**** BEGIN LOGGING AT Tue Oct 19 02:59:57 2010
Oct 19 04:54:47 <CIA-102> freesmartphone.org: 03Frederik.Sdun 07cornucopia * rdb8e51635311 10/tools/mdbus2/src/main.vala: mdbus2: Add missing dbus type 'x'
Oct 19 07:09:04 <daniele_athome> morning
Oct 19 07:24:21 <daniele_athome> why people don't answer in ML using reply all :(
Oct 19 07:24:21 <daniele_athome> uff
Oct 19 07:29:51 * JaMa|Zzz always does, even when some recipients complained why..
Oct 19 07:34:28 <daniele_athome> JaMa|Zzz, do always what? reply or reply all? :P
Oct 19 07:34:36 <daniele_athome> (aren't you slepping btw? :P )
Oct 19 07:34:43 <playya__> i used reply all
Oct 19 07:34:56 <daniele_athome> playya__, of course not you i'm talking about people who doesn't :D
Oct 19 07:35:15 <daniele_athome> the problem is that once just 1 person do reply, all other rcpts are lost
Oct 19 07:35:26 <daniele_athome> so the next person will not be able to reply all correctly
Oct 19 07:35:27 <daniele_athome> :)
Oct 19 07:35:30 <playya__> about aytana: e.g. mokowm could implement the dbusmenu thingy
Oct 19 07:35:39 <daniele_athome> mmm
Oct 19 07:36:18 <daniele_athome> playya__, please make an example
Oct 19 07:36:23 <daniele_athome> a practical one
Oct 19 07:36:53 <playya__> start an ubuntu live cd and take a look into the panel
Oct 19 07:37:01 <daniele_athome> playya__, i have ubuntu installed
Oct 19 07:37:05 <daniele_athome> it says indicator-applet
Oct 19 07:37:06 <JaMa|Zzz> daniele_athome: reply-all
Oct 19 07:37:12 <playya__> yes
Oct 19 07:37:15 <daniele_athome> JaMa|Zzz, don't worry just joking ;)
Oct 19 07:37:17 <JaMa|Zzz> no I'm not sleeping at work, right :)
Oct 19 07:37:32 * JaMa|Wrk is killing all openmoko recipes instead
Oct 19 07:37:43 <playya__> in the upper right corner, that's the indicator applet which uses the dbus menu
Oct 19 07:38:20 <JaMa|Wrk> feed almost done :) NOTE: Running task 18548 of 19279
Oct 19 07:38:33 <playya__> wow
Oct 19 07:39:02 <playya__> btw. i still have problems with gir
Oct 19 07:39:02 <daniele_athome> playya__, i don't think menus can be useful in a phone
Oct 19 07:39:11 <playya__> why?
Oct 19 07:39:13 <daniele_athome> i mean menus as gui widget should be dropped
Oct 19 07:39:23 <daniele_athome> because it's invasive and often too small
Oct 19 07:39:27 <daniele_athome> and bad integrated with the rest
Oct 19 07:39:41 <playya__> if you can add a icon which opens a fullscreen window with the options?
Oct 19 07:39:56 <daniele_athome> playya__, might be the answer :)
Oct 19 07:40:13 <daniele_athome> so you think i should implement dbus menu in mokopanel
Oct 19 07:40:16 <playya__> that's imho a big plus for dbus menu
Oct 19 07:40:49 <playya__> you only provide a description of the menu and the implementation decides a generic look
Oct 19 07:41:09 <playya__> s/decides/shows/
Oct 19 07:41:10 <apt> playya__ meant: you only provide a description of the menu and the implementation shows a generic look
Oct 19 07:42:28 <daniele_athome> ok
Oct 19 07:42:36 <daniele_athome> for now i will focus on the split thing and on notifications support
Oct 19 07:42:50 <daniele_athome> i would like to do some extensible stuff in mokohome
Oct 19 07:42:52 <daniele_athome> to write custom widgets
Oct 19 07:46:33 <[Rui]> JaMa|Wrk: awesome news (WRT tslib) :)
Oct 19 07:47:01 <daniele_athome> hey [Rui] 
Oct 19 07:48:23 <[Rui]> hi
Oct 19 07:48:44 <[Rui]> still so sleepy...
Oct 19 07:48:49 <daniele_athome> eheh
Oct 19 07:48:50 <daniele_athome> me too :S
Oct 19 08:12:04 <JaMa|Wrk> [Rui]: still building image.. :) NOTE: Running task 18763 of 19279
Oct 19 08:12:09 <JaMa|Wrk> +feed
Oct 19 08:21:53 <daniele_athome> TAsn, good morning
Oct 19 08:47:46 <JaMa|Wrk> [Rui]: you can upgrade for tslib nos
Oct 19 08:47:50 <JaMa|Wrk> s/nos/now
Oct 19 09:08:33 <daniele_athome> TAsn, i've sent you the entire patchset :)
Oct 19 09:55:15 <Azog> will there be a new shr-unstable-image in the near future?
Oct 19 09:56:22 <[Rui]> JaMa|Wrk: cool, test/shr-u upgrade or change to normal shr-u before upgrade?
Oct 19 09:57:12 <[Rui]> Azog: definitely, the new test/shr-u images are mostly working now, a rougher edge than usual here and there, but neat, I've been running a test/shr-u image for two days now
Oct 19 09:57:51 <[Rui]> Azog: a nice new feature is audible ring tones :) some alsa driver change or fine tunning? I don't know... but now I can hear it well enough not to miss a call
Oct 19 10:00:21 <daniele_athome> good to know [Rui] :)
Oct 19 10:00:46 <[Rui]> daniele_athome: I was actually surprised when I first heard it ring so loud :)
Oct 19 10:01:14 <daniele_athome> :)
Oct 19 10:01:40 <[Rui]> sadly the upgrade will have to wait. going now...
Oct 19 10:45:24 <JaMa|Wrk> Azog: it will take about 4 more days to recompile images+feed for other supported devices
Oct 19 11:35:58 <Azog> JaMa|Wrk: thanks for the info
Oct 19 11:50:34 <JaMa|Wrk> Azog: but if you want it sooner than feel free to reflash or upgrade from tests/shr-unstable
Oct 19 11:52:09 <Azog> JaMa|Wrk: i have a problem right now with the unstable-image from 2. of october: wlan and usb is not working, so upgrade is not possible :(
Oct 19 11:52:38 <JaMa|Wrk> tests image or normal one
Oct 19 11:52:40 <JaMa|Wrk> ?
Oct 19 11:53:15 <Azog> JaMa|Wrk: shr-full-eglibc-ipk--20101002-om-gta02.rootfs.jffs2
Oct 19 11:54:42 <Azog> the one from unstable
Oct 19 11:57:59 <JaMa|Wrk> this old image should be safe.. did you flash also coresponding uImage?
Oct 19 11:58:43 <JaMa|Wrk> and can you check if g_ether module is loaded ok (in lsmod)?
Oct 19 11:59:21 <JaMa|Wrk> also ie mrmoku had problem when flashing with dfu-util.. if you have some sane distribution in uSD you can reflash with nandwrite instead (it's even a lot faster then dfu-util)
Oct 19 12:03:40 <Azog> thanks for the tip, i try, when i am at home again
Oct 19 12:14:43 <JaMa|Wrk> Martix: images from yesterday already have tslib :)
Oct 19 13:05:22 <Martix> great
Oct 19 13:49:00 <Heinervdm> i really hate my cardreader...
Oct 19 14:23:08 <Heinervdm> cdc_subset: probe of 8-1:1.0 failed with error -22
Oct 19 14:23:30 <Heinervdm> this happes when i connect my FR, does anyone know what this means?
Oct 19 14:40:01 <anarsoul> -22 is -EINVAL
Oct 19 14:40:08 <anarsoul> not sure where it comes from
Oct 19 14:40:48 <Heinervdm> i'ts working, so nvm
Oct 19 15:07:23 <morphis> heyho
Oct 19 15:08:54 <[Rui]> now with the 2010/10/18 image
Oct 19 15:08:56 <[Rui]> nice to see tslib is back :)
Oct 19 15:09:13 <[Rui]> still would be nice to have emulaterightbutton on by default
Oct 19 15:09:21 <[Rui]> (and less one icon in the toolbar).
Oct 19 15:09:56 <[Rui]> however, and this also happened with evdev in the other test image), the touchscreen is still a pain to use
Oct 19 15:10:08 <[Rui]> specially on the keyboard
Oct 19 15:16:27 <[Rui]> interesting... tslib is still not usable after all
Oct 19 15:16:35 <[Rui]> there's a problem with emulaterightbutton 1
Oct 19 15:21:05 <[Rui]> apparently, the problem is with restarting X
Oct 19 15:21:28 <[Rui]> because I commented the driver option and I still don't get a responsive screen. I'll put emulation on again and reboot
Oct 19 15:23:27 <[Rui]> hms... nothing
Oct 19 15:24:32 <[Rui]> argh... my bad, I had /dev/inputevent1 (missing a / there...) that's what I get for editing from within vala-terminal with a hard to use touchscreen
Oct 19 15:26:33 <[Rui]> BUT emulaterightbutton doesn't seem to work anyway
Oct 19 15:36:18 <JaMa|Off> [Rui]: but tslib behaves much better then evdev right?
Oct 19 15:40:18 <[Rui]> JaMa|Off: yeah, and the problem is the filtering, after all
Oct 19 15:40:22 <[Rui]> I got a menu by accident
Oct 19 15:40:36 <[Rui]> so the option is indeed working, but the ts filtering... ouch :)
Oct 19 15:41:07 <[Rui]> it's a kernel issue, right? one would need to finetune it there, yes?
Oct 19 15:42:31 <JaMa|Off> it's different ts driver in kernel which doesn't do filtering, right
Oct 19 15:42:49 <JaMa|Off> but tslib does filter a bit IMHO (maybe it's configurable)
Oct 19 15:42:59 <[Rui]> JaMa|Off: I wonder if the astonishing speed of the new illumehome has anything to do with the icons being displayed in a single columns...
Oct 19 15:43:07 <GNUtoo|laptop> about tslib vs evdev
Oct 19 15:43:10 <[Rui]> JaMa|Off: I can check on input-tslib
Oct 19 15:43:38 <GNUtoo|laptop> I switched back to evdev for htcdream because xorg people told me to do so and because of xf86-video-fbdev rotations issues 
Oct 19 15:43:57 <[Rui]> GNUtoo|laptop: if evdev can have rightbutton emulation, I have no preference between both of them!
Oct 19 15:44:02 <GNUtoo|laptop> not sure
Oct 19 15:44:05 <GNUtoo|laptop> I know tslib can
Oct 19 15:44:15 <[Rui]> GNUtoo|laptop: that's why I prefer it :)
Oct 19 15:44:17 <GNUtoo|laptop> but I've no ideas for fbdev
Oct 19 15:44:21 <GNUtoo|laptop> s/fbdev/evdev
Oct 19 15:44:23 <[Rui]> with it we can get rid of that icon on the toolbar
Oct 19 15:46:23 <JaMa|Off> [Rui]: evdev doesn't do any filters.. that's why it's even worse with 2.6.34
Oct 19 15:46:41 <JaMa|Off> GNUtoo|laptop: both rotation problems should be fixed by those OE patches we have
Oct 19 15:46:52 <GNUtoo|laptop> ok
Oct 19 15:48:28 <[Rui]> it still doesn't rotate properly on freerunner :(
Oct 19 15:48:48 <[Rui]> only knows rotations 0 and 1, 2 and 3 are 0 and 1 anyway
Oct 19 15:51:32 <[Rui]> JaMa|Off: oh, and if elmdentica is still pinned, could you unpin it?
Oct 19 15:51:35 <[Rui]> please? :)
Oct 19 15:52:20 <JaMa|Off> it isn't pinned, using global EFL_SRCREV
Oct 19 15:54:53 <JaMa|Off> this is updated patch from vanous (which should fix calibration after rotation) http://git.openembedded.org/cgit.cgi/openembedded/tree/recipes/xorg-xserver/xserver-xorg-1.9.0.902/randr-support.patch
Oct 19 15:56:34 <[Rui]> ah that would be a nice to have :)
Oct 19 15:56:43 <[Rui]> JaMa|Off: oh... so we're still with such an old efl?
Oct 19 15:56:44 <JaMa|Off> we have it
Oct 19 15:57:03 <[Rui]> JaMa|Off: not on 2010/10/18 image
Oct 19 15:57:04 <JaMa|Off> EFL_SRCREV ?= "53489"
Oct 19 15:57:23 <JaMa|Off> we have it
Oct 19 15:57:32 <[Rui]> I mean that elmdentica still has that elm_notify_thing_init issue
Oct 19 15:57:34 <JaMa|Off> for last 3 months or so
Oct 19 15:57:50 <[Rui]> JaMa|Off: then it doesn't work well :(
Oct 19 15:58:01 <JaMa|Off> [Rui]: didn't you install elmdentica from normal feeds instead tests/?
Oct 19 15:58:39 <[Rui]> JaMa|Off: oops.. maybe, the test/shr-u images point to the shr-u repo rather than test/shr-u repo?
Oct 19 15:59:09 <JaMa|Off> yes they does
Oct 19 15:59:52 <JaMa|Off> ok real off now.. bye
Oct 19 16:00:00 <[Rui]> see ya!
Oct 19 18:37:57 <[Rui]> hi
Oct 19 19:40:28 <GNUtoo|laptop> hi JaMa|Off 
Oct 19 19:40:33 <GNUtoo|laptop> | gee/Makefile.am:95: HAVE_INTROSPECTION does not appear in AM_CONDITIONAL is still there
Oct 19 19:40:38 <GNUtoo|laptop> with oe dev HEAD
Oct 19 19:41:32 <GNUtoo|laptop> libgee-native-1_0.6.0-r6.1
Oct 19 19:42:34 <JaMa> GNUtoo|laptop: are you sure it's head?
Oct 19 19:42:56 <GNUtoo|laptop> I'll look
Oct 19 19:42:57 <JaMa> GNUtoo|laptop: gobject-introspection-native in DEPENDS should be enough to pass libgee do_configure
Oct 19 19:43:17 <JaMa> GNUtoo|laptop: if you started build before then clean&build again
Oct 19 19:43:37 <GNUtoo|laptop> ok
Oct 19 19:43:43 <JaMa> details are on oe-devel ML
Oct 19 19:44:13 <GNUtoo|laptop> I think I've something strange
Oct 19 19:44:15 <GNUtoo|laptop> I'm not sure
Oct 19 19:44:31 <GNUtoo|laptop> but I have the impression that bitbake -k can pass the error in some way
Oct 19 19:44:44 <GNUtoo|laptop> I mean it does the images
Oct 19 19:44:49 <GNUtoo|laptop> if I remember well
Oct 19 19:47:56 <GNUtoo|laptop> strange strange
Oct 19 19:48:05 <GNUtoo|laptop> it seem that it ignores the errors
Oct 19 19:48:11 <GNUtoo|laptop> I've a libsamplerate patch error
Oct 19 19:48:16 <GNUtoo|laptop> and after it do_compile it
Oct 19 19:49:29 <JaMa> patch error? isn't it wrong checksum?
Oct 19 19:49:31 <GNUtoo|laptop> http://pastebin.com/6Py0BhwP
Oct 19 19:50:02 <GNUtoo|laptop> ah not patch
Oct 19 19:50:22 <GNUtoo|laptop> maybe it's because I use bitbake HEAD
Oct 19 19:50:51 <JaMa> libsamplerate do_patch looks ok, gnutls do_package failed
Oct 19 19:51:12 <GNUtoo|laptop> ah ok
Oct 19 19:51:19 <GNUtoo|laptop> I'm very tired sorry
Oct 19 19:51:39 <GNUtoo|laptop> and I hate beeing tired as I'm embedded-work-addicted
Oct 19 19:57:03 <Heinervdm> JaMa: is buildhost still building?
Oct 19 19:57:17 <JaMa> Heinervdm: yes
Oct 19 19:57:22 <JaMa> Heinervdm: now n900
Oct 19 19:57:24 <JaMa> then palmpre
Oct 19 19:57:25 <Heinervdm> ok
Oct 19 19:57:45 <JaMa> it's about 3-4 days for image+feed :/
Oct 19 19:57:54 <JaMa> per machine..
Oct 19 19:57:57 <Heinervdm> that's really long time
Oct 19 19:58:07 <JaMa> yes I/O is terrible
Oct 19 19:58:19 <JaMa> even removing tmpdir takes about half an hour
Oct 19 19:58:50 <JaMa> good that om-gta0[12] and n900/palmpre are the same arch
Oct 19 19:59:47 <Heinervdm> is tsfilter activated for tslib on gta02?
Oct 19 19:59:57 <Heinervdm> because there is still a lot of jitter
Oct 19 20:00:15 <[Rui]> Heinervdm: and that jitter makes it almost impossible to use the rightbuttonemulation
Oct 19 20:00:17 <Heinervdm> if you try aceofpenguins you can see it
Oct 19 20:00:22 <JaMa> with download stats we could create task-shr-feed-extras (and build those even after rsyncing)
Oct 19 20:00:51 <[Rui]> Heinervdm: just put a pointer in illumehome and try to stay fixed in one point, you'll see illumehome being dragged up and down a bit, constantly
Oct 19 20:00:53 <JaMa> Heinervdm: I didn't enable anything special in tslib.. just replaced xorg.conf and added kernel patch from qtmoko
Oct 19 20:01:38 <Heinervdm> playya__: how do i activate tsfilter in tslib.conf?
Oct 19 20:01:39 <[Rui]> JaMa: what is needed to setup a buildhost for constant builds?
Oct 19 20:01:52 <[Rui]> I have a powerufll enough ubuntu computer at home sitting idly most of the day
Oct 19 20:01:58 <[Rui]> only plays video at night...
Oct 19 20:02:24 <[Rui]> I mean that in the "lending cpu power for a buildhost" kind of way
Oct 19 20:02:27 <JaMa> what do you mean by constant?
Oct 19 20:02:32 <Heinervdm> hmm, dejitter is activated...
Oct 19 20:02:38 <Heinervdm> delta=100
Oct 19 20:03:04 <[Rui]> JaMa: I see it going a bit up and down all the time if I try to get it's context menu
Oct 19 20:04:15 <JaMa> [Rui]: there is problem with sharing SRCPV database cache/spitz/bb_persist_data.sqlite3 between multiple buildhosts..
Oct 19 20:04:39 <mrmoku> [Rui]: problem is connection to internet
Oct 19 20:04:46 <mrmoku> buildhost needs a fast one :;P
Oct 19 20:05:05 <[Rui]> download is more or less ok here, upload... not so muc
Oct 19 20:05:06 <JaMa> which could be fixed by gitpkgv.bbclass
Oct 19 20:05:25 <mrmoku> [Rui]: that's the problem... you have to sync the built stuff somewhere
Oct 19 20:06:02 <[Rui]> mrmoku: the built stuff is much, much smaller than the downloaded stuff :)
Oct 19 20:06:17 <JaMa> [Rui]: but you have to sync also sysroots
Oct 19 20:06:23 <mrmoku> yes, but when it's downloaded it's downloaded
Oct 19 20:06:25 <JaMa> [Rui]: not only resulting packages
Oct 19 20:06:47 <GNUtoo|laptop> ok now it's doing the image
Oct 19 20:06:57 <GNUtoo|laptop> I just pulled from oe and bitbaken again shr-image
Oct 19 20:07:02 <playya__> Heinervdm, what do you need?
Oct 19 20:07:18 <JaMa> or we could split feeds for each arch (like current buildhost of gta0[12] and another one for pre+n900 and another for htcdream
Oct 19 20:07:19 <playya__> you need to add modules after module_raw
Oct 19 20:07:24 <Heinervdm> playya__: how to configure the dejitter module
Oct 19 20:07:49 <mrmoku> JaMa: you mean to split on multiple buildhosts?
Oct 19 20:08:04 <Heinervdm> playya__: i changed delta from 100 to 200 but the jitter got even worse...
Oct 19 20:08:47 <playya__> I'm using: variance delta=30
Oct 19 20:08:53 <[Rui]> JaMa: splitting could probably be done
Oct 19 20:08:55 <playya__> dejitter delta=100
Oct 19 20:09:01 <playya__> and linear
Oct 19 20:09:21 <Heinervdm> playya__: that's the default
Oct 19 20:09:31 <playya__> yes
Oct 19 20:10:08 <Heinervdm> with those values there is still  a lot of jitter
Oct 19 20:10:08 <mrmoku> JaMa: we should try how long it takes to sync one arch over a DSL line
Oct 19 20:10:27 <playya__> do you use tsmd?
Oct 19 20:10:39 <Heinervdm> playya__: no, it's on gta-02
Oct 19 20:11:07 <JaMa> mrmoku: yes as different archs share almost "only" native recipes
Oct 19 20:11:16 <playya__> ah. ok.
Oct 19 20:11:23 <JaMa> mrmoku: but still it's a bit sad
Oct 19 20:11:28 <[Rui]> I'm not sure how big my upload is ,it's either 256 or 512
Oct 19 20:11:35 <[Rui]> kbps
Oct 19 20:11:41 <playya__> brb. need to find a nokia phone with a bl-5c
Oct 19 20:11:46 <JaMa> mrmoku: because now ie when some package is not built for armv7t you can still download it from armv4t feeds
Oct 19 20:12:12 <mrmoku> JaMa: well in the end we would have them in one place for download
Oct 19 20:12:31 <JaMa> mrmoku: and any sane box would finish rebuilding from scratch in 3-4 days for all currently supported machines (not just one)
Oct 19 20:12:42 <mrmoku> yeah :/
Oct 19 20:12:56 <JaMa> mrmoku: but without consistent versioning (due to SRCPV)
Oct 19 20:13:26 <JaMa> mrmoku: so if you put armv4 feeds on armv7 target then you can get "upgrade" to older versions with higher PV
Oct 19 20:15:08 <mrmoku> http://www.hetzner.de/en/hosting/produkte_rootserver/eq4/
Oct 19 20:15:14 <mrmoku> :P
Oct 19 20:16:35 <mrmoku> we would have to collect some donations though ;)
Oct 19 20:19:46 <JaMa> even X2 would be probably faster then current :)
Oct 19 20:20:33 <[Rui]> 738/y is something, indeed
Oct 19 20:22:21 <playya__> Heinervdm, if the distance to the last recorded point is bigger delta, dejitter discards the history
Oct 19 20:22:32 <playya__> maybe you should use a smaller value?
Oct 19 20:23:12 <mrmoku> JaMa: yeah probably :P
Oct 19 20:23:39 * JaMa thinks that biggest pain is from virtualized IO
Oct 19 20:23:59 <playya__> grml. damn ecrypfs and long filenames
Oct 19 20:24:05 <[Rui]> current build host is a vm?
Oct 19 20:24:10 <mrmoku> yup
Oct 19 21:25:02 <[Rui]> JaMa|Zzz: g'night
Oct 19 22:33:35 <pespin> mrmoku, JaMa|Zzz, hi could some of you review this patch when you have some free time please? http://pespin.espeweb.net/patches/0001-Add-efm-.desktop-file-icon-to-e-wm.patch
Oct 19 22:33:53 <pespin> I'm not sure if everything is set correctly 
**** ENDING LOGGING AT Wed Oct 20 02:59:56 2010