**** BEGIN LOGGING AT Wed Apr 27 02:59:58 2011 Apr 27 04:15:29 >nf Apr 27 04:15:36 ~nf Apr 27 04:15:37 The #openmoko-cdevel Newsflash Bulletin Board. (continued at ~NF2. For help see ~NF-help) - - - Recommended URLs and channels(chanlogs etc):see ~RL - - - NEWSFLASH [2010-10-07] wiki is down, see http://lists.openmoko.org/pipermail/community/2010-October/063395.html Apr 27 04:16:00 ~nf2 Apr 27 04:16:01 (here you'll find additional info, if newsflash on ~NF is too short to hold all the info) Apr 27 04:16:11 :-) Apr 27 07:28:53 freesmartphone.org: 03mickey 07aurora * r10c1ac3b4fdb 10/tools/vala-qt-binding-tool/ (11 files in 3 dirs): vala-qt-binding-tool: start w/ some basic thoughts Apr 27 07:34:55 SHR: 03Martin.Jansa 07shr-chroot * rf8716f47a3ca 10/ (1073 files in 28 dirs): system upgrade Apr 27 08:04:51 heyho Apr 27 08:05:13 mickey|bbl: do you saw this http://live.gnome.org/GObjectIntrospection/GObjectConsume ? Apr 27 08:11:40 mickey|office: heyho Apr 27 08:11:48 mickey|bbl: do you saw this http://live.gnome.org/GObjectIntrospection/GObjectConsume ? Apr 27 08:12:52 moin, yeah, someone on the vala list hinted me at that, i'm not sure whether i can make sense out of that though Apr 27 08:13:46 i'll definitely have a look though Apr 27 08:15:20 it compiles fine, as long as you make it suitable for newer gobject-introspection (some type names have changed) Apr 27 08:15:46 feel free to upload it somewhere Apr 27 08:15:53 http://pastie.org/1838469 Apr 27 08:16:17 I just removed the type tags which are no longer available Apr 27 08:17:44 ok, cool Apr 27 08:21:16 mickey|office: I added you with your current tasks to the task list in the wiki Apr 27 08:26:39 right Apr 27 08:27:27 When you start w/ the QML stuff, we need to think about keyboard navigation Apr 27 08:27:46 it looks to me that there hasn't been much thought to that in the existing components Apr 27 08:27:53 probably since all are using touchscreens Apr 27 08:29:58 s/since/because7 Apr 27 08:36:43 jepp Apr 27 08:37:22 what do you mean when talking about keyboard navigation? Apr 27 08:37:31 moving around with the cursor and pressing enter? Apr 27 08:38:28 something like that, yes. the general question of which hardware keys are mapped to which functions in our application framework. how do we move from functions to functions, how do we open menus on non-ts systems, how do we go back one layer etc. Apr 27 08:39:21 even on ts-systems we should at least allow keyboard navigation as well Apr 27 08:39:23 do you already have something in mind when talking about the problems how to solve it? Apr 27 08:39:29 jepp Apr 27 08:40:02 btw. which is the machine you want to develop aurora for? Apr 27 08:40:06 i played around with some things, but due to my mimal amount of QML experience i'm not sure that qualifies as a solution :) Apr 27 08:40:17 ok Apr 27 08:40:29 it looks like the Pre is in best state Apr 27 08:40:46 i'd like to get the E2 running as well Apr 27 08:40:53 but it depends on how much stuff we can share Apr 27 08:40:59 that's why i want to start with statusbar Apr 27 08:41:02 which can be shared Apr 27 08:41:11 ok Apr 27 08:41:15 but the E2 needs a softkey system on the bottom of the key Apr 27 08:41:17 do you saw this: http://wiki.meego.com/QML/Keyboard_navigation ? Apr 27 08:41:20 which is probably not what you want for the others Apr 27 08:41:24 yep Apr 27 08:41:30 ok Apr 27 08:42:01 a softkey system should not be a problem as the launcher is the different component between the machines Apr 27 08:42:10 *nod* Apr 27 08:42:23 we just need to find a way to 'send' the softkey labels to the launcher then Apr 27 08:42:37 from every page for every state the page is in Apr 27 08:42:45 ok Apr 27 08:42:58 and then toggle the corresponding actions, much like if someone would have 'clicked' on the softkey Apr 27 08:43:05 so something to keep in the back of your head Apr 27 08:43:19 i don't mind the softkey approach for important functions Apr 27 08:43:24 even on ts systems it has some virtue Apr 27 08:43:30 since people know exactly where to look for actions Apr 27 08:43:47 but we can defer that discussion to when we start the HIG Apr 27 08:43:49 :) Apr 27 08:43:52 can you maybe do some sketchup how do you think the system should look like Apr 27 08:44:01 yeah. i suck at drawing, but i will Apr 27 08:44:03 :) Apr 27 08:44:10 me too Apr 27 08:44:31 so I started the first aurora build on the amethyst Apr 27 08:44:44 at first only the minimal-image Apr 27 08:44:46 ok Apr 27 08:45:21 a good time for a semipublic announcement might be when we have layed out some sketches and a basic HIG Apr 27 08:45:30 because that's when people can actually start to help Apr 27 08:45:40 w/ the QML stuff Apr 27 08:47:57 jepp Apr 27 08:48:12 people should be able to see how it look and fell Apr 27 08:48:15 s/fell/feel/ Apr 27 08:48:16 morphis meant: people should be able to see how it look and feel Apr 27 08:55:37 yes Apr 27 08:56:24 do we have a rough outline when we can deliver anything? like... when are the preparations (OE, C++ bridge) are being done? what do you think Apr 27 08:56:49 I want something I can use for making a phone call in august Apr 27 08:57:03 so we have three months Apr 27 08:57:37 if we don't get anymore help (with OE for example) it will take some time Apr 27 08:58:00 I already asked stefan if he's interested in doing some OE bits but he don't have any time Apr 27 08:59:09 the preparations should be done quickly Apr 27 09:00:48 maybe JaMa|Wrk has some time to support us with OE work Apr 27 09:02:08 yeah, the ubuntu font is already in OE! Apr 27 09:02:18 ok. in general we should try to recrute new people, since the usual suspects (mroku, gnutoo, etc.) are busy with device support Apr 27 09:02:26 morphis: guess who added it... Apr 27 09:03:15 :) Apr 27 09:03:41 I see you love this font too Apr 27 09:03:57 ubuntu-sans and droidmono are two exciting fonts for mobile systems Apr 27 09:04:05 jepp Apr 27 09:04:25 thanks to my roots, i'm somewhat of a font zealot Apr 27 09:04:27 do you have someone in mind when talking about new people? Apr 27 09:04:37 i want great font rendering Apr 27 09:06:29 (new people) well, i think there are a couple of folks hanging around in here and on the various mailing lists that do not know how to help, since all of our areas (FSO, SHR, OE) require quite a substantial amount of existing knowledge to contribute. If we can take those by the hand and show them that we're starting a new project and start all with the same level of knowledge (zero), it might be a good chance t Apr 27 09:06:29 o get someone new on board Apr 27 09:06:49 morphis: I'll be quite busy with daywork and shr-core for next month and then on holidays and then offline in new flat unless housing association allows me to install antenna on roof and run cable to my flat in first floor.. Apr 27 09:07:22 JaMa|Wrk: ok, too bad :) Apr 27 09:08:35 mickey|office: ok, then we should really do our annoucement as quick as possible to get some new folks on board Apr 27 09:10:27 morphis: that's what i was thinking, although i would wait until we have the sketches. too many degrees of freedom will not lead to constructive discussions IMO Apr 27 09:10:50 so we need a bit more of a plan Apr 27 09:10:56 beforehand Apr 27 09:11:09 right Apr 27 09:11:18 as we want to get something running someday ... Apr 27 09:11:22 indeed Apr 27 09:11:31 this time i want to use my stuff... really. Apr 27 09:13:27 :) Apr 27 09:15:22 it's now over a year since my original announcement Apr 27 09:15:26 (http://www.vanille-media.de/site/index.php/2010/02/27/qt-suddenly-got-interesting-again/) Apr 27 09:15:32 so its time this gets done :) Apr 27 09:16:24 :) Apr 27 09:20:34 btw. I will concentrate for now on the x11 variant, ok? Apr 27 09:20:52 so the aurora-image will curently require x11 Apr 27 09:21:19 we should later divide it by a machine feature which variant we want to build Apr 27 09:21:34 or we do aurora-x11-image Apr 27 09:21:47 and aurora-image Apr 27 09:21:54 to different flavours Apr 27 09:22:26 aurora-x11-image and aurora-fb-image sound good to me Apr 27 09:22:28 it's explicit Apr 27 09:22:39 the aurora-fb-image will require qt-embedded instead of qt then Apr 27 09:22:40 etc. Apr 27 09:22:55 ok Apr 27 09:22:55 it might be best to factor out the commonalities into aurora-image.inc Apr 27 09:23:06 and i'll take care of the -fb variant then Apr 27 09:23:22 however if you could take a look at my qt-embedded problem on the pre... ;) Apr 27 09:23:33 i'd be grateful. Apr 27 09:28:32 hehe, yes I don't forget it :) Apr 27 09:33:00 GarthPS: heyho Apr 27 09:33:14 morphis: hey! Apr 27 09:33:36 morphis: what's up ? Apr 27 09:34:59 GarthPS: it's all fine Apr 27 09:35:03 and you? Apr 27 09:38:01 morphis: well all is getting better and better! with always some challenges of course but what is a life without it? Apr 27 09:38:36 hehe, yeah thats right, we need the little challanges in everyday life Apr 27 09:40:01 morphis: so what's is hot new in SHR/FSO/pre field ? Apr 27 09:40:21 I saw that you did some work on new stuff. waiting to see thtat :) Apr 27 09:41:05 GarthPS: in SHR field nothing is new Apr 27 09:41:17 as you already mentioned mickey|office and I kicked of a new project Apr 27 09:41:38 to get a more stable and less complex base for FSO development Apr 27 09:41:47 and we need help :) Apr 27 09:43:27 morphis: aurora right ? so this will impact the reste of FSO ? I was not sure.. Apr 27 09:43:42 jepp it's named aurora Apr 27 09:43:46 you know zhone2? Apr 27 09:45:17 morphis: nope Apr 27 09:45:23 morphis: euu$ Apr 27 09:45:25 not sure Apr 27 09:45:33 it's the simple FSO test UI Apr 27 09:45:50 yep I know that but I am not sur to have seen it Apr 27 09:46:55 ok Apr 27 09:47:04 aurora is about creating a better zhone2 Apr 27 09:47:22 so we can develop FSO while we see it in action Apr 27 09:47:43 developing a middleware without knowing the right needs of the system on top is not really possible Apr 27 09:47:53 so we need some more simple UI than SHR which uses FSO Apr 27 09:48:01 but is still extensible for new features Apr 27 09:48:33 aurora will run within one process and will be written in Qt/QML C++ + Vala Apr 27 09:48:57 GarthPS: thats zhone2: http://amethyst.openembedded.net/~morphis/images/palmpre-zhone2.jpg Apr 27 09:51:47 morphis: I do know a bit of Qt/C++ programming . is that what you need ? (If I find the time to do that) Apr 27 09:52:11 morphis: yeah I do know zhone2 :) Apr 27 09:53:54 what is the openembedded branch in FSO ? Apr 27 09:55:10 it's for aurora development Apr 27 09:55:22 we will create a image later for supported devices Apr 27 09:55:28 and we will have some release cycle for aurora Apr 27 09:55:38 http://wiki.freesmartphone.org/index.php/Aurora Apr 27 10:08:05 morphis: yop already seen in fact :) and about pre field ? Apr 27 10:16:47 GarthPS: in pre field no news Apr 27 10:19:04 morphis: ok I will try to make a video of the install process soon and then a video of both pre2(SHR vs webos) since I have doctored my pre2 Apr 27 10:19:23 hopping that the ppp connection will work this time Apr 27 10:19:31 ok Apr 27 10:20:49 GarthPS: do have some time to help us getting the OE things for aurora right? Apr 27 10:21:04 that means, testing the image on the device, adjust configuration etc. Apr 27 10:22:30 hmm see week i am nor sure but perhaps . I will ping you . or give me a todo list and I will see Apr 27 10:23:07 in general there is a lot of basic work to do Apr 27 12:24:22 morphis: I will try to do what I can , make me a list and I will tell you more.(a mail ?) Apr 27 12:41:52 dos1: hi Apr 27 12:42:37 dos1: can you check old/dos* on buildhost and remove stuff you don't need anymore (it's mostly ancient..) Apr 27 12:42:49 JaMa|Wrk: hi Apr 27 12:42:59 JaMa|Wrk: ok Apr 27 12:46:02 JaMa|Wrk: done - removed everything, it was all ancient :) Apr 27 12:51:45 dos1: :) you can remove even more if you know why we still kept some of those old/* bits :) Apr 27 12:52:33 dos1: but it has only 16M now.. so sstate-cache dir is growing faster than this every hour :) Apr 27 12:53:27 antrik: I mean it would have been nice to make SHR on palm pre popular before meego ate up the attention. Apr 27 12:53:33 hi Apr 27 12:54:44 JaMa|Wrk: git/shr-overlay is mostly empty, but it has hidden files in .git directory with root as owner Apr 27 12:55:40 should I remove it? Apr 27 12:59:12 JaMa|Wrk: yes, i think Apr 27 13:17:42 <[Rui]> JaMa|Wrk: so how's the new build environment working? :) Apr 27 13:32:47 mickey|office: hey, what's the command that turns on the normal 'verbosity level' (e.g. OK instead of 0) in qualcomm radios and where is it written in fsogsmd ? Apr 27 13:44:24 [Rui]: chroot or shr-core? both fine :) Apr 27 13:47:13 hi JaMa|Wrk, everything that used to build with the old bitbake builds also with my new meta-kde layer. But i need to test if it really works on the freerunner Apr 27 13:48:05 darkh: great, then publish it :) Apr 27 13:48:11 after test ofc Apr 27 13:48:56 darkh: meta-gnome is already in meta-openembedded repo Apr 27 13:49:25 i hope i did not reinvent the wheel :/ Apr 27 13:50:58 that's why I think you should publish it soon (to prevent someone else to reinvent it) Apr 27 13:51:13 good idea Apr 27 13:52:44 oO i have ~50 recipes in my meta-kde layer Apr 27 13:57:45 what's progress with N900 ALSA? Apr 27 13:57:53 and cmtspeech? Apr 27 14:05:42 dcordes: ATV1 Apr 27 14:06:07 dcordes: this is probably in modem_init, which is part of fsogsmd.conf Apr 27 14:06:56 mickey|office: thanks. it doesn't seem to be in fsogsmd.conf : Apr 27 14:06:58 modem_init = E0V1;+CMEE=1;+CRC=1;+CSCS="UCS2";@BRIC=0;+GTKC=2;+CLIP=1 Apr 27 14:07:44 it is Apr 27 14:07:52 v250.ter allows simple commands to be concatenated Apr 27 14:08:01 E0V1 = ATE0; ATV1 Apr 27 14:09:14 ah ok :) Apr 27 14:11:27 is there any way to prevent opkg installing recommended stuff like -dev or -dbg packages? Apr 27 14:14:10 darkh: on target or ie to some OE image? Apr 27 14:14:42 there is BAD_RECOMMENDATIONS for later Apr 27 14:14:46 on target opkg install Apr 27 14:16:02 mickey|office: I am trying to go through the htcleo modem init manually, basing on what I see in fsogsmd Apr 27 14:16:23 ok, good Apr 27 14:16:30 mickey|office: I sent all the commands in modem_init but when trying to dial, I get an error Apr 27 14:16:35 which one? Apr 27 14:16:56 ERROR Apr 27 14:16:59 :( Apr 27 14:19:03 heh Apr 27 14:19:18 AT+CMEE=1 Apr 27 14:19:20 AT+CRC=1 Apr 27 14:19:22 should help Apr 27 14:19:56 if they don't something is wrong Apr 27 14:20:13 since they enable proper error codes Apr 27 14:20:22 or are supposed to Apr 27 14:25:34 I send both commands, still only see "ERROR" on following dial command "ATD*99***1#" Apr 27 14:28:21 mickey|office: ah I already sent CMEE=1 and CRC=1 (as I sent everything from modem_init Apr 27 14:28:34 hmm Apr 27 14:28:39 get me the full log Apr 27 14:28:42 mickey|office: maybe I miss some generic init that is not in the fsogsmd-conf ? Apr 27 14:28:45 are you registered at all? Apr 27 14:29:03 it's a bit difficulat as I'm using echo and cat :> Apr 27 14:29:05 and is the device up? Apr 27 14:29:08 uhm Apr 27 14:29:12 why are you not using a proper tool? Apr 27 14:29:18 any terminal emulator will do Apr 27 14:29:28 I'm not in oe rootfs Apr 27 14:29:36 not sure if I'm registered Apr 27 14:29:38 doesn't matter Apr 27 14:29:43 just build a static one Apr 27 14:29:46 AT+CREG? Apr 27 14:29:47 and copy it over Apr 27 14:29:52 AT+CFUN=1 Apr 27 14:29:59 AT+COPS=0,0 Apr 27 14:30:03 AT+COPS? Apr 27 14:30:37 +CREG: 2 Apr 27 14:30:37 +CREG: 3 Apr 27 14:30:37 OK Apr 27 14:30:37 +CREG: 1,"03BC","0E1EC91A" Apr 27 14:30:55 mickey|office: do you really think it's a good idea to use a common UI for TS devices and traditional keypad phones?... Apr 27 14:32:05 I wasn't registered before it seems Apr 27 14:33:01 mickey|office: now it works, thanks a lot Apr 27 14:33:11 antrik: no, that's probably not a good idea Apr 27 14:33:21 antrik: still reusing as much as possible is a good idea Apr 27 14:33:34 antrik: and adding keypad navigation to touchscreen devices is also Apr 27 14:33:36 dcordes: good Apr 27 14:37:30 mickey|office: I did not think about the fact, that before we had a directy bootloader on the device, windows would register to the network for me. Apr 27 14:38:26 mickey|office: since I can't move forward much with SHR due to lack of sound driver, I'm playing with other distros Apr 27 14:39:59 k Apr 27 14:40:49 it would be quite nice to install recent fso in ubuntu Apr 27 14:45:10 sure, why not Apr 27 14:45:16 all that it needs is packaging i guess Apr 27 14:46:49 mickey|office: then the question is if I can install a decent gui Apr 27 14:50:28 mickey|office: BTW, if you want new contributors, you have to do some outreach. presently we have pretty much an underground community here... Apr 27 14:52:12 antrik: the question is not if we need outreach but how to reach out to people Apr 27 14:52:32 antrik: personally I tried to polish the wiki and motivate people to put easily testable downloads Apr 27 14:54:24 well, the press release idea in february was pretty good IMHO... too bad it didn't actually happen Apr 27 14:55:50 dcordes: BTW, I still don't get your point about MeeGo. what does MeeGo got to do with SHR on Pre?... Apr 27 14:56:00 hi dcordes Apr 27 14:56:14 any news about alsa ? Apr 27 14:56:48 antrik: yeah, underground is right. i have no idea how to improve that though. Apr 27 14:57:09 antrik: is there anything you did in that context ? Apr 27 14:57:14 vicious circle. without something to show you hardly get contributors, and without contributors you can hardly improve your stuff so that you have something to show... Apr 27 14:58:03 GNUtoo|laptop: since you ask cotulla more frequently than me, I doubt I will have news earlier than you Apr 27 14:58:13 GNUtoo|laptop: no news from my side sorry Apr 27 14:58:36 from my side, the alsa was made but not tested on real target Apr 27 14:58:41 that's the last news I got Apr 27 14:58:55 ah I read it in the #htc-linux logs Apr 27 14:59:00 ok Apr 27 14:59:14 I'm sure cotulla will ping us once he has something usable Apr 27 14:59:19 ok Apr 27 14:59:42 mickey|office: well, SHR is not exactly amazing; but it's sure good enough to get contributors. the problem is that the vast majority of people simply do not know it even exists Apr 27 15:00:47 antrik: in that case some folks would need to build more images and advertise them anywhere and advocate the use of them Apr 27 15:01:44 antrik: I don't see what is so hard to understand about it. Apr 27 15:02:03 14:53 < dcordes> antrik: I mean it would have been nice to make SHR on palm pre popular before meego ate up the attention. Apr 27 15:10:34 meego on palm pre....hmmm....but ofono doesn't have good support for the palm pre...they use the AT part of the palm pre modem....which is kind of incomplete and has huge issues Apr 27 15:10:49 s/doesn't have good/none at all/ :) Apr 27 15:11:49 <[Rui]> JaMa|Off: back from sleeping, not 100% cured of the migraine Apr 27 15:12:04 mickey|office: does this look correct? http://paste.pocoo.org/show/379088/ Apr 27 15:12:05 <[Rui]> JaMa|Off: so I should just erase my buildenv and install the new one, right? Apr 27 15:13:10 dcordes: are you saying nobody is interested in SHR on the Pre, because everyone wants to run MeeGo on it?... Apr 27 15:13:58 mrmoku: which problem do you want to solve? Apr 27 15:14:05 antrik: no Apr 27 15:14:26 * mickey|office thinks noone is interested in SHR on the palm, because WebOS is not too bad and semi-hackable as well... Apr 27 15:14:26 mickey|office: fsogsmd aborting due to the net subsystem not yet reachable Apr 27 15:14:40 mrmoku: whois querying for the signal strength? Apr 27 15:14:55 mickey|office: phoneuid probably Apr 27 15:15:16 in any case it should not lead to fsogsmd aborting, right? Apr 27 15:16:27 mickey|office: just to explain... phoneuid is querying signal strength once on startup and then listens to the signals Apr 27 15:16:38 mrmoku: yes, but i'm not sure whether it should rather be intercepted beforehand Apr 27 15:17:20 if you check the dbus_service, we have a state check for every dbus command Apr 27 15:17:34 it might not be a good idea to duplicate those checks Apr 27 15:17:40 but rather intercept the call earlier Apr 27 15:17:41 that sounds even better :) Apr 27 15:17:52 no need to do it in every mediator then Apr 27 15:17:57 yep Apr 27 15:18:09 <[Rui]> mickey|office: judging from a post from Harald Welte, it *may* be possible to eventually get shr into Samsung Galaxy S Apr 27 15:18:29 we can safely assume that getting the signal strength is something we can ignore unless we're in ALIVE_REGISTERED Apr 27 15:18:44 [Rui]: cool, but it's just yet another android system which will give us nightmares Apr 27 15:18:47 [Rui], if you do the same kind of work than morphis with the palm pre, yes Apr 27 15:18:50 (the hardware) Apr 27 15:19:06 our problem is no longer the lack of good hardware to demonstrate Apr 27 15:19:10 mickey|office, I think it should be more standard, but the modem is the big blocker Apr 27 15:19:14 our problem is now to get something finished and to polish Apr 27 15:19:19 yes Apr 27 15:19:19 [Rui]: what's special about Galaxy S ? Apr 27 15:19:23 <[Rui]> mickey|office: yes. Apr 27 15:19:33 <[Rui]> dcordes: it's my job's phone, but it's a very powerfull computer Apr 27 15:20:17 mickey|office: hmm... looking at checkAvailability I don't understand why that happens in the first place Apr 27 15:20:36 if not Modem.Status is wrong somehow Apr 27 15:21:36 mickey|office, mrmoku [fsodevice.kernel_idle] makes fsogsmd segfault on nokia900 Apr 27 15:21:44 what should I do between: Apr 27 15:21:46 *remove it Apr 27 15:21:51 *trace it Apr 27 15:22:30 GNUtoo|laptop: trace it :) Apr 27 15:22:42 ok Apr 27 15:23:43 mickey|office: http://paste.pocoo.org/show/379097/ Apr 27 15:23:45 better? Apr 27 15:25:16 [Rui]: cool Apr 27 15:28:37 [Rui]: do you have a link to thest post ? Apr 27 15:28:42 the Apr 27 15:28:43 bbl Apr 27 15:30:34 mickey|office: actually, I think the big strength of SHR/FSO is the platform support. the more different devices it runs on, the more attractive it gets Apr 27 15:30:49 (as for *demontstration*, I still believe FreeRunner is perfectly fine...) Apr 27 15:31:49 http://www.pastie.org/1839670 Apr 27 15:32:55 http://www.pastie.org/1839672 Apr 27 15:33:32 heh Apr 27 15:33:39 another string is null segfault :P Apr 27 15:33:39 <[Rui]> dcordes: sure... http://laforge.gnumonks.org/weblog/2010/08/21/ Apr 27 15:34:04 indeed Apr 27 15:34:34 antrik: while this is true, you see how crazy platform support is. it takes loads of resources :/ Apr 27 15:34:44 mrmoku: yes, that is good Apr 27 15:34:49 bbiab Apr 27 15:35:37 <[Rui]> antrik: agree perfectly. freerunner is quite nice Apr 27 15:35:41 mrmoku, could you fix it? Apr 27 15:35:42 <[Rui]> not powerful but... Apr 27 15:35:51 and some hardware bugs Apr 27 15:35:58 such as headphone bass filtred Apr 27 15:36:09 that may sound a good idea but it's very bad in practice Apr 27 15:37:18 GNUtoo|laptop: yup, looking into it Apr 27 15:37:35 <[Rui]> captainigloo: elfe looking good (but slower) Apr 27 15:38:15 GNUtoo|laptop: stateIgnoreById[idlestatus.status].length == 0 Apr 27 15:38:17 is the culprit Apr 27 15:39:35 GNUtoo|laptop: does it still segfault if you add Apr 27 15:39:36 ignore_by_id = lis302 Apr 27 15:39:39 to the conf? Apr 27 15:39:45 or ignore_by_id = foobar Apr 27 15:39:51 just to have it filled Apr 27 15:45:25 ok I'll look Apr 27 15:45:43 it probably would segfault on gta02 too without in configured ignores Apr 27 15:46:04 and looking at the code I _think_ I know the correct fix Apr 27 15:46:50 seems to work Apr 27 15:48:27 good Apr 27 15:48:30 heh Apr 27 15:48:30 Collected errors: * satisfy_dependencies_for: Cannot satisfy the following dependencies for fsogsmd: * fsogsmd-module-lowlevel-nokia900 * Apr 27 15:51:06 GNUtoo|laptop: thanks for tracing :) Apr 27 15:51:16 * mrmoku off to give his fish fresh water Apr 27 15:51:17 bbiab Apr 27 15:55:22 mrmoku, do you have an openBSD puffy fish? Apr 27 16:04:34 GNUtoo|laptop: I don't think water agrees well with the habitat of the OpenBSD puffy fish ;-) Apr 27 16:04:57 well, except perhaps in the server's liquid cooling system... Apr 27 16:09:02 GNUtoo|laptop: no, no such fish in our aquarium :) Apr 27 16:09:26 mickeyl: GNUtoo|laptop found out why the kernel_idle plugin segfaults on n900 Apr 27 16:09:45 mickeyl: it segfaults because there are no ignore_by_id configured Apr 27 16:11:06 lol Apr 27 16:11:15 bbl Apr 27 16:11:21 mickeyl: does the following segfault if there are no ignores Apr 27 16:11:22 string ids = stateIgnoreById[ idlestatus.status ]; Apr 27 16:11:22 string[] idsToIgnore = ids.split( ";" ); Apr 27 16:11:44 ids would be null then Apr 27 16:11:54 * mrmoku just tries Apr 27 16:12:31 i would expect an empty array Apr 27 16:12:41 ah, wait Apr 27 16:12:52 ids can be null, yes Apr 27 16:12:59 in that case this code would SIGSEGV Apr 27 16:13:07 ok Apr 27 16:13:10 funny we didn't spot that earlier Apr 27 16:13:27 should i fix or do you want? Apr 27 16:14:15 mickeyl: http://paste.pocoo.org/show/379127/ Apr 27 16:14:20 would be my fix Apr 27 16:15:30 it is stateIgnoreById[idlestatus.status].length == 0 which makes it segfault first Apr 27 16:15:36 can anyone tell me how to deploy all packages with bitbake? i accidently delete deploy/ipk but i know they are still somewhere in some staging directory Apr 27 16:15:45 and unneeded anyway because included by checkForIgnoreInputEvent Apr 27 16:15:57 mrmoku: looks good, thanx Apr 27 16:16:04 good, will commit :) Apr 27 16:17:00 freesmartphone.org: 03mok 07cornucopia * r3c09812e8fa2 10/fsodeviced/src/plugins/kernel_idle/plugin.vala: Apr 27 16:17:00 freesmartphone.org: fsodeviced: kernel_idle: don't segfault when no ignores are configured Apr 27 16:17:00 freesmartphone.org: Signed-off-by: Klaus Kurzmann Apr 27 16:18:46 freesmartphone.org: 03mickey 07utilities * re4cf49712b2c 10/nokian900/ (3 files in 2 dirs): nokian900/cmtspeechtest: add manual mode to synthesize call active signal Apr 27 16:18:48 hmm? Apr 27 16:19:03 *shrug* Apr 27 16:19:43 ? Apr 27 16:20:02 just puzzled why the mac in the office turns on Apr 27 16:20:04 [Rui]: sorry I was on my way home.. Apr 27 16:20:08 ahh, ok :) Apr 27 16:20:14 [Rui]: what is your old buildenv? Apr 27 16:20:17 transfering data to apple probably ;) Apr 27 16:20:20 hahaha Apr 27 16:20:21 :D Apr 27 16:20:29 [Rui]: you can just update shr-chroot if you have it already Apr 27 16:20:50 mickeyl: the thing in fsogsmd did not help though (dbus_service) Apr 27 16:21:16 why not? Apr 27 16:21:41 it still tries to execute the mediator Apr 27 16:21:58 even if the state is less? Apr 27 16:22:48 looking at checkAvailability I don't understand how that should work anyway Apr 27 16:23:05 let me have a look Apr 27 16:23:07 as it does not check for the required argument Apr 27 16:24:01