**** BEGIN LOGGING AT Tue Dec 21 02:59:59 2010 Dec 21 06:25:27 JaMa|Off: ok :/ Dec 21 07:47:04 * JaMa|Off still too busy with daywork to be able to resolve u-boot before holidays :/ Dec 21 07:47:14 JaMa|Off: np Dec 21 07:47:41 * mrmoku will be away from 24th to 3rd anyway Dec 21 07:47:55 italy? Dec 21 07:47:58 yup Dec 21 07:48:07 cold there too :/ Dec 21 07:48:13 only family visits or skiing too? Dec 21 07:48:21 * mrmoku would prefer caribics... or even a visit to TAsn :P Dec 21 07:48:25 no only family Dec 21 07:48:43 but my laptop will come with me ;) Dec 21 07:49:20 yeah visit TAsn and try water skiis for change :) Dec 21 07:49:28 :) Dec 21 07:49:57 * mrmoku would already be happy to be lying on a warm beach and watching TAsn doing the fancy stuff :-) Dec 21 07:53:07 hehe Dec 21 08:30:56 Hey :) What OE distro is recommended for a custom embedded project? Angstrom bleeding edge would require updates (long recompiles) too often. Dec 21 08:31:58 PaulFertser: you can use angstrom-2008 Dec 21 08:32:10 JaMa|Off: ain't that too old? It's 2011 coming. Dec 21 08:32:27 PaulFertser: or whatever distro in own branch based on release-2010.12 tag Dec 21 08:32:49 I'm also "suprised" i can't really build an image of angstrom because it depends on ipkg and handhelds cvs is unavailable :/ Dec 21 08:32:51 and then upgrade it to next release when that happens Dec 21 08:33:09 you have to use PREMIRRORS Dec 21 08:33:40 Not many distros to choose from, heh. Dec 21 08:34:10 shr/angstrom-20*/minimal/micro Dec 21 08:35:13 JaMa|Off: what would you recommend? Dec 21 08:35:35 JaMa|Off: btw, i can specify handhelds cvs in bitbake.conf but i do not know the mirror's url. Dec 21 08:36:01 I thought angstrom is OE's mainline distro and hence is the most suitable. Dec 21 08:36:22 depends on project :) but from that 5 I would expect most problems with micro (due to /usr flattened to /) Dec 21 08:37:23 minimal is also pretty good, general-purpose distro Dec 21 08:38:34 SHR won't have many benefits unless your project wants latest xorg stuff and telephony Dec 21 08:39:15 PaulFertser: ipkg dependency is probably from used image (not really distro-specific) Dec 21 08:39:44 JaMa|Off: i'm trying console-image here. Dec 21 08:40:02 JaMa|Off: aha, so you generally recommend minimal. Ok, thanks. Dec 21 08:42:46 PaulFertser: is it this file you see missing? http://permalink.gmane.org/gmane.comp.handhelds.openembedded/40248 Dec 21 08:44:52 hmm libphone-ui failing today | phoneui-utils.c:368:11: error: 'FSO_FRAMEWORK_PIM_DatesServicePath' undeclared (first use in this function) Dec 21 08:58:49 JaMa|Off: yeah, exactly so. Dec 21 08:59:39 JaMa|Off: but anyway that stupid handhelds.org is unreachable. Dec 21 09:00:51 JaMa|Off: and yeah, thanks for the hint :) Dec 21 09:01:34 resolved with PREMIRRORS? Dec 21 09:01:48 that file is on sources.openembedded.org Dec 21 09:05:43 JaMa|Off: not yet, i did git pull and now it rebuild the toolchain again etc.. Dec 21 09:31:18 larsc: just looking at upstream..qtmoko-v31. do you think 'S3C: Allow to completly disable low-level messages' would be something that should be sent upstream? looks innocent to me and probably would only need to spellcheck the commit message? Dec 21 09:32:04 larsc: for easy access it's http://paste.debian.net/102903/ Dec 21 09:32:35 larsc: since it modifies head.S in http://paste.debian.net/102900/ I became interested Dec 21 10:34:31 morning folks Dec 21 10:34:37 did you read my stuff yesterday nite? Dec 21 10:41:49 the gdbus changes? Dec 21 11:00:26 morning Dec 21 12:00:45 Hi, is this the correct channel to ask about compiling FSO? (cornucopia) Dec 21 12:01:34 playya_: yes, juergbi added the missing bits. we can switch soon now Dec 21 12:01:40 guyvdb_: worth a try Dec 21 12:04:32 I did a git clone of cornucopia and the readme has a build order... which specifies libgsm0710 before libgsm0710mux but 0710 is not part of the cornucopia repos. Should I check it out separately? I cannot ignore it and continue with the build order as I get a missing dependency Dec 21 12:04:45 mickey|office, cool. I'm writing a testcase if interface -> dbus -> object works, but no success yet :( Dec 21 12:07:40 Hey guys. Is anyone working on getting SHR to the HTC Hero? I have a spare Hero to play with now. Dec 21 12:12:44 mickey|office: great :-D Dec 21 12:13:20 guyvdb_: yeah, checkout separately Dec 21 12:14:35 guyvdb_: http://git.freesmartphone.org/?p=libgsm0710.git;a=summary Dec 21 12:14:42 guyvdb_: http://git.freesmartphone.org/?p=libgsm0710mux.git;a=summary Dec 21 12:15:36 thx Dec 21 12:23:42 mrmoku, guyvdb_: the second one is deprected Dec 21 12:23:59 use the cornucopia version instead Dec 21 12:27:29 ahh, right Dec 21 12:27:33 HeinervdmWork: thanks :) Dec 21 12:29:24 yea i used libgsm0710 for own repos and libgsm0710mux from cornucopia ... correct? Dec 21 12:31:08 correct Dec 21 14:10:13 Hi, I am just starting to look at fso. I am using a mini2440. I want to establish a ppp connection (always on) and suspend it during voice calls. I want to communicate via dbus. Which elements in the fso stack should i be looking at? (Abyss??). Dec 21 14:15:41 hmm. fsodeviced fails for HEAD on the pre, because of missing i2c header Dec 21 14:16:12 playya_: ya, OE broke it Dec 21 14:16:15 should i add a switch or disable i2c plugins if the headers are missing? Dec 21 14:16:18 oh. ok Dec 21 14:16:29 copy the i2c-userdev.h to i2c-dev.h Dec 21 14:16:31 then it should build Dec 21 14:16:36 OE renamed it to i2c-dev-user.h Dec 21 14:18:30 or include tmp/sysroots/palmpre-oe-linux-gnueabi/kernel/include/linux/i2c-dev.h Dec 21 14:19:33 hmm. it's available in the arm sysroot, too Dec 21 14:27:59 you need the one from sysroot (staged by i2c-tools) Dec 21 14:33:40 ah. ok Dec 21 15:02:24 even with install i2c-tools there's no i2c-user*.h Dec 21 15:02:29 playya@thinky:~/pre$ find tmp/sysroots/ -name '*i2c*dev*.h' Dec 21 15:02:29 tmp/sysroots/x86_64-linux/usr/include/linux/i2c-dev.h Dec 21 15:02:29 tmp/sysroots/armv7a-oe-linux-gnueabi/usr/include/linux/i2c-dev.h Dec 21 15:02:29 tmp/sysroots/palmpre-oe-linux-gnueabi/kernel/include/linux/i2c-dev.h Dec 21 15:06:46 playya_: 3.0.2 version? 3.0.1 keeps old i2c-dev.h name Dec 21 15:07:26 playya_: ie /OE/tmpdir-shr/sysroots/armv4t-oe-linux-gnueabi/usr/include/linux/i2c-dev-user.h Dec 21 15:08:11 i2c-tools-3.0.2-r2 Dec 21 15:09:10 then rebuilding it is enough Dec 21 15:09:19 it was changed in r2 and then pr bumped to r3 Dec 21 15:09:41 so you probably have OE before 48e6a063370a38a35f31a28efd8f6ce6ebf00840, right? Dec 21 15:11:27 yes Dec 21 15:11:41 still on morphis' oe branch on fso Dec 21 15:11:54 playya_: then rebuilding i2c-tools after every linux-libc-headers build is enough Dec 21 15:12:08 but it seems to work after i cleaned kernel and i2c-tools Dec 21 15:12:59 btw. shouldn't some of the dependencies conditional for different machine? Dec 21 15:13:25 e.g. pre doesn't depend on androidrpc and i2c-tools Dec 21 15:43:10 freesmartphone.org: 03morphis 07cornucopia * rf6f518737c3a 10/fsodeviced/src/plugins/palmpre_quirks/ (plugin.vala powercontrol.vala powersupply.vala): Dec 21 15:43:10 freesmartphone.org: fsodeviced: palmpre_quirks: be more verbose Dec 21 15:43:10 freesmartphone.org: Signed-off-by: Simon Busch Dec 21 15:47:15 freesmartphone.org: 03morphis 07cornucopia * r9094938e5110 10/fsodeviced/src/plugins/palmpre_quirks/powersupply.vala: Dec 21 15:47:15 freesmartphone.org: fsodeviced: palmpre_quirks: fix bug in battery authentication handling Dec 21 15:47:15 freesmartphone.org: Signed-off-by: Simon Busch Dec 21 17:03:38 SHR: 03mail 07libphone-ui * r3690983fd837 10/src/phoneui-utils.c: phoneui-utils: disable dates support Dec 21 18:06:02 heyho Dec 21 18:34:08 freesmartphone.org: 03mickey 07gdbus * r7e9d3cb83895 10cornucopia/libfsoframework/fsoframework/dbus.vala: libfsoframework: repair DBusServiceNotifier for gdbus Dec 21 18:35:07 freesmartphone.org: 03morphis 07msmcomm * r64a21070b7d9 10/msmcommd/configure.ac: Dec 21 18:35:07 freesmartphone.org: msmcommd: add libfsobasics as dependency Dec 21 18:35:07 freesmartphone.org: Signed-off-by: Simon Busch Dec 21 18:35:11 freesmartphone.org: 03morphis 07msmcomm * rccd459ddcc42 10/msmcommd/src/lowlevelcontrol.vala: Dec 21 18:35:11 freesmartphone.org: msmcommd: lowlevel control: write directly to sysfs nodes Dec 21 18:35:11 freesmartphone.org: Writing to sysfs nodes via calling echo with the system(...) function does not work Dec 21 18:35:11 freesmartphone.org: probably so we do that here directly and reduce the dependency on any other userland tool. Dec 21 18:35:11 freesmartphone.org: Signed-off-by: Simon Busch Dec 21 18:35:13 freesmartphone.org: 03morphis 07msmcomm * r8684dfbf87b6 10/msmcommd/src/ (dbusservice.vala util.vala): Dec 21 18:35:13 freesmartphone.org: msmcommd: fix several vala compilation warnings Dec 21 18:35:13 freesmartphone.org: Signed-off-by: Simon Busch Dec 21 18:35:55 mrmoku, morphis, playya_, JaMa|Off, gnutoo, heinervdm, etc.: can we meet later for a bit? i'd like to discuss our gdbus migration strategy Dec 21 18:36:53 mickeyl: ok Dec 21 18:37:13 mickeyl: can you define 'later'? Dec 21 18:37:23 mickeyl: yup Dec 21 18:37:32 lets do a quick poll Dec 21 18:37:46 20:00, 20:30, 21:00, 21:30 ? Dec 21 18:37:58 whenever :) Dec 21 18:38:06 ok, i'm fine with all of that as well Dec 21 18:38:10 morphis: when is it good for you? Dec 21 18:38:16 20:00 but I can not gurantee that I am not leaving suddenly Dec 21 18:38:20 gnutoo will probably appear later as well Dec 21 18:38:24 ok, then lets aim for the earliest Dec 21 18:38:45 mickeyl: gnutoo is on a trip to bergamo to day... to attend some RMS event Dec 21 18:38:50 dunno when he'll be back Dec 21 18:38:50 ok Dec 21 18:39:14 ah, ok, doesn't matter too much. it's more important to get the distro guys Dec 21 18:39:24 JaMa|Off: ping :) Dec 21 18:39:26 but maybe we can discuss the basic bits here and then publish everything to the ML Dec 21 18:39:44 that for sure Dec 21 18:39:52 ok Dec 21 18:40:46 ok, some house duties then Dec 21 18:40:47 cu in a bit Dec 21 18:40:52 ok Dec 21 18:42:53 mrmoku: some question about machine handling in OE Dec 21 18:43:05 freesmartphone.org: 03morphis 07msmcomm * r687af3449b21 10/msmcommd/conf/msmcommd.conf: Dec 21 18:43:05 freesmartphone.org: msmcommd: mention defauls values for lowlevel sysfs nodes in configuration file Dec 21 18:43:05 freesmartphone.org: Signed-off-by: Simon Busch Dec 21 18:43:08 if you have some time for me Dec 21 18:44:51 morphis: sure Dec 21 18:45:10 mrmoku: I need to introduce a machine config for the palmpre2 Dec 21 18:45:19 ok Dec 21 18:45:22 the palmpre2 is exactly the same as palmpre but just needs another kernel Dec 21 18:45:31 how to handle this best in OE? Dec 21 18:45:40 introducing palmpre2.conf which requires palmpre.conf? Dec 21 18:45:56 hmm... let me check if machine confs have .inc Dec 21 18:46:01 will then any _append_palmpre even work for a palmpre2.machine? Dec 21 18:46:53 morphis: there is conf/machine/include Dec 21 18:46:59 jepp Dec 21 18:47:08 I think putting a palmpre.inc into there would be the best way Dec 21 18:47:24 and then any *_append_palmpre is still working? Dec 21 18:47:33 when using palmpre2 machine? Dec 21 18:47:51 hmm... no Dec 21 18:48:02 thats the problem ... Dec 21 18:48:47 * mrmoku digs a bit Dec 21 18:49:19 I even searched for a similar case but found nothing Dec 21 18:49:36 the problem is even that we have many machine independet things for the palmpre Dec 21 18:49:55 morphis: MACHINE_OVERRIDES += "palmpre palmpre2" Dec 21 18:49:57 something like that Dec 21 18:50:04 would be my guess Dec 21 18:50:36 is that possible? Dec 21 18:50:45 let me ask in #oe Dec 21 18:50:50 a right Dec 21 18:50:55 look at nslu2be.conf Dec 21 18:50:59 it is nslu2 Dec 21 18:51:03 thats nice Dec 21 18:51:06 really nice Dec 21 18:51:51 yeah Dec 21 18:51:58 OE is a quite flexible beast :P Dec 21 18:52:41 mrmoku: pong Dec 21 18:52:46 JaMa|Off: hey Dec 21 18:53:02 JaMa|Off: we would like to have a 'switch to gdbus meeting' :-) Dec 21 18:57:14 ok, so I need more coffee :) Dec 21 18:59:42 JaMa: was late last night, he :P Dec 21 19:04:33 ok, lets get started Dec 21 19:05:06 as you might know yesterday juergbi committed something that (at this point of time) I consider being a fix for the last showstopper wrt. gdbus migration Dec 21 19:05:21 i have done all the work in the branch gdbus up to now Dec 21 19:05:29 the problem w/ gdbus is that it is an all-or-nothing thing Dec 21 19:05:38 we can't switch on an individual daemon basis Dec 21 19:05:53 since it has consequences for the basic infrastructure, such as in libfsoframework and libfso-glib Dec 21 19:05:59 (which will then be renamed libfso-gdbus) Dec 21 19:06:01 so Dec 21 19:06:10 there are two options now and I'd like your input Dec 21 19:06:25 a) put the gdbus stuff in OE and add some infrastructure to switch between the two Dec 21 19:07:56 this will affect a dozen of recipes Dec 21 19:07:56 b) merge gdbus and deal with the consequences.. Dec 21 19:07:56 a) is the safe bet as it will keep the current system working Dec 21 19:07:56 but a) also means much more work Dec 21 19:07:57 b) is the best way to make sure the gdbus problems are being recognized asap Dec 21 19:07:58 and... Dec 21 19:07:58 given the timeframe, it may be just about the right time to do the switch Dec 21 19:07:59 since i have a couple of free days between the years Dec 21 19:08:00 an opportunity which might not happen again for the next months Dec 21 19:08:01 in any case i need your support Dec 21 19:08:08 what do you think? Dec 21 19:09:19 I think b) is the way to go then Dec 21 19:09:29 we can warn people not to upgrade ;) Dec 21 19:09:35 and sort out stuff Dec 21 19:09:44 though Dec 21 19:09:53 I'm away from 24th to 3rd Dec 21 19:09:59 won't be of much help there Dec 21 19:10:37 I even think b) is the best way Dec 21 19:10:50 a) will produce to much work and problems Dec 21 19:11:20 and when you say you have some time between the years and after not really for months, let us do that now Dec 21 19:11:30 it's something we wait for a long time now Dec 21 19:11:45 +100 Dec 21 19:12:21 ok, is that the general consensus or are there any major objections? Dec 21 19:12:21 ok Dec 21 19:12:31 we are three people Dec 21 19:12:41 and I think most of the other would say the same Dec 21 19:12:45 JaMa is still searching for coffee Dec 21 19:13:09 if you're in short before a release, we can postpone a couple of days, otherwise i'd start with merging gdbus back to head and separating libfso-glib / libfso-gdbus on 23th Dec 21 19:13:28 HeinervdmOff did alread branch off for a testing release Dec 21 19:13:32 so we're fine Dec 21 19:13:36 ah, that's good Dec 21 19:13:44 so i don't feel guilty for breaking all your phones ;) Dec 21 19:13:56 :) Dec 21 19:13:59 mickeyl: why is it a all or nothing switch? Dec 21 19:14:00 maybe we can freeze the master branches? Dec 21 19:14:05 mickeyl: PaulFertser would say it's broken anyway ;) Dec 21 19:14:17 and do development in a gdbus-based branch Dec 21 19:14:58 morphis: i will tag master Dec 21 19:15:03 or this way Dec 21 19:15:08 should suffice Dec 21 19:15:19 but what we have to do then Dec 21 19:15:20 DieMumiee: gdbus and dbus-glib have different semantics when operating from Vala Dec 21 19:16:05 maybe we should count the tasks and process and write it down somewhere so we have an overview and anyone who wants can pick up a task out of it Dec 21 19:16:06 DieMumiee: so to make it compile, you have to do changes in almost all FSO libs and subsystems Dec 21 19:16:13 ok, just reading about gdbus. Wasnt aware that the gnome world has another dbus impl Dec 21 19:16:31 DieMumiee: that's all new, but it's supposed to fix a bunch of nitty gritty bugs Dec 21 19:16:40 and probably introduce new ones, but hey... that's life Dec 21 19:16:41 but it is only per process switch? or somehow system wide? Dec 21 19:16:48 :) Dec 21 19:17:02 mickeyl: list all things which could be a problem Dec 21 19:17:06 i am trying to move my company into using dbus for some of the application interfaces Dec 21 19:17:12 well, it's dbus, so you can switch per-process Dec 21 19:17:20 however... most of FSO is written in vala now Dec 21 19:17:26 and i don't have the energy to maintain two branches :) Dec 21 19:17:42 aye understood Dec 21 19:17:45 morphis: ok, i will sum up our discussion and write a mail to SHR-devel, OM-devel, and FSO-userland Dec 21 19:17:47 * JaMa also votes for b) Dec 21 19:17:53 JaMa: ok, excellent Dec 21 19:18:19 mickeyl: thats one point, we should even sum up all things which are covered by this merge, applications, daemons and so on Dec 21 19:18:34 to have an overview where we have to fix things or better where we have to look for bugs Dec 21 19:19:29 morphis: right. the good thing is that the dbus interfaces all stay the same Dec 21 19:19:34 jepp Dec 21 19:19:41 so all external processes should be unaffected Dec 21 19:19:42 but the apps are using libfso-glib Dec 21 19:19:55 yes, they might have to adapt Dec 21 19:19:59 or almost certainly Dec 21 19:20:01 mrmoku: coffee found, few more u-boot revs tested too :) Dec 21 19:20:06 :) Dec 21 19:20:10 successfull or not? Dec 21 19:20:21 mickeyl: should we start some wiki page for this? Dec 21 19:20:34 morphis: sounds good! Dec 21 19:20:39 ok Dec 21 19:20:41 we create one Dec 21 19:20:42 can you go ahead and create one? I'll fill the details Dec 21 19:21:15 mrmoku: sofar mostly successfull slowly bisecting Dec 21 19:21:29 mickeyl: will put it under Implementations, ok? Dec 21 19:21:38 or should it be an "event"? Dec 21 19:21:52 no, implementations should be ok Dec 21 19:22:04 ok Dec 21 19:22:07 btw., yesterday also was a breakthrough for gta04 Dec 21 19:22:10 JaMa: ok Dec 21 19:23:10 mickeyl: that remindes me to ask him if he wants the money :-) Dec 21 19:23:22 right Dec 21 19:23:28 fwiw: http://shr.pastebin.ca/2025604 Dec 21 19:24:21 mickeyl: nice :) Dec 21 19:24:41 yeah, he did some magic in software to cope with some funny wiring bugs Dec 21 19:24:54 hns' stubbornness rocks Dec 21 19:24:56 :) Dec 21 19:24:59 hehe Dec 21 19:25:12 Wo ein Wille da ein Weg :P Dec 21 19:25:16 so isses Dec 21 19:26:57 playya_: what's the state of the async delegates and signal handlers, btw? Dec 21 19:27:04 didn't hear about that for quite a while now Dec 21 19:28:24 mickeyl: http://wiki.freesmartphone.org/index.php/Implementations/gdbus-migration Dec 21 19:28:37 cool, thanks Dec 21 19:29:28 mickeyl: gta04: yeah! Dec 21 19:29:53 mickeyl: is that u-boot or real linux console Dec 21 19:29:58 u-boot Dec 21 19:30:01 ok Dec 21 19:30:09 but this is hardware verification stage Dec 21 19:30:23 it took some months to get to that stage :) Dec 21 19:30:50 drivers and stuff will almost be simple compared to what lays behind... Dec 21 19:31:12 exactly as many drivers should be already there Dec 21 19:31:19 omap3 is well supported these days Dec 21 19:31:27 and most people has many experience with it Dec 21 19:33:29 hello!!! :) Dec 21 19:33:39 daniele_athome: hey Dec 21 19:33:42 hi morphis Dec 21 19:34:13 i've just committed a patch with some big changes Dec 21 19:34:15 :) Dec 21 19:34:23 JaMa, mrmoku i've them sent to ML :) Dec 21 19:38:35 daniele_athome: why some big patch ? Dec 21 19:38:53 morphis, phone and panel have been removed from appbunch Dec 21 19:39:03 panel has many new stuff Dec 21 19:39:10 desktop app too Dec 21 19:39:12 and so on :) Dec 21 19:39:18 mrmoku, keep your hopes down... it's rare that we have surfable waves here, and even if we do, the bitches aren't that warm :| I wear a full swimsuit, and it's too cold to just sit on the beach :) Dec 21 19:39:22 (in the winter) Dec 21 19:39:27 p.s I want to go to italy :) Dec 21 19:39:31 daniele_athome: ok, but why do you put all these changes on one big patch? Dec 21 19:39:39 Bitches not warm? :-O Dec 21 19:39:43 and not makes on patch per change Dec 21 19:39:44 morphis, no it's just a recipe patch Dec 21 19:39:59 morphis, because it's just a couple of srcrev bump Dec 21 19:40:07 and 2 new recipes, all concerning mokosuite2 Dec 21 19:40:13 right Dec 21 19:40:27 but I would prefer to split them up and submit them in the right order Dec 21 19:40:31 helps to track down bugs later Dec 21 19:40:43 morphis, yes you're right but i'm too lazy :P Dec 21 19:40:51 hehe Dec 21 19:40:59 then hope that the patch is applied as it is :) Dec 21 19:41:14 I know some people who would never apply it as it is :) Dec 21 19:41:39 morphis, such as? :) Dec 21 19:42:10 me or other OE developers Dec 21 19:43:52 eheheh Dec 21 19:43:58 but anyway are there some recent screenshot of mokosuite somewhere? Dec 21 19:44:06 morphis, anyway i'll keep in mind the next time Dec 21 19:44:13 ok Dec 21 19:44:14 PaulFertser, lol, oops, the bitches are warm, typo. :) Dec 21 19:44:17 morphis, no because the gui hasn't changed Dec 21 19:44:22 daniele_athome: ok Dec 21 19:44:25 it's just a major refactoring Dec 21 19:44:29 very big actually Dec 21 19:44:30 ok Dec 21 19:44:40 implementing some standards ecc. Dec 21 19:44:45 you heard about the fso gdbus merge? Dec 21 19:44:45 TAsn: what a relief, that sounds encouraging Dec 21 19:44:57 that will happen in the next weeks? Dec 21 19:45:17 morphis, yes i heard about that Dec 21 19:45:20 ok Dec 21 19:45:20 PaulFertser, yep, but you can't take them to the beach, cause it's too cold this season. Dec 21 19:45:23 i guess so Dec 21 19:45:42 I am right that mokosuite is affected by it? Dec 21 19:46:16 or do you don't use vala for dbus communication? Dec 21 19:47:00 morphis, yes I use vala for dbus servers, but i don't use gdbus yet Dec 21 19:47:17 for the client part is all pure C Dec 21 19:47:28 I don't think using dbus-glib for a time will break anything Dec 21 19:47:34 s/time/while/ Dec 21 19:47:35 daniele_athome meant: I don't think using dbus-glib for a while will break anything Dec 21 19:59:08 JaMa: hello, I have a strange problem: If I compile a fresh shr image "make image" it does not compile (glib-2.0, missing dbus/dbus.h), if I bitbake glib-2.0 in a fresh environment it compiles. Do you have the same problem ? Dec 21 20:00:41 nschle85: yes I have, reported to OE few days ago Dec 21 20:01:04 nschle85: http://comments.gmane.org/gmane.comp.handhelds.openembedded/40368 Dec 21 20:02:14 nschle85: workarround is to build dbus manually first (with -b it ignores dependencies) and then you can build glib and the rest Dec 21 20:03:28 JaMa: thank you Dec 21 20:04:18 JaMa, http://patchwork.dev.bearstech.com/patch/819/ Dec 21 20:04:21 abort this patch please Dec 21 20:04:29 i have applied it upstream Dec 21 20:04:41 (it was actually a mokohome patch) Dec 21 20:08:52 mrmoku: hmm first commit which breaks newer u-boot (still with old toolchain) is http://git.denx.de/u-boot.git/?p=u-boot.git;a=commit;h=29844707469854d9fab181edd6abe2f25fb5d208 Dec 21 20:09:29 daniele_athome: ? gry theme patch was for mokohome? Dec 21 20:09:41 JaMa, yes :) trevino mistake Dec 21 20:09:56 ah oki Dec 21 20:10:56 JaMa, that's because my app is themable, and the default theme is named "gry", which is supposed to be like the elementary gry theme Dec 21 20:12:00 but in mokohome I see theme.edc not home.edc, but ok I belive you :) Dec 21 20:12:53 JaMa: hmm Dec 21 20:14:25 mrmoku: but maybe it gets fixed 2 commits later with 0c0a0e07811965188d5f64cdbc186331c0598fa6 Dec 21 20:19:27 JaMa, the patch was actually for mokohome in appbunch Dec 21 20:19:36 :) Dec 21 20:33:03 daniele_athome: btw mokohome and mokophone are not in any recipe depends, should I put those to task-shr-feed? Dec 21 20:33:18 hi GNUtoo|laptop Dec 21 20:33:23 how was your trip? Dec 21 20:33:27 fine Dec 21 20:33:28 JaMa, yes please, i forgot that :( Dec 21 20:33:34 s/mokohome/mokopanel/ Dec 21 20:33:35 JaMa meant: daniele_athome: btw mokopanel and mokophone are not in any recipe depends, should I put those to task-shr-feed? Dec 21 20:33:35 I've to write a mail now Dec 21 20:33:48 ok Dec 21 20:40:04 daniele_athome: did you also forget to add libnotify to DEPENDS? :) Dec 21 20:40:21 JaMa, on mokopanel? no I added it! Dec 21 20:40:28 libmokosuite Dec 21 20:40:46 azz forgot on mokopanel Dec 21 20:40:47 gggrrrr Dec 21 20:41:00 but i don't think it's of any use Dec 21 20:41:16 because mokopanel depends on libmokosuite which in turn depends to libnotify Dec 21 20:41:20 JaMa, am i right? Dec 21 20:41:43 so you have to put it in libmokosuite (I already did) Dec 21 20:45:02 JaMa, ah yes sorry i added it in my local copy of the recipe Dec 21 20:45:03 but not to oe Dec 21 20:45:04 :( Dec 21 20:45:38 no problem Dec 21 20:49:35 daniele_athome: pushed and build started Dec 21 20:49:45 ok nice :) Dec 21 21:02:02 JaMa: next Problem: fsodeviced configure fails, configure: error: Could not find i2c functions in linux/i2c-dev.h...please install i2c-tools Dec 21 21:02:22 JaMa: but i2c-tools are installed Dec 21 21:03:03 nschle85: cp tmp/sysroots/armv4t-oe-linux-gnueabi/usr/include/linux/i2c-dev-user.h tmp/sysroots/armv4t-oe-linux-gnueabi/usr/include/linux/i2c-dev.h Dec 21 21:04:13 nschle85: http://www.mail-archive.com/openembedded-devel@lists.openembedded.org/msg13983.html Dec 21 21:26:01 mrmoku, hi Dec 21 21:28:41 hi GNUtoo|laptop Dec 21 21:28:48 any progress? Dec 21 21:28:53 yup Dec 21 21:29:00 wow /me listens Dec 21 21:29:13 all depends on config :) Dec 21 21:29:25 moment Dec 21 21:29:45 ok Dec 21 21:29:54 http://paste.pocoo.org/show/308091/ Dec 21 21:30:04 that's what I have right now Dec 21 21:30:12 g_nokia works Dec 21 21:30:18 it charges even Dec 21 21:30:27 you mean with the script? Dec 21 21:30:32 no, with fsodeviced Dec 21 21:30:37 ok nice Dec 21 21:31:01 searching for the sysnodes to power on the modem for over an hour now Dec 21 21:31:12 and some minute ago I discovered /dev/cmt :P Dec 21 21:31:18 ok Dec 21 21:31:39 an ofono has if (gpio) do gpio stuff else write something to /dev/cmt Dec 21 21:31:45 ,logs? Dec 21 21:31:45 trying to understand that Dec 21 21:31:56 logs? Dec 21 21:31:59 mrmoku, look at lowlevel_nokia900 Dec 21 21:32:00 ~logs Dec 21 21:32:00 logs are on 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. logs are updated daily --- livelogs are http://logs.nslu2-linux.org/livelogs/openmoko-cdevel.txt and http://hentges.net/tmp/logs/irc/livelogs/#openmoko-cdevel.livelog Dec 21 21:32:02 it should work Dec 21 21:32:13 but it's untested Dec 21 21:32:31 you mean it should power on the modem? Dec 21 21:32:33 i need normal logs, not livelogs Dec 21 21:32:48 mrmoku, it should handle power on/off Dec 21 21:32:54 it's taken from ofono Dec 21 21:33:02 GNUtoo|laptop: nice thing is there is something new in sysfs :) Dec 21 21:33:09 root@nokia900 /sys/devices/platform/omap_ssi.0/ssi0/port0/ssi_protocol/net/phonet0 # ls Dec 21 21:33:09 ah Dec 21 21:33:12 addr_assign_type broadcast device features ifindex mtu speed tx_queue_len Dec 21 21:33:15 addr_len carrier dormant flags iflink operstate statistics type Dec 21 21:33:18 address dev_id duplex ifalias link_mode power subsystem uevent Dec 21 21:33:33 ok Dec 21 21:33:43 is it the normal sys node? Dec 21 21:33:44 but, when I'm trying to ifconfig phonet0 up I get the following in dmesg Dec 21 21:33:47 [ 1294.572479] ADDRCONF(NETDEV_UP): phonet0: link is not ready Dec 21 21:33:55 normal sysnode? Dec 21 21:33:56 then power it on Dec 21 21:34:09 I mean maybe every interface have the same node Dec 21 21:34:11 like wlan0 Dec 21 21:34:28 I think so, yes Dec 21 21:35:42 1970-01-01T00:00:18.711334Z [INFO] NokiaIsiModem <>: Modem Status changed to FSO_GSM_MODEM_STATUS_INITIALIZING Dec 21 21:35:46 1970-01-01T00:00:18.711639Z [INFO] DBusServiceDevice <>: Modem opened successfully Dec 21 21:35:49 GNUtoo|laptop: does that mean it should be powered on? Dec 21 21:35:55 add more debug Dec 21 21:36:00 that's nokia isi modem Dec 21 21:36:06 not lowlevel_nokia900 Dec 21 21:36:12 and remember.... Dec 21 21:36:13 ahh... fsodeviced Dec 21 21:36:17 you have the test vala stuff Dec 21 21:36:24 like authenticate Dec 21 21:36:28 and network Dec 21 21:36:31 written in vala Dec 21 21:36:34 from libisi Dec 21 21:37:25 GNUtoo|laptop: on the device? Dec 21 21:37:48 yes that would be a good start Dec 21 21:38:30 ohh... I need to add the lowlevel plugin Dec 21 21:38:44 GNUtoo|laptop: at what point to add it? Dec 21 21:39:14 it's already shipped and in the config Dec 21 21:39:24 just add DEBUG Dec 21 21:39:25 and see Dec 21 21:39:45 no, it's not in my config Dec 21 21:40:01 ah Dec 21 21:40:05 pastebin fsogsmd.conf Dec 21 21:40:25 * mrmoku confused Dec 21 21:41:18 brb Dec 21 21:45:29 mmm Dec 21 21:52:45 re Dec 21 21:52:49 ok Dec 21 21:53:17 2010-12-21T21:39:11.360260Z [DEBUG] NokiaIsiModem <>: ISICHANNEL: initialize ... Dec 21 21:53:20 2010-12-21T21:39:11.360412Z [DEBUG] NokiaIsiModem <>: ISICHANNEL: modem initialization is finished! Dec 21 21:53:33 pastebin fsogsmd.conf Dec 21 21:53:42 basically I'll explain Dec 21 21:53:50 imagine you've got 2 phones Dec 21 21:53:53 a nokia n900 Dec 21 21:54:02 and a nexusS Dec 21 21:54:08 imagine that they have the same modem Dec 21 21:54:10 the nokia isi Dec 21 21:54:19 but that the way of activating it is different Dec 21 21:54:26 [fsogsm] Dec 21 21:54:26 lowlevel_type = nokia900 Dec 21 21:54:26 modem_type = nokia_isi Dec 21 21:54:32 ok Dec 21 21:54:34 that's fine then Dec 21 21:54:40 add DEBUG output Dec 21 21:54:45 and pastebin the whole log Dec 21 21:54:55 global DEBUG is on Dec 21 21:55:03 ok then pastebin the whole log Dec 21 21:55:03 does the lowlevel plugin have it's own debug level? Dec 21 21:55:06 ok Dec 21 21:55:07 no Dec 21 21:55:09 it's global Dec 21 21:55:17 the lowlevel plugin is for activating the modem Dec 21 21:55:23 the nokia isi plugin for using it Dec 21 21:55:46 http://shr.pastebin.com/0cqpwa4D Dec 21 21:55:55 yup Dec 21 21:56:22 let me try on .28 Dec 21 21:56:38 0x6c was phonet address? Dec 21 21:56:55 I think so Dec 21 21:56:57 PN_DEV Dec 21 21:56:59 grep for that Dec 21 21:58:07 GNUtoo|laptop: can't set it there anyway Dec 21 21:58:19 root@nokia900 /sys/devices/platform/omap_ssi.0/ssi0/port0/ssi_protocol/net/phonet0 # echo 0x6c > address Dec 21 21:58:22 -sh: can't create address: Permission denied Dec 21 21:58:29 I'll look Dec 21 21:58:34 but first I compile shr-image Dec 21 21:58:51 * mrmoku looks at what meego does in udev.rules Dec 21 21:59:00 ok Dec 21 21:59:41 ohh... interesting Dec 21 21:59:55 http://paste.pocoo.org/show/308111/ Dec 21 22:00:16 ok Dec 21 22:00:23 very interesting indeed Dec 21 22:02:54 after that ifconfig phonet0 up works :-) Dec 21 22:03:07 ok Dec 21 22:04:08 root@nokia900 /sys/devices/platform/omap_ssi.0/ssi0/port0/ssi_protocol/net/phonet0 # cat address Dec 21 22:04:11 15 Dec 21 22:05:52 hmm... the test just hangs now Dec 21 22:06:56 hmmm Dec 21 22:07:06 fsogsmd seem not to handle power on/off anymore Dec 21 22:10:44 switching GSM on and off doesn't produce any dbus calls Dec 21 22:10:59 trough shr-settings Dec 21 22:11:19 hmm Dec 21 22:12:02 but other stuff do Dec 21 22:12:07 like auto-dimming Dec 21 22:18:41 GNUtoo|laptop: ohh... [ 7851.637634] ADDRCONF(NETDEV_CHANGE): phonet0: link becomes ready Dec 21 22:18:44 :) Dec 21 22:18:56 ok Dec 21 22:19:13 but normally fsogsmd's lowlevel_nokia900 should take care of it Dec 21 22:20:00 yeah, but it can't without either having the gpio-switch or the links in /dev Dec 21 22:20:06 and it was fsogsmd doing that :) Dec 21 22:20:16 I just added those links and restarted fsogsmd Dec 21 22:25:38 yes but it doesn't even work on 2.6.28 Dec 21 22:25:40 goodnight guys! Dec 21 22:25:44 did you look at the code? Dec 21 22:27:22 http://git.freesmartphone.org/?p=cornucopia.git;a=blob;f=fsogsmd/src/plugins/lowlevel_nokia900/nokia-gpio.c;h=95ff0326d9f6960040dabecb3d7dd386733b501d;hb=HEAD Dec 21 22:32:37 GNUtoo|laptop: yeah Dec 21 22:33:23 if cmtdir exists Dec 21 22:33:25 it uses that Dec 21 22:33:32 else it fallsback to gpiodir Dec 21 22:33:49 if only it ran Dec 21 22:33:56 I mean the dbus stuff is not called Dec 21 22:34:03 the other way round Dec 21 22:34:16 if gpio does not exist it tries dev Dec 21 22:34:27 ok Dec 21 23:28:16 mrmoku: hey :-D Dec 21 23:28:19 nice **** ENDING LOGGING AT Wed Dec 22 02:59:57 2010