shr dev :) Jan 14 13:47:01 daniele_athome: if EFL bump helps I'll bump it in SHR.. Jan 14 13:47:33 JaMa, can I post libedje binary .so file in shr-devel? Jan 14 13:48:07 you can.. it depends on readers if they trust binaries.. :) Jan 14 13:48:20 ok Jan 14 13:48:43 daniele_athome: good idea Jan 14 13:49:56 JaMa|W, lazy shr devs! :P Jan 14 13:50:55 hmm :/ Jan 14 13:51:14 daniele_athome: and shr buildhost is still building.. Jan 14 13:51:26 JaMa|W, just a little revenge, just joking... Jan 14 13:51:48 daniele_athome: and it's not about being lazy :P Jan 14 13:52:06 nevermind, i don't want to know :P Jan 14 13:52:20 daniele_athome: if your patch is missing in upstream e (and EFL bump won't help) then we(YOU) should try to push it there.. Jan 14 13:52:44 JaMa|W, it is in upstream Jan 14 13:52:56 but in the meantime... Jan 14 13:53:10 JaMa|W, my message has not been posted because of excessive size Jan 14 13:53:17 700k :S Jan 14 13:53:47 daniele_athome: aha, then write me the working EFL rev :) Jan 14 13:54:08 JaMa|W: http://trac.enlightenment.org/e/changeset/45060 Jan 14 13:54:20 as you can see, a very little change Jan 14 13:54:26 von_fritz: or ping me as soon as you can confirm that 45111 is even better Jan 14 13:54:28 HEAD seems to be quite solid after lots of problems even on x86 Jan 14 13:54:57 JaMa|W ok Jan 14 13:58:48 JaMa|W, that's not the actual problem Jan 14 13:59:06 a temporary patch is always useful in the meantine that the system get upgraded Jan 14 13:59:30 for some lower issues I guess there is no need for patched, but for such a (imho) big issue it is necessary Jan 14 13:59:32 always imho Jan 14 14:00:09 s/no need for patched/no need for patches/ Jan 14 14:01:19 sorry I have to finish something at work.. so no time to add additional patch.. all I would be able to do now would be just change 1 number and run one line command on shr buildhost but even pushing that to oe repo takes time because I have few other patches in my queue etc.. :/ Jan 14 14:01:44 if you do it in the next days, it's ok Jan 14 14:01:52 so no need for patches :) Jan 14 14:02:35 at least can you tell me if there is someone here who can approve my post in shr-devel JaMa|W? Jan 14 14:03:11 imho Ainulindale, but better upload it somewhere and send link => that will be faster Jan 14 14:04:47 What? :-) Jan 14 14:05:03 I raised the size limit two times already Jan 14 14:05:11 If it's too big then it's too big :-) Jan 14 14:05:22 Don't force me to stop slacking :-) Jan 14 14:05:29 Ainulindale, what's the problem? Jan 14 14:05:41 Well I don't know, I have none Jan 14 14:05:43 it's a binary patch Jan 14 14:05:52 not an image, or video, or so Jan 14 14:05:54 :S Jan 14 14:06:01 Ainulindale, so approve it... Jan 14 14:06:15 Well you could have mailed shr-devel-owner, you didn't Jan 14 14:06:18 So don't complain Jan 14 14:06:24 And you could, as well, say please Jan 14 14:06:36 That's basic conventions when one's polite Jan 14 14:08:07 daniele_athome: PLEASE upload it somewhere on the Web and send us the link Jan 14 14:08:15 Well you could have mailed shr-devel-owner, you didn't <--- what!? Jan 14 14:08:29 Ainulindale, I guess you are the shr-devel-owner, aren't you? Jan 14 14:09:14 Yep. Jan 14 14:09:31 :) Jan 14 14:09:45 Ainulindale, so I'm talking to you directly, what's the difference? Jan 14 14:10:06 daniele_athome: The difference is that you're not polite, and I don't like that. Start trying, and I'll stop whining. Jan 14 14:10:26 spaetz_: I'm not a psychic but I can foresee I won't be happy in the nearest future Jan 14 14:10:42 * misc give a beer to Ainulindale Jan 14 14:10:48 Ainulindale, never mind... Jan 14 14:11:00 maybe you can't see by your own eyes... Jan 14 14:11:01 misc: w00t :-) Jan 14 14:11:19 Ainulindale: why, what's happening? Jan 14 14:11:22 daniele_athome: Yeah, that must be me, can't be you of course :-> Jan 14 14:11:24 spaetz_: nothing :-) Jan 14 14:11:30 I'm happy, changed jobs, it's great here :-) Jan 14 14:11:30 ahh :) Jan 14 14:11:50 well, that's a reason to be happy then Jan 14 14:12:08 Yep Jan 14 14:12:16 already at the new job ? Jan 14 14:12:32 ok let's say we are different people Ainulindale and close this discussion here... Jan 14 14:12:33 Yep, started on monday Jan 14 14:12:57 daniele_athome: Yes, we're different. I, for instance, say "hi ! could you please ..." instead of giving orders :-) Jan 14 14:13:07 Ainulindale: not bearstech anymore? Jan 14 14:13:12 or stillß Jan 14 14:13:12 Nope Jan 14 14:13:14 Ainulindale, so you really should read what happened before Jan 14 14:13:23 ahh, didn't know that Jan 14 14:13:27 and yesterday, and the day before yesterday, and the day before, and so on... Jan 14 14:13:40 daniele_athome: you posted your patch today at 2h51 Paris time Jan 14 14:13:42 Ainulindale, I assure you that you will lose the patience too Jan 14 14:13:47 That is 20 minutes ago Jan 14 14:14:11 Ainulindale, the patch is only the latest thing Jan 14 14:14:17 And you didn't bother to ask politely, I can't see why I'll bother to listen to you then Jan 14 14:14:57 spaetz_: well for a lot of reasons I decided to quit some time ago Jan 14 14:15:04 spaetz_: and I started monday in a new company :-) Jan 14 14:15:15 exciting Jan 14 14:15:18 Yep Jan 14 14:15:23 good luck then :) Jan 14 14:15:45 Now I just have to get back to my SHR stuff which won't be that easy with my serveal months gap Jan 14 14:15:49 s/serve/sever/ Jan 14 14:15:50 Ainulindale meant: Now I just have to get back to my SHR stuff which won't be that easy with my several months gap Jan 14 14:16:17 At least I kept using it in the meantime :-) Jan 14 14:16:24 cool, welcome back Jan 14 14:16:31 hehe Jan 14 14:16:35 Don't say that, wait until I do something :-p Jan 14 14:17:48 spaetz_: anyways I have 2 GTA02s and 1 GTA01 covered with dust at home Jan 14 14:17:53 I should find some use to them Jan 14 14:26:01 valos, http://lists.shr-project.org/pipermail/shr-devel/2010-January/001901.html Jan 14 14:26:16 posted to filefactory (the first free hoster that I found :) Jan 14 14:26:42 Funny to see you were to proud to even type "please" =) Jan 14 14:27:06 +o Jan 14 14:31:06 daniele_athome: bump estimate is as soon as von_fritz confirms its better with it (no new terrible bug somewhere eles) Jan 14 14:31:32 JaMa|W, ok good :) Jan 14 14:31:54 JaMa|W, i'm just looking through the eyes of a simple user Jan 14 14:32:00 daniele_athome: bumping only 10 revs is safer.. but with time how long it will take to build I guess that von_fritz can confirm it sooner :) Jan 14 14:32:14 I would feel safer if in the meantine there is a patch for temporary fixing an issue :) Jan 14 14:32:35 that's why I posted the binary Jan 14 14:32:51 ok.. and I'll work now :) Jan 14 14:35:22 Ainulindale: hey, good to hear you new job is fun :) Jan 14 14:35:25 +r Jan 14 14:35:30 JaMa|W ...well slider are ok now with EFL_SRCREV ?= "45111" but another little bug .... :( Jan 14 14:35:44 von_fritz, what is it? Jan 14 14:35:55 JaMa|W everytime i press on a button or on the screen keyboard shows up and when i repress button keyboard disappears. :( Jan 14 14:35:55 mrmoku: well I'll see in a month or so if it keeps being that way Jan 14 14:36:00 mrmoku: hopefully it will Jan 14 14:36:11 wow, that is sow Jan 14 14:36:14 oops Jan 14 14:36:17 JaMa|W see http://scap.linuxtogo.org/files/5e122a23386bed8395432adb109fe2f3.png this is very annoying :( Jan 14 14:36:38 von_fritz: ohh, ouch :P Jan 14 14:36:43 These issues with EFL_SRCREV are really annoying Jan 14 14:37:01 They kept bugging us for a year and a half Jan 14 14:37:08 Well, mostly bugging mrmoku though :-) Jan 14 14:37:16 heh Jan 14 14:37:20 yep, but only 8k emails Jan 14 14:37:21 * mrmoku passed bugging to JaMa|W ;) Jan 14 14:37:29 oops, wrong win again... Jan 14 14:37:33 (and spaetz_ for testing :P) Jan 14 14:37:42 * JaMa|W is creating some imunity for that :P Jan 14 14:37:45 mrmoku: delegating is good isn't it? :-) Jan 14 14:37:49 daniele_athome: thanks Jan 14 14:38:02 Ainulindale: yeah, very good :P Jan 14 14:38:04 You have no idea how much time I spent in the beginning into stabilizing this EFL stuff Jan 14 14:38:21 Seriously, a huge tentacular thorn in my ass Jan 14 14:38:33 Just impacts on everything Jan 14 14:38:52 if only we had this ncurses library ... :-> Jan 14 14:39:01 EFl is pretty fricking unstable in that things start and stop working all the time Jan 14 14:39:03 we don't ? :) Jan 14 14:39:07 hehe Jan 14 14:39:18 mrmoku: well I don't. Is there any? Jan 14 14:39:20 SHR, takes the graphics out of your phone Jan 14 14:39:30 no idea... Jan 14 14:39:34 SHR-ncurses Jan 14 14:39:45 spaetz_: well I'd sign for a phone like that, I still have bits of code to create mouse aware ncurses window with libvte Jan 14 14:39:54 at least there would be a networkmanager console client ;) Jan 14 14:39:57 Shaped into an old libframeworkd-phonegui-ncurses library Jan 14 14:40:00 With no logic though Jan 14 14:40:05 mrmoku: true :) Jan 14 14:40:33 As for EFL, seriously guys, we need QA procedures for EFL bumps Jan 14 14:40:51 We're still at the same point we were 6 months ago regarding EFL Jan 14 14:40:51 btw, don´t put much hope in me getting a connman frontend out of the door. i got so frustrated by vala and autotools that i already stopped again ... Jan 14 14:41:01 spaetz_: heh :-) Jan 14 14:41:51 spaetz_: ohh... ouch... how's your c++? ;) Jan 14 14:42:13 mrmoku: I though PROLOG was the way for SHR now... :-> Jan 14 14:42:37 Ainulindale: your way :P Jan 14 14:42:40 (I'm reaaaaally glad we forgot all about that crappy stuff :-p) Jan 14 14:42:54 Nah, you'll never see me code using PROLOG for that kind of software Jan 14 14:43:00 hehe, i have a c++ book in my shelve Jan 14 14:43:04 Mainly because nobody knows how to properly develop using it Jan 14 14:43:16 * mrmoku never wrote a single line of prolog... Jan 14 14:43:29 (sadly, if we look at the realistic side, nobody knows vala either) Jan 14 14:43:35 running the UI in emacs Jan 14 14:43:39 (but hey, that's my himble opinion :-) ) Jan 14 14:43:43 s/hi/Hu/ Jan 14 14:43:45 Ainulindale meant: (but hey, that's my Humble opinion :-) ) Jan 14 14:43:45 shell scripting rules ;) Jan 14 14:43:59 mrmoku: yeah we should recode all the UI stuff in perl Jan 14 14:44:10 that's a well known, easily readable language :-> Jan 14 14:44:17 heh :P Jan 14 15:17:24 ~logs Jan 14 15:17:25 All conversations are logged to http://ibot.rikers.org/channel, where "channel" is replaced by the URL-encoded channel name, such as %23freenode for #freenode. Lines starting with spaces are not logged. Jan 14 15:18:45 !logs Jan 14 15:18:47 Channel logs for #openmoko-cdevel are archived at: Jan 14 15:18:48 http://hentges.net/tmp/logs/irc/%23openmoko-cdevel Jan 14 15:18:49 Live-logs are available at Jan 14 15:18:51 http://hentges.net/tmp/logs/irc/livelogs/%23openmoko-cdevel.livelog Jan 14 15:18:52 See ?? help-logs for usage instructions Jan 14 15:20:40 max_posedon, I don't have access, it's an fso repo, mrmoku` is the man :) Jan 14 15:24:31 spaetz: what's wrong w/ vala and autotools? can i help? Jan 14 15:38:08 hmm, 2.6.32.3 does not boot from SD here Jan 14 15:38:19 always reports ext2 errors even on a freshly formatted SD Jan 14 15:42:18 mickeyl: try kernel options Jan 14 15:42:25 such as... ? Jan 14 15:42:33 mickeyl: i had same problem recent. Jan 14 15:43:06 mickeyl: glamo_mci.sd_drive=2 glamo_mci.sd_max_clk=10000000 Jan 14 15:43:22 mickeyl: i used this one. dont know if this is optimal Jan 14 15:44:20 valos, 403 forbidden :D Jan 14 15:44:50 mickeyl: I don't know. perhaps. I copied my autotools stuff from libefl anyway, so you should be familiar with it :) Jan 14 15:44:50 gena2x: okies, trying... Jan 14 15:45:07 daniele_athome: yes, fixed now, thx Jan 14 15:45:56 mickeyl: http://dpaste.com/145119/ Jan 14 15:46:02 error: evas not found in specified Vala API directories or GObject-Introspection GIR directories Jan 14 15:46:13 it works fine when compiling locally o my host Jan 14 15:46:27 but when compiling with OE, it complains about that Jan 14 15:46:33 spaetz, you need to pass --vapidir I guess Jan 14 15:47:13 and indeedand, indeed: ls /home/spaetz/src/shrbuildenv/shr-unstable/tmp/staging/armv4t-oe-linux-gnueabi/usr/share/vala/vapi/ only contains Jan 14 15:47:23 fso-glib-1.0.deps fso-glib-1.0.vapi gee-1.0.vapi Jan 14 15:47:31 so, elm.vapi etc are not in staging Jan 14 15:47:48 daniele_athome: that's what I figured, but where should I point the vapidir? Jan 14 15:48:01 hmm Jan 14 15:48:04 in OE certainly not to my /usr/local/ stuff. Jan 14 15:48:05 let me check the recipe Jan 14 15:48:44 anyone here willing to test an improved tangogps? Jan 14 15:48:46 spaetz, try in i686-linux Jan 14 15:48:51 who should install those .vapi files (the elm ones) Jan 14 15:48:53 accumulator, that is? :) Jan 14 15:49:00 spaetz, libeflvala Jan 14 15:49:02 ? Jan 14 15:49:11 accumulator, what do you mean by "improved"? Jan 14 15:49:14 mmh, ok, let me clean and reinstall libefl Jan 14 15:49:17 libeflvala Jan 14 15:49:20 speed mainly Jan 14 15:49:27 spaetz: strange, it should work, since vala.bbclass stages the .vapi and .deps into the appropriate dir Jan 14 15:49:29 now feels very snappy on the FR Jan 14 15:49:30 spaetz, first look also in i686-linux Jan 14 15:49:33 hmm Jan 14 15:49:34 I mean staging/i686-linux Jan 14 15:49:35 no go with the kernel Jan 14 15:49:38 * mickeyl tries another SD Jan 14 15:49:43 around 50KB patch ;) Jan 14 15:49:46 daniele_athome: will do Jan 14 15:49:54 accumulator, can we see it? Jan 14 15:50:03 sure : http://3v8.net/~sander/openmoko/tangogps/tangogps_0.99.2-r2.4_armv4t.ipk Jan 14 15:50:44 DOH! Jan 14 15:50:49 ow, before I forget, based on SHR-T Jan 14 15:50:53 * spaetz feels very stupid now. Jan 14 15:50:58 accumulator, source code please :) Jan 14 15:51:08 indeed the staging/x86_64-linux/usr/share/vala/vapi is full of stuff Jan 14 15:51:18 and that makes sense as I use vala-native ! Jan 14 15:51:18 I can put the patches online, but they're quite big Jan 14 15:51:35 so, I need to point my vapidir there Jan 14 15:51:45 accumulator, no problems, I'm curious to look at source code :) Jan 14 15:51:52 what variables can I use to get that dir? Jan 14 15:51:55 spaetz, :) Jan 14 15:52:01 spaetz, variables? Jan 14 15:52:05 sorry, I am really still an OE/autotools noob Jan 14 15:52:21 ah sorry spaetz I don't know Jan 14 15:52:21 yes, obviously I don't want to hardcode that path in my Makefile.am Jan 14 15:52:31 I manually cross-compile, I do not use autotools to cross-compile Jan 14 15:52:31 daniele_athome: check the same URL, without the filename Jan 14 15:52:50 daniele_athome: 2 patches there, first should be patchset99 Jan 14 15:52:56 all the libeflvala examples only use $topdir/vapi :-) Jan 14 15:52:58 thanks accumulator Jan 14 15:53:02 daniele_athome: diff relative to 1.99.2 Jan 14 15:53:11 spaetz: are you compiling for cross or nateive? Jan 14 15:53:15 spaetz, I got crazy trying to do it by using autotools Jan 14 15:53:26 compiling for arm Jan 14 15:53:27 vala.bbclass copies all stuff from x86 to arm directory Jan 14 15:53:29 so the paths are ok Jan 14 15:53:35 eventually I fall down to shr/bitbake Jan 14 15:53:39 after do_stage they need to be in both dirs Jan 14 15:53:44 if they aren't then it's a bug Jan 14 15:53:55 mickeyl: they are not, I guess Jan 14 15:54:03 doh. finished moving. Now for home improvement :-) Jan 14 15:54:06 still reinstalling libeflvala Jan 14 15:54:39 hmm, ok, 2.6.32.3 doesn't like me Jan 14 15:54:40 haven't updated my build env for quite some time. Will let you know if it works now Jan 14 15:54:46 what kernel are you guys recommending these days? Jan 14 15:54:49 spaetz: oki Jan 14 15:55:33 oooh, 2.6.32? does that mean fso2 will work with that eventually? Jan 14 15:55:42 * spaetz dances a happy dance of happiness :) Jan 14 15:55:49 no Jan 14 15:55:51 since i can't boot Jan 14 15:55:55 mickeyl: .29 with nodebug and nopreempt :) Jan 14 15:56:00 hmm, k Jan 14 15:56:05 so back to the old and trusted Jan 14 15:56:40 it depend on that do you want to do i think Jan 14 15:56:45 well Jan 14 15:56:50 i had hoped to fix ogsmd for .32 Jan 14 15:56:53 and to get rid of udev Jan 14 15:57:07 but that will not work without .32 Jan 14 15:57:14 so seem you need .32 :) Jan 14 15:57:44 yep Jan 14 15:58:31 :-( Jan 14 15:58:35 would have been nice Jan 14 15:59:05 udev takes about 30 seconds in bootchart which certainly does not look right. Jan 14 15:59:25 let's remove udev!!!! :D Jan 14 15:59:45 daniele_athome: some time ago, mrmoku tried mdev Jan 14 15:59:56 but with little success, if I remember right Jan 14 15:59:58 cool... Jan 14 16:00:11 it's mmc, and the missing rfkills for gps and gsm iirc Jan 14 16:00:14 spaetz: with nodebug whole boot is little bigger than minute Jan 14 16:00:24 should be the last blockers for .32 Jan 14 16:00:30 gena2x: looking forward to those improvements Jan 14 16:00:41 mickeyl: and the cannot suspend issue afaik Jan 14 16:00:50 right Jan 14 16:00:54 but from fso side, cool that you are close to ready Jan 14 16:00:58 spaetz: in fact .32 have all this enabled/disabled by default Jan 14 16:01:27 gena2x: that might be the reason why 2.6.32 seemed so much damn faster when we tried it Jan 14 16:01:40 spaetz: yes. only reason in fact Jan 14 16:02:14 spaetz: .29 have almost same speed Jan 14 16:02:29 gena2x: as the maintainer of shr-testing I am willing to cut out some more debug options there than we can in shr-unstable, where debug info will often be useful Jan 14 16:02:55 if you let me know which options to turn off, I could do that there. Jan 14 16:04:29 spaetz: i can prodive diff to defconfig Jan 14 16:04:44 spaetz: if you mean kernel Jan 14 16:04:50 yep Jan 14 16:05:21 a diff would be good Jan 14 16:06:17 spaetz: ok, moment Jan 14 16:12:25 default IO scheduler is "deadline". does that make sense on NAND/sd card? Jan 14 16:12:42 isn't stupid "noop" better? Jan 14 16:12:55 spaetz: in fact, because of speed change some bugs revealed - people reported this one: https://docs.openmoko.org/trac/ticket/2327 Jan 14 16:13:10 spaetz: is it ok from you pov? Jan 14 16:13:50 freesmartphone.org: 03mickey 07cornucopia * r6c69d213d8fd 10/tools/serial_forward/ (19 files in 2 dirs): tools: autotoolize serial_forward Jan 14 16:14:31 outch, non-working WiFi is bad. I do rely on wifi for updating etc Jan 14 16:14:49 spaetz: not strictly non-working Jan 14 16:14:51 probably uncovered some race condition... Jan 14 16:15:05 well, randomly failing is just as bad :) Jan 14 16:15:50 spaetz: :( Jan 14 16:16:04 spaetz: but all other things seem ok for people. Jan 14 16:16:08 but let's try anyway Jan 14 16:16:15 I can always revert Jan 14 16:16:21 spaetz: and 2x speed may worth it. Jan 14 16:19:10 hm, is it possible to run SHR not as root? Jan 14 16:19:18 is OABI needed on GTA02 ? Jan 14 16:19:25 рь Jan 14 16:19:26 hm Jan 14 16:19:30 WARNING: Erorr opening /usr/share/libphone-ui-shr/config Jan 14 16:19:37 phoneuid.log Jan 14 16:19:59 ThibG: try it, there are a few issue apparently. If you find things, report to trac so we can fix them. Jan 14 16:20:10 ok Jan 14 16:20:11 ThibG: we would like to switch away from being root in the long run Jan 14 16:21:35 I'll try in a few days I think Jan 14 16:21:48 as I want to install debian and share /home with SHR Jan 14 16:22:05 mickeyl: have you tried to increase rootdelay for .32? Jan 14 16:22:20 mickeyl: here it boots from uSD only for rootdelay=7 or longer Jan 14 16:22:47 ok, let me try that Jan 14 16:23:04 mickeyl: or with debug from Qi (with POWER buttong) Jan 14 16:23:22 well, i have loglevel=8 Jan 14 16:23:28 that doesn't buy me anything though Jan 14 16:23:35 as i just see it hanging after mounting rootfs Jan 14 16:24:11 AUX lock don't work also Jan 14 16:24:31 Q-Master: harmless warning Jan 14 16:24:48 config in libphone-ui-shr is only for theming (and that is not completely implemented) Jan 14 16:24:58 what is CONFIG_DAB ? Isn't that DAB radio stuff? Jan 14 16:25:07 why do we have that enabled on the FR? Jan 14 16:25:31 why not? Jan 14 16:25:33 mrmoku`: and AUX? seems that it is not automatically locking Jan 14 16:25:35 as long as it's a moudle.. Jan 14 16:25:45 spaetz: we also have HZ==200 Jan 14 16:26:07 spaetz: slow fp kernel emulation (shr uses -msoftfloat) Jan 14 16:26:16 mickeyl: CONFIG_DAB=y Jan 14 16:26:23 doesn't look very module like Jan 14 16:26:44 spaetz: s/ (shr uses -msoftfloat) / (shr uses -msoftfloat?) / Jan 14 16:27:48 mickeyl: especially as the DAB_USB drivers are disabled, so its pretty useless anyway Jan 14 16:28:01 true Jan 14 16:28:13 gena2x: don't ask me. I am no kernel guy and I have never looked at the FR kernel config before, basically Jan 14 16:28:27 Q-Master: no idea why AUX would not work anymore... Jan 14 16:28:31 spaetz: i'm not asking :) Jan 14 16:29:01 gena2x: ahh :) Jan 14 16:29:16 spaetz: i'm telling, that if you are looking to config, we have other interstion things where Jan 14 16:29:47 gena2x: nothing we can do about slow fp emulation, can we? It just has not fast hardware for that... Jan 14 16:29:52 s/not/no( Jan 14 16:29:56 s/not/no/ Jan 14 16:29:57 spaetz meant: s/no/no( Jan 14 16:30:03 bah Jan 14 16:30:10 spaetz: in kernel, 2 engines - one fast and one preciese :) Jan 14 16:30:23 spaetz: we now not using fast Jan 14 16:31:33 spaetz: but all this applies only for programs not build with -msoft-float Jan 14 16:31:41 mmh, I have no idea how bad it is to use fast but non-precise Jan 14 16:32:05 we could kick out OSS as well, and only leave in the alsa oss emulation Jan 14 16:32:11 did anyone tried an EFL build w/ fixed point? Jan 14 16:32:25 should bump performance a bit on our CPU Jan 14 16:32:36 but that is obviously nothing to try out for shr-testing :) Jan 14 16:33:00 much more interesting, does -msoft-float is used in current compilation procedure Jan 14 16:33:17 gena2x: I don't think it's a standard CFLAG, no Jan 14 16:33:49 spaetz: think it will extemly good for performance to enable this Jan 14 16:34:16 freesmartphone.org: 03mickey 07cornucopia * r4662b55fdba5 10/tools/serial_forward/ (.gitignore src/forward.c src/pty_forward.c): Jan 14 16:34:16 freesmartphone.org: tools: serial_forward: remove default arguments in favour of explit arguments Jan 14 16:34:16 freesmartphone.org: (necessary to forward multiple device nodes) Jan 14 16:34:35 gena2x: I'm not the one who is gonna try that, but it would be cool if someone did test this Jan 14 16:34:51 spaetz: i'm planning....... Jan 14 16:40:03 mickeyl: I rebuild libeflvala and the .vapi files are in arm staging now. Jan 14 16:40:11 Sorry for the noise... Jan 14 16:40:38 np, glad to hear Jan 14 16:42:35 shouldn't the vala class stage it into both arches? Jan 14 16:43:00 playya_: yes, I previously only had them in my x86-64 staging Jan 14 16:43:07 and on rebuilding they are in both now Jan 14 16:43:20 don't ask me why Jan 14 16:44:05 ok. brb. i need some beer for my schnweinshaxe Jan 14 16:51:35 gena2x: btw, thanks a lot for the measurements ! i'll have to plot them to get a better impression. some thing look quite interesting. Jan 14 16:53:52 wpwrak: i'm glad to help especially if this were interesting... and challenging in unexpected way. Jan 14 17:00:39 gena2x: seems that a number of options have exctly the same (bad) performance impact. so there must either be something common underneath (e.g., they all enable another option) or they manage to trigger the same effect in different ways (e..g, by crowding a cache) Jan 14 17:05:15 wpwrak: I think this is wrong way to use microbenchmark. Only way to understand this in true depth is to check kernel code. I think all this options are quite easy to understand. If in doubt or just to stop guesses, I think it's possible to profile kernel, but this may not reveal important impact on things like interrupt handlers. Jan 14 17:07:26 gena2x: did you look at the wifi bug that surfaced? ;) Jan 14 17:07:47 wpwrak: any attempt to accaunt something uses cache, i don't know cache configuration on our cpu in details, but even 4 bytes is 0.25% of data cache. Jan 14 17:08:31 wpwrak: as we have 1024 as far as i can recall. Jan 14 17:10:20 lindi-: if i'll start to fix bugs in areas unknown for me (complicated), i'll never finish other performace tests or other simple tests Jan 14 17:11:09 lindi-: don't recall - i'm not kernel developer. only developed driver once. Jan 14 17:11:28 lindi-: s/recall/forget/ Jan 14 17:12:12 gena2x: just asking :) Jan 14 17:12:49 lindi-: i will to try, but after all that relatevely simple things i planned. Jan 14 17:30:32 Heinervdm, what's the name of your protocol? Jan 14 17:30:46 playya_: PBAP? Jan 14 17:31:06 ok. thx Jan 14 17:36:33 gena2x: (microbench) i'm want them as an indicator where things are happening. particularly if you have an upper and a lower bound to compare to, they're not so bad. granted, you can still end up with cases where they don't tell you anything. but then you know at least where to look. Jan 14 17:37:52 gena2x: (cache) 64 way set-assoc, 16 kB i-cache, 16 kB d-cache. 8 words per cache line. wt or wb. (see also um_s3c2442b_rev12.pdf, page 6) Jan 14 18:06:53 mrmoku: TAsn: mplayer seems to be broken in shr-U Jan 14 18:08:05 alexxy: symptoms? Jan 14 18:08:36 root@om-gta02 ~ $ ldd /usr/bin/mplayer Jan 14 18:08:36 ..... Jan 14 18:08:44 libbz2.so.1 => not found Jan 14 18:08:48 =) Jan 14 18:09:18 so seems some deps are missing Jan 14 18:12:43 anyone here willing to test an improved tangogps? Jan 14 18:18:09 spaetz: here is config diff against gta02_packaging_defconfig, but paths are wrong in diff :(, full config, built kernel with modules. http://www.bsdmn.com/openmoko/andy_nodebug_nopreempt/ Jan 14 18:19:08 accumulator: improved how? Jan 14 18:19:19 speed Jan 14 18:19:25 maily.. Jan 14 18:19:28 and bugs solved Jan 14 18:19:53 accumulator: post the patch; I'd love to look at it tonight Jan 14 18:20:10 spaetz: spent some trying to generate oops to check for backtraces, but end up in checking code - all seems in place Jan 14 18:20:33 budfive: ok, patches and ipk are here : http://3v8.net/~sander/openmoko/tangogps Jan 14 18:20:50 accumulator: thanks! Jan 14 18:21:44 wpwrak: so, 0,025% is best case :) Jan 14 18:22:04 budfive: I just updated the patches, maybe reload to be sure Jan 14 18:23:52 alexxy: ohh, fallout from the bz2 bump :| Jan 14 18:25:11 mrmoku: =) Jan 14 18:25:25 also i think not only mplayer failng Jan 14 18:25:27 =) Jan 14 18:30:37 spaetz: s/spent some trying/spent some time trying/ Jan 14 18:30:40 accumulator: works nicely here Jan 14 18:31:04 faster? Jan 14 18:31:11 should be :) Jan 14 18:31:12 accumulator: is there anything in particular that i should try Jan 14 18:31:21 yeah, I *think* so Jan 14 18:31:24 yeah, scroll outside map boundary Jan 14 18:31:41 and try as many 'overlays' as possible, like friends, tracks, poi etc Jan 14 18:33:02 I'm also using the nodebug kernel (with preempt though), but the speedup is really noticable Jan 14 18:33:13 no more waits after each map-drag Jan 14 18:33:41 accumulator: we a talking about it for a week :( Jan 14 18:33:54 gena2x: about what? Jan 14 18:34:00 gena2x: the nodebug kernel? Jan 14 18:34:10 I'm using the stock shr kernel and can still scroll into white areas Jan 14 18:34:24 teythoon: yes that's ok Jan 14 18:34:27 k Jan 14 18:34:53 I had screen corruption before, when dragging into white areas, but I think it's solved now Jan 14 18:34:57 accumulator: yes, how social, phycoligical and weather on mars aspects are related to this problem. Jan 14 18:35:09 gena2x: hehehe Jan 14 18:35:20 gena2x: too deep for me :) Jan 14 18:35:34 gena2x: is it a problem? Jan 14 18:35:53 accumulator: yes Jan 14 18:36:29 gena2x: why? I'm using nodebug for a week now, without any problems. made my own nodebug patch though Jan 14 18:37:09 accumulator: everyone made their own :) Jan 14 18:37:56 accumulator: i did first in november. Jan 14 18:38:06 gena2x: I mean, instructions were posted on the mailing list Jan 14 18:42:25 spaetz: do you still want to apply nodebug/nopreempt patches? Jan 14 18:46:33 freesmartphone.org: 03mickey 07cornucopia * r16f1a9691a5e 10/fsogsmd/src/lib/ (modem.vala muxppp.vala pdp.vala): fsogsmd: refactor common ppp logic into AtPdpHandler; derive MuxPppPdpHandler from it Jan 14 18:48:51 max_posedon: serial_forward is now using autotools Jan 14 18:59:23 has anyone yet flashed the new image? Jan 14 18:59:25 ~hail autotools Jan 14 18:59:26 * infobot bows down to autotools and chants, "I'M NOT WORTHY!!" Jan 14 19:00:00 foundation of universe Jan 14 19:25:01 mrmoku: TAsn : also seems something strange with default shr illume theme Jan 14 19:25:24 alexxy: ? Jan 14 19:25:28 in past versions there were two buttons messages and contacts Jan 14 19:25:35 now there is line =) Jan 14 19:25:41 of buttons Jan 14 19:26:11 ahh, yes Jan 14 19:26:20 known (unfixed) problem... Jan 14 19:35:54 freesmartphone.org: 03mickey 07cornucopia * rc5130714b3da 10/fsodeviced/ (7 files in 4 dirs): fsodeviced: new audio router plugin: router_qdsp5 Jan 14 19:38:02 gena2x: thanks for the patch. Will apply some or all of it Jan 14 19:53:12 freesmartphone.org: 03morphis 07msmcomm * re3056d7db567 10/msmcommd/src/relay.c: msmcommd: use default port for network relay interface Jan 14 19:53:13 freesmartphone.org: 03morphis 07msmcomm * r5611295d1585 10/msmcommd/ (5 files in 2 dirs): msmcommd: add seq number handling - need to proof in reality Jan 14 20:22:28 TAsn, I would like use g_warn_if_fail instead g_assert Jan 14 20:22:34 will you happy with this? Jan 14 20:22:45 gena2x: echo l > /proc/sysrq-trigger Jan 14 20:23:51 max_posedon, kinda. Jan 14 20:23:57 I'd also like you to exit the function Jan 14 20:24:24 and in this case, you should probably just you a simple if and a g_error Jan 14 20:24:33 actually, an error is probably better than a warning Jan 14 20:24:47 and I think (hope) that we don't exist on errors Jan 14 20:24:51 just on criticals Jan 14 20:25:20 so, g_return_if_fail ? Jan 14 20:25:35 it logs and returns Jan 14 20:26:03 and what about cleanup Jan 14 20:26:03 ? Jan 14 20:26:04 :P Jan 14 20:26:06 just Jan 14 20:26:21 if (expr) { g_error(); cleanup; return; } Jan 14 20:26:49 what cleanup do?) Jan 14 20:26:56 depends on the function Jan 14 20:27:05 anyhow, we prefer to have the ability Jan 14 20:27:13 to control whether it's an error/warning/etc Jan 14 20:27:18 why are you so against it ? :P Jan 14 20:27:24 PaulFertser: very nice, thanks. Jan 14 20:27:45 because I'm doing little bit other things, just fixing warnings Jan 14 20:28:04 and I think most of that functions just *can't* fail in regular system Jan 14 20:28:16 so use the return_if_fails Jan 14 20:28:19 gena2x: not sure it works on FR though, at least with andy-tracking l doesn't (confused). Jan 14 20:28:26 worst case we'll change it if we ever find a reason :P Jan 14 20:28:34 freesmartphone.org: 03mickey 07cornucopia * r1f9837a0b50d 10/fsodeviced/src/plugins/audio/plugin.vala: fsodeviced: allow router_type = qdsp5 Jan 14 20:28:34 ok, nice, I'm happy now Jan 14 20:28:45 PaulFertser: yes outputs 'HELP' :) Jan 14 20:29:46 PaulFertser: but backtraces are ok Jan 14 20:30:14 btw, _vibration_on return 0; on success, so, I think I should return -1 on error, (if it will ever present), am I? Jan 14 20:30:44 or 1 Jan 14 20:30:52 PaulFertser: t works. Jan 14 20:31:12 max_posedon, just return a non-zero Jan 14 20:31:17 that'll be "good enough" Jan 14 20:32:30 gena2x: l is available only when SMP set Jan 14 20:33:50 PaulFertser: Nice, we have no SMP set ! ;) Jan 14 20:40:10 TAsn, done, plz apply Jan 14 20:40:32 I bet I will never use g_assert for this project anymore) Jan 14 20:40:41 hehe, thanks :P Jan 14 20:43:10 done Jan 14 20:43:28 SHR: 03maxposedon 07libphone-ui * re4b2e91a1d17 10/src/phoneui-utils-device.c: Use g_return_if_fail instead g_assert for fixing gcc-4.4 warnings Jan 14 20:43:44 ffs Jan 14 20:43:50 did you even try to compile it? Jan 14 20:43:59 g_return_if_fail(rc == 0); Jan 14 20:44:03 what about a , 1 ? :P Jan 14 20:44:16 will fix Jan 14 20:44:18 :| Jan 14 20:44:22 you forgot almost everywhere. Jan 14 20:44:53 oh Jan 14 20:44:53 voids Jan 14 20:44:53 forget about it Jan 14 20:44:56 I'm an idiot :P Jan 14 20:44:57 )) Jan 14 20:45:09 you are not, but yes, I tried compile it) Jan 14 20:45:50 Anyhow, sorry for being so reckless Jan 14 20:46:02 :P Jan 14 21:13:41 hi does someone knows libefso? Jan 14 21:13:45 any comment on that: http://pastebin.com/m2939a574 Jan 14 21:13:50 and on missing includes etc... Jan 14 21:14:05 I'll look if it's still maintained Jan 14 21:15:07 perror and etc, is common when you switching to gcc-4.4 Jan 14 21:15:30 ok Jan 14 21:15:49 or may be even gcc-4.3, it doesn't mean code not mainteiner, just developers don't follow recent compilers Jan 14 21:16:01 ok Jan 14 21:17:36 mmm Jan 14 21:17:43 I'll look at the svn info Jan 14 21:18:54 Last Changed Date: 2009-03-01 22:12:43 +0100 (Sun, 01 Mar 2009) Jan 14 21:18:57 it's not a matter of compilers, it just probably means the code is not perfect (missing includes, etc) Jan 14 21:19:26 maybe I should write a mail then for asking if no one object Jan 14 21:20:10 you also can do git-svn + any fixes you want Jan 14 21:21:43 I know but I was in a hurry Jan 14 21:21:55 git svn download all the revs Jan 14 21:23:56 I think there not so much revs, it isn't linux kernel Jan 14 21:31:22 ok Jan 14 21:57:28 'night Jan 14 22:11:07 what is the dbus invocstion for a contacts screen? Jan 14 22:42:46 booooooooooOOOOOO Jan 14 22:43:06 sorry too quiet Jan 14 22:43:29 :)) Jan 14 23:12:00 mrmoku|away, TAsn [shr-specs] warning xml dir doesn't contains recent xml's Jan 14 23:12:22 of course I'll run make check, but I think its a bug to have inconsist version Jan 14 23:24:07 max_posedon: hehe nice. prepare for impact everybody Jan 14 23:28:42 mrmoku|away, TAsn [shr-specs] what about finally deside where is the place for it in FS, and allow all use it? It will be nice have shr-1.0.pc which explains where is xml installed Jan 14 23:29:09 DocScrutinizer51, I just want do my little things) Jan 14 23:29:54 max_posedon, that's the only sane way to do it :P Jan 14 23:29:56 We should. Jan 14 23:29:56 mrmoku|away, TAsn also it will allows have shr-specs in deps, and check availability xmls on configure stage Jan 14 23:30:09 Yes. I agree. Jan 14 23:30:38 Anyhow, I'm off. Jan 14 23:30:50 I'll try make all shr compiles with hardcoded path... later may be do it myself Jan 14 23:52:39 TAsn: Jan 14 23:53:20 go surfing Jan 14 23:54:19 :) Jan 15 00:13:58 Hi, anyone familiar with edje/embryo here? Jan 15 00:14:46 I am having some problems with my OM application gui Jan 15 00:39:00 they should be here later Jan 15 00:46:26 mrmoku|away, TAsn shr-devel have one more patch for phoneuid Jan 15 00:46:35 gcc-4.4 warning/error, as usual) Jan 15 00:47:02 can someone confirm that if you Jan 15 00:47:15 turn gsm off in shr-settings Jan 15 00:47:21 and suspend the phone, Jan 15 00:47:39 when you unsuspend it, gsm reeablesitself? Jan 15 00:48:55 was great when trying to turm gsm off on a recent flight! - was a surprise. Jan 15 00:51:25 mrmoku|away, libframeworkd-glib, phonefsod, phoneui-apps, phoneuid, shr-specs, libphone-ui, libphone-ui-shr,libphone-utils ebuilds done! Jan 15 00:51:36 so, basically shr packaged to ebuilds Jan 15 00:51:45 max_posedon: good job! Jan 15 00:51:52 alexxy, you can try it!) Jan 15 00:52:12 JaMa|W, thanks, may be I'll able create some "image" soon Jan 15 00:52:22 gentoo can be nice dev env Jan 15 00:52:46 no it costs zero time install any fso/shr on PC/laptop and start develop Jan 15 01:01:38 max_posedon: and you can cross compile it or are you building on neo? Jan 15 01:03:31 ok..bed now.. bye Jan 15 01:11:12 I'm native compile on sheevaplug Jan 15 02:05:15 http://www.youtube.com/watch?v=eJN_uWaVRfo&NR=1&feature=fvwp Jan 15 02:29:45 http://www.youtube.com/watch?v=vQ6vl49r67I&feature=quicklist&playnext=3&playnext_from=QL **** ENDING LOGGING AT Fri Jan 15 02:59:57 2010