**** BEGIN LOGGING AT Fri Jan 07 02:59:58 2011 Jan 07 05:25:14 hello Jan 07 06:48:13 playya_ or whoever cares about libeflvala r55869 changed elementary API and vala bindings weren't updated Jan 07 07:40:13 playya_: patch for it http://gitorious.org/~jama/angstrom/jama-shr-experimental/blobs/93c60f83dbccb273b817403db69b569659770f55/recipes/efl1/libeflvala/0001-eflvala-adapt-to-elementary-API-change-from-r55869.patch Jan 07 07:40:21 now iliwi builds again Jan 07 08:27:00 heyho Jan 07 08:39:55 mickeyl: ping Jan 07 08:48:26 JaMa: has the new E some fix for the startup bug? Jan 07 08:55:07 Heinervdm: as I said in that bug I haven't seen that bug, so probably yes, but I don't reboot/reflash lately often enough to be sure Jan 07 08:55:51 JaMa: where is that bug? Jan 07 08:56:40 http://trac.shr-project.org/trac/ticket/1105 Jan 07 08:57:38 JaMa, we just need to add it to e's skel Jan 07 08:57:43 and to the e startup apps Jan 07 08:58:01 ~/.e/e/applications/startup/ Jan 07 08:58:06 put a desktop file there Jan 07 08:58:32 JaMa: ah you are testing that ne e since 3 weeks :) Jan 07 08:58:51 TAsn: we are already using ~/.e/e/applications/startup/ Jan 07 08:59:14 Heinervdm, so why is this bug still open? Jan 07 08:59:25 TAsn: because it's an E bug Jan 07 08:59:34 e has no bugs. Jan 07 08:59:52 btw guys, what's the state of libeflvala? Jan 07 08:59:53 E doesn't start applications form startup reliable Jan 07 09:00:05 Heinervdm, that's odd, on my computer it does. Jan 07 09:00:09 but perhaps it's fixed already Jan 07 09:00:19 at least from what I've noticed. Jan 07 09:00:30 maybe Jan 07 09:00:34 my app is not reliable? Jan 07 09:00:45 it's your app? Jan 07 09:00:48 try wrapping it in a script that echos date into a log Jan 07 09:00:58 Heinervdm, shr-elm-quickpanel Jan 07 09:01:01 or whatever the name is? Jan 07 09:01:56 ah, i thought the startup part Jan 07 09:01:58 TAsn: there is no e's skel.. but we're creating it in ~/.e/e/applications/startup/ for long http://git.shr-project.org/git/?p=shr-wizard.git;a=blobdiff;f=python/shr-wizard.py;h=a07d07f475b3458a4333314cb5d70e9d410ef31b;hp=6867acbc3319b4663a92789d4c73bd367cf023b6;hb=0342a0b36ab3df7e980fe7a045c905cee8f2a7e8;hpb=6a85f0b064e859aef5ae4b504f01d189de5bfbd7 Jan 07 09:02:32 TAsn: and see that bug for info from wrapped script Jan 07 09:03:17 JaMa, i.c Jan 07 09:03:20 and imho it's related to efreet cache and I have efreet-icon-cache enabled only locally (that's maybe reason why I haven't seen it) Jan 07 09:03:24 ok, I have nothing to contribute then. :P Jan 07 09:04:05 JaMa: does the efreet cache has disadvantages? Jan 07 09:04:31 somehow I'm missing icon theme configuration in illume config.. so I have no icons on desktop (unless .desktop file has full path to icon) Jan 07 09:05:03 and I haven't time to check why it's missing or why no icon is found Jan 07 09:05:05 that's bad :D Jan 07 09:18:52 JaMa, is you patch already upstream? Jan 07 09:21:14 playya_: no, I don't have RW there Jan 07 09:22:13 ok. cloning e again Jan 07 09:41:47 *yawn* morning Jan 07 09:42:06 moin Jan 07 09:43:49 won the fight against GValues? Jan 07 09:44:26 somehow Jan 07 09:44:32 at least phoneui is working again :-) Jan 07 09:44:38 phonefsod is still segfaulting Jan 07 09:44:44 does not on my laptop though :/ Jan 07 09:44:55 I don't think it has to do with gvalues though Jan 07 09:45:15 hmm, ok Jan 07 09:45:28 race condition due to different timtings? Jan 07 09:45:36 hmm Jan 07 09:46:00 I think more that on my laptop glib is getting away with some critical warning where on the target that will cause a segfault Jan 07 09:46:24 ok Jan 07 09:46:25 could be Jan 07 09:46:27 * mrmoku remembers to have had that already Jan 07 09:49:03 debugging on the target is a pain though :( Jan 07 09:49:12 it is Jan 07 09:49:18 maybe I should come back to Gnutoo and set up gdbserver correctly Jan 07 09:49:34 yeah, for some problems there is no alternative Jan 07 09:49:41 although you can of course make critical fail on the laptop as well Jan 07 09:49:45 g_critical, that is Jan 07 09:49:59 hmm... that might be an idea too Jan 07 09:50:10 or run it from gdb Jan 07 09:50:13 and set a breakpoint Jan 07 09:50:14 maybe first get it error free on the laptop Jan 07 09:50:20 yeah Jan 07 09:52:46 freesmartphone.org: 03morphis 07next * r4a7d1cd4d438 10msmcomm/msmcommd-specs/ (9 files in 2 dirs): msmcommd-specs: remove old stuff for a fresh start Jan 07 10:06:10 mickey|office: can you cherry-pick bdb28f886018e88066628ad72a01e0b2ce12121c to dbus-glib branch of cornucopia? Jan 07 10:09:24 hm.. Jan 07 10:09:30 we need a network of "trusted people" Jan 07 10:09:33 in the open source world Jan 07 10:09:40 freesmartphone.org: 03mickey 07dbus-glib * r463448789e33 10cornucopia/libfsobasics/ (6 files in 3 dirs): libfsobasics: make it compile against Vala 0.11.2 Jan 07 10:09:41 so people that are known to be legit Jan 07 10:09:44 done Jan 07 10:09:46 will have access to do Jan 07 10:09:48 some actions Jan 07 10:09:50 mickey|office: thx Jan 07 10:09:54 in all the open source projects in the world. Jan 07 10:10:00 :P Jan 07 10:10:07 such as this cherry pick. Jan 07 10:10:16 TAsn: the opensource elite ;) Jan 07 10:10:17 true Jan 07 10:10:31 which is a matter of expertise too though Jan 07 10:10:32 not only trust Jan 07 10:10:59 Heinervdm, I was thinking something more in the lines of "The Justice League" Jan 07 10:11:04 and my expertise is for sure not spanning over the whole open source world :P Jan 07 10:11:05 mrmoku, well, trust someone to not fuck up Jan 07 10:11:13 yeah, ok Jan 07 10:11:15 as a fuck up Jan 07 10:11:18 I consider Jan 07 10:11:22 cherry picking a good feature Jan 07 10:11:27 into a branch with a feature freeze Jan 07 10:11:28 :) Jan 07 10:11:28 I probably would be sane enought to not try to cherry pick gcc commits ;) Jan 07 10:11:32 hehe Jan 07 10:11:48 that settles it then. Jan 07 10:12:33 We will create this system and OSS will advance once more. Jan 07 10:13:59 mickey|office: ping Jan 07 10:14:06 at your servie Jan 07 10:14:09 great Jan 07 10:14:09 service, even Jan 07 10:14:16 good morning btw ... Jan 07 10:14:20 morsche Jan 07 10:14:36 :) Jan 07 10:14:49 as you may know I am currently reworking most msmcomm stuff Jan 07 10:15:03 yeah. tearing it apart and building it up -time Jan 07 10:17:48 and I want to touch the msmcomm dbus interface too Jan 07 10:18:14 ok Jan 07 10:19:10 what do you think about a more generic API like enqueueCommand("change_operation_mode", new HashTable { "mode" : "reset" }) Jan 07 10:19:34 would cut down the effort to maintain the dbus api Jan 07 10:20:12 which includes adding new structs for message arguments and so on ... Jan 07 10:21:48 hmm Jan 07 10:21:54 in general i'd say this is wrong wrong wrong Jan 07 10:22:01 Hi, I would like to recover my old gta01v4 from the dust and use it for my mockup project for which I need rs232, so the question is, where I could easily solder/hookup tx/rx/gnd pins? :) instead of the agps? (I don't need GPS) Jan 07 10:22:03 since essentially you're killing the API contract then Jan 07 10:22:12 it's like making everything accepting a{sv} Jan 07 10:22:15 which is equally bad Jan 07 10:22:17 _however_ Jan 07 10:22:33 in the specific situation of fsogsmd communicating with msmcommd Jan 07 10:22:38 i'd say we can do that Jan 07 10:22:40 however Jan 07 10:22:53 please use string marshalling or enums Jan 07 10:22:57 for the command Jan 07 10:23:01 so we get at least a bit of safety Jan 07 10:23:05 ynezz: you can use the serial console pins Jan 07 10:23:33 morphis: the problem with such flexible interfaces is that while they are appealing, you will eventually have to add parameter validation in both places for the server and the client Jan 07 10:23:45 by using stricter interfaces, you can get dbus/the compiler do these checks for you Jan 07 10:23:47 ynezz: it would be the easiest. But you'll lose connectivity with the GSM modem afair, they're sharing the same port. Jan 07 10:23:55 but as i said, in the specific situation, this might be ok Jan 07 10:24:35 PaulFertser: thanks, couldn't find it in the schematics and I don't need GSM anyway Jan 07 10:26:28 ynezz: there're big easily accessible testpoints for that near the debug connector. Jan 07 10:26:58 ynezz: but that's not rs232 of course, rather a 2.8V (or was it 3.3?) UART. Jan 07 10:27:28 yep, I need ttl uart :) Jan 07 10:32:24 mickey|office: the parameter validation is one point but I think its not as important in this situation as on others Jan 07 10:32:44 as we are the only one using the dbus api Jan 07 10:33:37 yes Jan 07 10:34:22 hm Jan 07 10:35:03 it occurs to me that by having kept it as a library it would have been easier ;) Jan 07 10:35:06 but don't let that stop you :) Jan 07 10:35:24 hm Jan 07 10:36:03 TAsn: today is friday... you could make libphone-ui work with different backends ;) Jan 07 10:36:26 IIRC your argument for that was the multiple clients, but the library could have kept the command queues so that the assignment for which response packet goes to which client would have been mappable Jan 07 10:36:38 mrmoku, ha, I wish I had the time :| Jan 07 10:36:47 I'm currently doing tons of h.w Jan 07 10:36:48 :| Jan 07 10:36:51 s/library/msmcommd server side/ Jan 07 10:36:51 mickey|office meant: IIRC your argument for that was the multiple clients, but the msmcommd server side could have kept the command queues so that the assignment for which response packet goes to which client would have been mappable Jan 07 10:36:51 s/library/server/ Jan 07 10:36:51 :P Jan 07 10:36:52 mickey|office meant: IIRC your argument for that was the multiple clients, but the server could have kept the command queues so that the assignment for which response packet goes to which client would have been mappable Jan 07 10:36:55 mickey|office: thats even the same with the generic command above Jan 07 10:36:55 err Jan 07 10:37:25 yes, sorry, this is completely orthogonal Jan 07 10:37:37 :) Jan 07 10:37:43 i just remembered that i wanted to say this at some point of time ;) Jan 07 10:38:51 you are right, we can always do the mapping of command/responses in the command queue and don't need the dbus stuff, I just thought it is a better abstraction and dealing with a network interface Jan 07 10:39:22 it could have been interesting to make it a real network interface Jan 07 10:39:34 and talk a command language Jan 07 10:39:38 as doing the commandqueue stuff in any fso component we need the msmcomm protocol to do thinks Jan 07 10:39:38 that's basically what PHONET is Jan 07 10:39:43 would be nice Jan 07 10:39:50 but phonet is kernelspace, right? Jan 07 10:39:55 yes Jan 07 10:40:10 the transport mechanism is done in kernel space Jan 07 10:40:23 is it possible to implement network interface in userland? Jan 07 10:40:27 tun/tap? Jan 07 10:40:33 yeah possibly Jan 07 10:40:44 but that will throw us back even harder Jan 07 10:40:52 right Jan 07 10:40:52 i.e. will take much longer to get to something usable Jan 07 10:41:00 maybe Jan 07 10:41:15 so I'm in favor of the minimum amount of changes to get things working, but also maintainable Jan 07 10:41:27 so in essence you need to decide as you know the architecture best Jan 07 10:41:36 as I very demotivated when it's to much effort to implement new messages / test new stuff ... and this it is in the current situation Jan 07 10:41:41 yep Jan 07 10:41:51 s/as I/as I am/ Jan 07 10:41:52 morphis meant: as I am very demotivated when it's to much effort to implement new messages / test new stuff ... and this it is in the current situation Jan 07 10:42:00 if it brings you something in the midterm Jan 07 10:42:02 by all means do it Jan 07 10:42:18 i rewrote the plugin architecture in FSO 4 times before committing it Jan 07 10:42:20 :) Jan 07 10:43:48 hehe Jan 07 10:44:06 and it's great :) Jan 07 10:45:22 it works, yeah ;) Jan 07 10:45:33 i love things you can forget about... Jan 07 10:45:43 since i do anyways hehe Jan 07 10:46:00 hehe Jan 07 10:46:12 anyways, comeing back to the issue Jan 07 10:46:19 you can do everything you like until we have many mediators Jan 07 10:46:21 thats what I am trying with libmsmcomm-next ... it's too much effort to implement messages in C as in Vala Jan 07 10:46:33 once we have many mediators implemented, every refactoring is going to be hard work Jan 07 10:46:38 right Jan 07 10:46:39 s/refactoring/change/ Jan 07 10:46:39 mickey|office meant: once we have many mediators implemented, every change is going to be hard work Jan 07 10:47:12 do you ever worked with tun/tap? Jan 07 10:47:30 not really, no Jan 07 10:48:11 but honestly it wouldn't bring too much Jan 07 10:48:22 what we gain with dbus, we need to reimplement then Jan 07 10:48:33 like, process control, activation/deactivation, protocol marshalling Jan 07 10:48:46 I am even on the way reimplement the dbus API Jan 07 10:48:54 heh Jan 07 10:49:02 but you are right we are more flexible with the dbus api Jan 07 10:49:19 as we need even some control mechanisms for the msmcommd daemon Jan 07 10:49:20 it would be another situation, if we were to implement PDP Jan 07 10:49:45 for PDP dbus would mean a _huge_ overhead compared to an interface Jan 07 10:49:59 but we're only using it for control messages which are relatively small Jan 07 10:49:59 but we don't have to Jan 07 10:50:05 right Jan 07 10:50:40 I think I will implement the dbus api as before but better designed Jan 07 10:51:02 ok, sounds good Jan 07 10:52:12 will split it into org.msmcomm.Misc, org.msmcomm.State, org.msmcomm.Call, ... instead of a simple org.msmcomm.Commands Jan 07 10:53:02 good Jan 07 10:54:49 btw. OE supports now the palmpre2 too Jan 07 10:59:16 ya, seen the commits. nice work Jan 07 10:59:30 will there be any other differences but kernel? Jan 07 11:01:08 as far as I know, no Jan 07 11:01:19 3d Jan 07 11:01:25 is one point Jan 07 11:01:30 so another kernel module Jan 07 11:01:45 touchscreen seems to be the same Jan 07 11:02:06 but I would really like to put my fingers on a palmpre2 device :) Jan 07 11:04:02 ebay? Jan 07 11:05:00 money? Jan 07 11:05:17 ya, 400 is quite a lot Jan 07 11:05:33 I am waiting for palm offering a pre2 for developers for a cheaper price (like 200$ in US) Jan 07 11:05:47 http://cgi.ebay.com/Palm-Pre-2-Unlocked-/270689185814?pt=Cell_Phones&hash=item3f06553016 Jan 07 11:05:56 US $51.00 Jan 07 11:06:12 that's not too bad Jan 07 11:06:19 everything under 200 would be good Jan 07 11:06:20 ah, its a coupon code Jan 07 11:06:32 oh Jan 07 11:06:45 * pabs3 bah Jan 07 11:06:49 freesmartphone.org: 03morphis 07next * r390c4810539a 10msmcomm/msmcommd-specs/src/ (Makefile.am misc.vala): msmcommd-specs: add specification for misc commands Jan 07 11:06:51 freesmartphone.org: 03morphis 07next * r7b44a62bdeb4 10msmcomm/ (28 files in 4 dirs): Rename msmcommd-specs to msmcomm-specs Jan 07 11:07:14 pabs3: right its coupon, but you only can use it when you are in US Jan 07 11:07:46 mickey|office: maybe we should a start a donation offer like webos-internals when we have something for the pre which is usable Jan 07 11:08:17 there is one for $213.50 (box unopened) Jan 07 11:12:40 morphis: yes, and also for the n900 Jan 07 11:14:37 right Jan 07 11:14:48 flattr would be a nice option for this Jan 07 11:14:58 as we can get montly donations Jan 07 11:15:18 and not only to specific points in time a defined amount Jan 07 11:20:48 morphis: hi ! You are welcome in Bretagne :) Jan 07 11:21:00 Brest is not that far :p Jan 07 11:21:12 GarthPS: :) Jan 07 11:21:37 GarthPS: I am wanting to go there but that far away .... Jan 07 11:22:01 morphis: but did you asked to rwhitby ? perhaps there is one for you in their donation operation ? Jan 07 11:22:59 morphis: I read they colected a lot of money , around 8000$ Jan 07 11:24:05 hm, maybe we should ask him, mickey|office? Jan 07 11:24:42 GarthPS: btw. OE has now support for the palmpre2 Jan 07 11:25:02 morphis: always worth a try Jan 07 11:25:08 GarthPS: but please test the kernel from here first: http://amethyst.openembedded.net/~morphis/oe/palmpre/deploy/images/palmpre2/ Jan 07 11:25:12 however it'd be better if we could already show something before asking Jan 07 11:25:28 morphis: yop read too! so I can build me a kernel for my pre2 directly right ? Jan 07 11:25:36 GarthPS: right Jan 07 11:25:40 mickey|office: would be better Jan 07 11:26:24 morphis: other than kernel , pre2 in OE does not differe from pre1 for the moment right? after I tweak some config files it will be different right ? Jan 07 11:26:42 which config files? Jan 07 11:26:55 it's the same kernel with a different defconfig Jan 07 11:27:03 everything else is the same Jan 07 11:27:08 freesmartphone.org: 03Frederik.Sdun 07cornucopia * rc371bb1c08e5 10/libfsobasics/fsobasics/binary.vala: fsobasics: Add BinReader class Jan 07 11:27:10 freesmartphone.org: 03Frederik.Sdun 07cornucopia * r921dcbb69b56 10/libfsobasics/tests/ (Makefile.am testbinreader.vala): fsobasics: Add BinReader tests Jan 07 11:27:10 morphis: i mean in rootfs Jan 07 11:27:11 freesmartphone.org: 03Frederik.Sdun 07cornucopia * r62a0d169a685 10/fsodeviced/ (4 files in 4 dirs): Merge branch 'master' of fso:cornucopia Jan 07 11:27:42 GarthPS: ok, but which? Jan 07 11:27:48 you mean for fso? Jan 07 11:27:48 morphis: mickey|office : I have plan to make a video of FSO/SHR on my pre2 on day Jan 07 11:28:03 it could be a point to show to rwhitby Jan 07 11:28:22 would be nice Jan 07 11:28:41 morphis: like powerbutton briging the e menu Jan 07 11:32:43 GarthPS: so you have local modifications to the fso config? Jan 07 11:35:04 morphis: no no i have nothing right now. I wanted to say that after making fso-installer complet and some howto install on a pre i will help to adapt FSO/SHR to th pre2, which means some tweaking : resolution font etc etc Jan 07 11:35:47 ah ok Jan 07 11:35:58 I will try your new kernel in half an hour Jan 07 11:36:00 jepp, then we have to decide which stuff is device specific Jan 07 11:36:03 ok Jan 07 11:39:03 GarthPS: (video) pretty cool, yes please Jan 07 11:40:14 larsc: hi, i am now testing 2.6.37, with your gta02_defconfig it works ok, but if i try mine config used in qtmoko distro with 2.6.34 kernels i get WS Jan 07 11:40:42 larsc: i can ssh to neo - so it's not WSOD, but i wonder what can be wrong Jan 07 11:43:06 radekp: uhm, a lot of things ;) try to make a diff of both configs Jan 07 11:43:28 larsc: i am doing, but i cant see anything obvious :) Jan 07 11:43:44 larsc: you just regenerated the old config for 2.6.37? Jan 07 11:44:07 freesmartphone.org: 03morphis 07next * r986fced83ce6 10msmcomm/msmcomm-specs/src/Makefile.am: msmcomm-specs: fix makefile for compilation issues Jan 07 11:46:07 larsc: the diff is like this http://pastebin.com/6KmGTz08 Jan 07 11:47:32 larsc: if you dont have time never mind, i'll try to find what is wrong using combinations :) Jan 07 11:48:11 CONFIG_SAMSUNG_GPIO_EXTRA=64 Jan 07 11:48:47 i am testing it now :) Jan 07 11:49:23 hmm still the same Jan 07 11:50:52 mickey|office: can you cherry-pick bdf1abf9c0843a14d6f50bf5275db796d980b43e and 759585d937828095f60d503a1e4b683bf51e843e too? Jan 07 12:30:14 morphis: new kernel booted Jan 07 12:30:24 morphis: what do you want me to check ? Jan 07 12:30:42 that it boots :) Jan 07 12:31:11 as we now can build kernel for palmpre and palmpre2 out of the same git branch with only another defconfig Jan 07 12:32:44 freesmartphone.org: 03morphis 07next * rf287e3648c2a 10msmcomm/msmcomm-specs/ (4 files): msmcomm-specs: some more fixes for compilation issues Jan 07 12:32:46 freesmartphone.org: 03morphis 07next * r4145be3c89ef 10msmcomm/msmcommd/ (9 files in 2 dirs): msmcommd: make it compile with next generation msmcomm-specs and libmsmcomm Jan 07 12:32:47 freesmartphone.org: 03morphis 07next * r12acf73fb46c 10msmcomm/msmcomm-specs/src/management.vala: msmcomm-specs: add not implemented exception Jan 07 12:34:32 morphis: ok so that is all for the moment ? no log. can I reboot my phone ? :) Jan 07 12:40:12 GarthPS: log would be great if you have some time Jan 07 12:40:21 about SMS sending and phonecalls Jan 07 12:40:29 and evtl. GPRS Jan 07 12:40:47 larsc: hmm interesting, it seems that if compile bluetooth as module it does WS Jan 07 12:41:08 morphis: yeah sure but for that i need to reboot into webos right :) perhaps this week end Jan 07 12:41:10 freesmartphone.org: 03Frederik.Sdun 07cornucopia * r424daf45e128 10/libfsobasics/fsobasics/ (Makefile.am checksum.vala): fsobasics: Import crc16 code of morphis msmcomm implementation Jan 07 12:41:26 larsc: if you use the gta02_defconfig and select bluetooth as module it will WS Jan 07 12:41:26 GarthPS: would be really nice! Jan 07 12:41:43 GarthPS: and I hope I will have webos 2.0 some day ... Jan 07 12:41:50 morphis: ok I will ping you in time Jan 07 12:42:25 morphis: yeah you saw that there is a webos 2.0 for a pre somewhere ? don't rememeber where Jan 07 12:43:05 radekp: thats rather strange Jan 07 12:43:13 GarthPS: no there is no, I tried the webos 2.0 from the pre2 on my pre Jan 07 12:43:31 but i'll try it later today Jan 07 12:43:31 morphis: ahh :) Jan 07 12:43:49 larsc: i think so too, maybe it has to do something with kernel size Jan 07 12:44:23 morphis: http://www.precentral.net/webos-2-0-spotted-telcel-pre-mexico Jan 07 12:44:28 larsc: oki, thanks, i'll let you know if i find something interesting Jan 07 12:46:17 GarthPS: but nothing to download Jan 07 12:47:55 morphis: ... :) Jan 07 12:48:18 freesmartphone.org: 03morphis 07next * rf06c948921cb 10msmcomm/msmcommd/src/ (4 files): msmcommd: add stub implementation for misc service Jan 07 12:49:53 morphis: I need your help 2" Jan 07 12:50:31 how to make this work as generic? novaterm boot mem:// < build/unpacked-doctor/resources/nova-installer-image-*.uImage as it could be #nova-installer-image-castle.uImage Jan 07 12:50:31 or #nova-installer-image-roadrunner.uImage Jan 07 12:51:13 I want it to be user'sbrainless Jan 07 12:51:16 :) Jan 07 12:55:50 GarthPS: add a variable for it depending on the set TARGET Jan 07 12:56:05 when TARGET = palmpre then NOVA_ISNTALLER=nova-installer-iamge-castle.uImage Jan 07 12:56:43 morphis: juste like this (syntax correct?)? Jan 07 12:57:36 because i already tried to define a variable's value depending of TARGET but did not succed so ... Jan 07 12:58:32 show me the code Jan 07 13:00:50 morphis: of my try ? already away because not working. but here is what i have acutally http://pastebin.com/t3anTNnB Jan 07 13:01:48 CONFIG_WEBOS_MEDIA_TARGET_SIZE = 13.0G Jan 07 13:01:55 thats something which is device specific too Jan 07 13:02:04 larsc: if i increase the value in msleep(100) in your WS fix it starts working Jan 07 13:02:22 GarthPS: on Pre you only have 8GB, on Pre Plus and Pre2 you ahve 16GB Jan 07 13:02:27 morphis: yop that is why I did some warning Jan 07 13:02:43 set another default depending on TARGET Jan 07 13:03:36 GarthPS: btw. here is a good example for writing such a Makefile: http://git.webos-internals.org/?p=tools/meta-doctor.git;a=tree Jan 07 13:03:53 it will help you on many cases Jan 07 13:04:11 mickey|office: ping Jan 07 13:04:49 hmm if it wasnt random... Jan 07 13:05:52 morphis: crap. i don't understand i did exacly the same as rwhitby to define his target with some ifeq but it did not work .... Jan 07 13:06:05 I will retry it Jan 07 13:06:17 hm Jan 07 13:11:46 freesmartphone.org: 03Frederik.Sdun 07cornucopia * rf520d1e72354 10/libfsobasics/fsobasics/binary.vala: fsobasics: add crc16 methods for BinBuilder Jan 07 13:11:52 freesmartphone.org: 03Frederik.Sdun 07cornucopia * r9ca6d6556ac6 10/libfsobasics/fsobasics/binary.vala: fsobasics: add crc16_verify for BinReader Jan 07 13:13:36 morphis_: ok ..it work.. perhaps a retunr missing.. about CONFIG_WEBOS_MEDIA_TARGET_SIZE it could be easier do it at the opposite: define SHR space instead , no ? Jan 07 13:14:14 maybe Jan 07 13:14:31 yeah you are right, that should be the correct way Jan 07 13:14:51 morphis_: because yes it is device specifique but what to do if the user specified a bad size ? Jan 07 13:15:22 larsc: i think i found it, i am using qi with 242 timings here, so maybe for you it will work Jan 07 13:15:37 larsc: but after applying this patch https://github.com/radekp/linux-2.6/commit/475a5b42a45c7935adfca735fa6caf8e3c12c7bd it works ok Jan 07 13:15:51 I think the best way is to check it and trow an error when it is too big Jan 07 13:16:02 morphis_: I think the best way is to check it and trow an error when it is too big Jan 07 13:16:09 larsc: btw athor of that patch is gena2x not me (forgot to add --author) Jan 07 13:16:58 GarthPS: jepp Jan 07 13:17:29 morphis_: or yeah juste do what i just said because if we define ths SHR space it will not resolv the issue of user specifying wrong size Jan 07 13:18:04 morphis_: so for me we just need to check and trow away if wrong Jan 07 13:18:12 good morning Jan 07 13:18:26 good afternoon :) Jan 07 13:19:07 what would take to use desktop notifications in phoneuid? Jan 07 13:19:13 does it still use the system dbus? Jan 07 13:19:33 GarthPS: ok Jan 07 13:19:37 i have my panel working with freedesktop notifications, it's a shame i can't use it for messages :) Jan 07 13:19:39 morphis_: hm...in fact we need to check too thing : enough space for webos and enough space for FSO/SHR Jan 07 13:19:55 mrmoku, hey Jan 07 13:20:38 GarthPS: enough space only when the internal partition has very limited free space Jan 07 13:20:58 but I think it's something the user should care about on it's own Jan 07 13:21:14 daniele_athome, an applications should be able to use system and session bus at the same time Jan 07 13:21:18 if he is using the makefile he should do a backup anyway and deal with the consequences Jan 07 13:22:07 playya_, i know, but this will make an inconsistency, right? (sorry if i mispelled) Jan 07 13:22:28 i mean ui comunication should happen on session bus too Jan 07 13:22:43 i guess for now i should do that way (use both as you said playya_ :) Jan 07 13:23:10 yes Jan 07 13:23:34 but an application can use fso on system bus and send notifications on session bus Jan 07 13:24:04 phew. still importing efl into git Jan 07 13:24:30 indeed Jan 07 13:24:35 playya_, efl into git? Jan 07 13:25:13 the enlightenment svn repository into a local git repository Jan 07 13:25:30 playya_: full efl into git? Jan 07 13:25:51 why? why not a only project? Jan 07 13:25:58 btw. maybe we should generate an additional library with common dbus interfaces Jan 07 13:26:31 morphis_: how much space do we for FSO/SHR ? Jan 07 13:26:31 iirc git svn doesn't support it Jan 07 13:26:58 no? I imported only elementary a long time ago Jan 07 13:27:11 http://git.gravedo.de/?p=elementary.git;a=summary Jan 07 13:28:12 GarthPS: for FSO/SHR, hm 300-500Mb should be enough for / Jan 07 13:30:24 morphis_: for /dev/store/fso Jan 07 13:30:36 it is the same right Jan 07 13:31:01 ok Jan 07 13:33:21 JaMa, morning Jan 07 13:35:01 daniele_athome: moin Jan 07 13:38:58 freesmartphone.org: 03morphis 07next * r7ea437639164 10msmcomm/libmsmcomm-next/msmcomm/messagetype.vala: libmsmcomm-next: add more type to string translations for message types Jan 07 13:39:39 JaMa, i was thinking about adding support for freedesktop notifications at least to messages Jan 07 13:39:52 any prior indications before i start digging into the code? :) Jan 07 13:48:51 freesmartphone.org: 03mickey 07dbus-glib * rd1dd25ba29bc 10cornucopia/tools/mterm2/src/reader.vala: mterm2: catch up with vala changes Jan 07 13:48:53 freesmartphone.org: 03mickey 07dbus-glib * r0eb6f8ada786 10cornucopia/tools/mterm2/src/reader.vala: mterm2: fix damage introduced by last commit Jan 07 13:49:20 daniele_athome: hey Jan 07 13:49:41 daniele_athome: are there still any reasons for you not to use libfso-glib with gdbus? Jan 07 13:53:34 mrmoku, what do you mean? Jan 07 13:54:03 i didn't dig into dbus anyhow yet, actually mokosuite development is quite slowing down this month, because of other work :( Jan 07 13:54:09 freesmartphone.org: 03morphis 07next * r988a4ee15f62 10msmcomm/libmsmcomm-next/ (6 files in 3 dirs): libmsmcomm-next: add all known misc subsystem unsolicited responses Jan 07 13:54:12 s/dbus/gdbus/ Jan 07 13:54:12 daniele_athome meant: i didn't dig into gdbus anyhow yet, actually mokosuite development is quite slowing down this month, because of other work :( Jan 07 13:54:38 mrmoku, any persuasive thoughts? :P Jan 07 13:55:08 daniele_athome: nothing special on my mind Jan 07 13:57:33 daniele_athome: we dumped libframeworkd-glib as libfso-glib in the gdbus version works fine for everything Jan 07 13:57:50 daniele_athome: question is if you still need your own autogenerated lib? Jan 07 14:05:20 mrmoku, well libframeworkd-glib is quite old and unmaintained Jan 07 14:05:41 but i will indeed take a look to gdbus port of libfso-glib Jan 07 14:06:15 and then decide to drop my lib :) Jan 07 14:06:51 oki :) Jan 07 14:11:50 * mrmoku off to buy some fishes :) Jan 07 14:11:56 -es Jan 07 14:12:24 mrmoku, the fact is i don't like all the *_finish stuff Jan 07 14:12:34 and freeing all the things Jan 07 14:12:40 i'd like the library do those things for me :) Jan 07 14:12:46 that's why i built that generator Jan 07 14:13:01 so i can rid of most tedious things about dbus callbacks Jan 07 14:13:39 ok, will take a look at your generator then... to see what it does :) Jan 07 14:15:32 :P Jan 07 14:15:33 :) Jan 07 14:15:58 mrmoku, anyway i think it's far from perfect, but it does some funny things :) Jan 07 14:33:28 mickey|office: so, sending/recieving message with new msmcommd/libmsmcomm is working Jan 07 14:35:12 freesmartphone.org: 03morphis 07next * rb17f05a42bad 10msmcomm/msmcommd/src/ (7 files): msmcommd: adopt message pack and unpack logic to new libmsmcomm style Jan 07 14:35:12 freesmartphone.org: 03morphis 07next * r5754b4157c5f 10msmcomm/msmcomm-specs/src/misc.vala: msmcomm-specs: add unsolicited response for misc service Jan 07 14:42:19 This segfault is module.so, isn't it? http://shr.pastebin.com/nfYyxpXP Jan 07 14:44:01 depends on how badly is stack corrupted Jan 07 14:45:59 JaMa: so i should install evas-dev? Jan 07 14:51:32 so I am off for today Jan 07 14:51:34 byebye Jan 07 14:52:55 Heinervdm: looks like you have it already Jan 07 14:53:06 Hi Jan 07 14:53:17 JaMa: nothing changed Jan 07 14:53:36 Heinervdm: but I would check stack before segfault if you can reproduce it in evas_common_font_int_reload Jan 07 14:54:03 I have a little problem: SHR-t seems to be working fine without a SIM card present, but whenever I put a SIM card into my neo, enlightenment keeps crashing. Is there a way to solve this? Or is SHR-u better for daily use? Jan 07 14:55:00 JaMa: do you have your Neo with you? Jan 07 14:55:48 no, but it's at home running Jan 07 14:56:54 the bug occurs when you power up the phone then press power and choose shutdown in quickmenu Jan 07 14:57:02 then phoneuid segfaults Jan 07 14:57:18 new phoneuid with gdbus? Jan 07 14:57:27 no shr-testing phoneuid Jan 07 14:57:37 and only after boot Jan 07 14:57:42 ah :/ Jan 07 14:57:49 if you try it again after that it's working fine Jan 07 14:58:15 so it's a bit strange :) Jan 07 14:59:12 does it happen with current shr-u too? Jan 07 14:59:32 don't know Jan 07 14:59:46 but i can try Jan 07 14:59:58 that would be something I can test on n900 if gf has some uSD reader at home :) Jan 07 15:00:12 i need a second uSD :) Jan 07 15:00:42 JaMa: tests/shr-u or normal shr-u? Jan 07 15:01:02 better start with normal shr-u Jan 07 15:01:17 with tests/shr-u you maybe won't get that far :) Jan 07 15:01:30 ok :) Jan 07 15:02:25 it should have same E version Jan 07 15:06:10 yamitenshi: that problem is strange Jan 07 15:06:22 never heard of sth like that Jan 07 15:07:08 Heinervdm, maybe it's a problem with my SIM card then? Jan 07 15:07:16 it could be related to signal indicator Jan 07 15:07:31 we had similar issue at least once before Jan 07 15:08:39 yamitenshi: you can try shr-u or shr-testing from tests Jan 07 15:08:53 and try if the error occours there too Jan 07 15:10:12 Heinervdm, Will do, thanks. Didn't know there was a tests directory too, is it for development builds? Jan 07 15:11:42 Yay! Pizza! Jan 07 15:12:47 yamitenshi: yes, so that's not stable and when you want to install packages you have to change the config Jan 07 15:13:10 add the tests/ to the configs in /etc/opkg Jan 07 15:15:15 JaMa|Off: hmm... somebody reenabled introspection? Jan 07 15:15:21 libgee build is failing Jan 07 15:16:34 | /OE/tmp/sysroots/x86_64-linux/usr/bin/g-ir-compiler: error while loading shared libraries: libffi.so.4: cannot open shared object file: No such file or directory Jan 07 15:18:09 mrmoku: it gets autodetected as enabled if you build gobject-introspection first, but nothing should depend on non-native gobject-introspection now Jan 07 15:18:37 hmm Jan 07 15:18:40 * JaMa|Off is just finishing rebuild from scratch without libgee issue Jan 07 15:18:55 * mrmoku manually built libffi Jan 07 15:19:06 but if you've built gobject-introspection manually then you have to disable introspection in few recipes Jan 07 15:19:31 my armv4t build was quite old... don't remember Jan 07 15:20:45 gtg.. Jan 07 15:21:34 cu Jan 07 15:21:42 will rebuild from scratch then :P Jan 07 15:22:11 mrmoku, i have a mokophone bump here, could you send it to main feeds? Jan 07 15:22:19 i mean without waiting for your gdbus stuff :) Jan 07 15:22:23 no Jan 07 15:22:25 not possible Jan 07 15:22:33 we can either sync all or nothing Jan 07 15:22:38 mrmoku, because of rsyncing Jan 07 15:22:39 i see Jan 07 15:22:56 i'll send you the bump patch, then i'll build on my own for now Jan 07 15:23:09 ok Jan 07 15:23:16 I hope gdbus will be sorted out soon Jan 07 15:23:19 so that we can sync again Jan 07 15:23:21 me too :) Jan 07 15:23:24 can't wait Jan 07 15:25:40 gena2x: yeah I think I managed to configure watchdog to start on resume :) Jan 07 15:26:30 mrmoku, patch sent Jan 07 15:32:19 mrmoku, what do you think about a generated dbus library with common dbus interfaces? Jan 07 15:32:40 playya_: what do you man by common ? Jan 07 15:32:44 +e Jan 07 15:33:11 what we need. e.g. bluez, notofications, ... Jan 07 15:33:22 ahh Jan 07 15:33:28 s/notofications/notofications/ Jan 07 15:33:36 damn it ;) Jan 07 15:33:38 :P Jan 07 15:34:14 hmm Jan 07 15:34:44 so far we (as in SHR phonestack) use only fso and ourself via dbus Jan 07 15:35:02 ok Jan 07 15:35:03 but nofifications are something we probably want too Jan 07 15:35:07 mrmoku, why all these weirdnesses to me? :S Jan 07 15:35:09 perl: error while loading shared libraries: libperl.so: cannot open shared object file: No such file or directory Jan 07 15:35:10 lol Jan 07 15:35:31 ahh, that one I remember ;) Jan 07 15:35:38 mrmoku, you do? :P Jan 07 15:35:49 lindi-: hope it will help you to fight resume bug :) Jan 07 15:35:50 well... the problem... the solution a lot less Jan 07 15:36:06 ehm Jan 07 15:36:12 daniele_athome: I think adding some link was the solution Jan 07 15:36:13 gena2x: it will, now I just need to add some printfs() to the resume path and look at ramconsole after crash Jan 07 15:36:21 radekp: anyway need to really all patches larsc doesn't like Jan 07 15:36:30 *to reapply Jan 07 15:36:31 as I just wiped my tmp for a rebuild from scratch... can't check anymore :/ Jan 07 15:36:47 gena2x: i am nearly done with it Jan 07 15:36:55 GNUtoo|laptop: do you remember the 'fix perl link' ? Jan 07 15:36:58 mrmoku, rebuilding from scratch would a solution too? Jan 07 15:37:02 I think it was you pointing that one out... Jan 07 15:37:03 radekp: here it is not working :) Jan 07 15:37:05 *would be Jan 07 15:37:08 i mean .37 Jan 07 15:37:09 I think so, yes Jan 07 15:37:20 can tell you in about 5 hours ;) Jan 07 15:37:42 well i could use a rm -fR tmp :) i've not been building from scratch for much time Jan 07 15:38:48 radekp: it says me "UBIFS: read-only UBI device" then "UBIFS error cannot mount read-write - read-onlt media" then "cannot open root device" :) Jan 07 15:39:09 also it can't find any mmcblkps Jan 07 15:39:27 gena2x: interesting i could boot my ubifs nand root Jan 07 15:39:33 may be it is my hwecc stopper working Jan 07 15:39:36 mrmoku, yes? Jan 07 15:39:45 *stopped Jan 07 15:39:47 gena2x: i'll finish porting the ugly patches and will test some... Jan 07 15:39:59 GNUtoo|laptop: the problem with perl... you told me to do a link between 2.6.3 and 2.6.2 or somethink like that Jan 07 15:40:04 or was that a different problem? Jan 07 15:40:09 hmm Jan 07 15:40:13 lets call em 'ugly from larsc point of view' Jan 07 15:40:17 thinking about that it might have been a different one Jan 07 15:40:24 ~lart mrmoku's memory Jan 07 15:40:24 * apt stabs mrmoku's memory Jan 07 15:40:34 mrmoku, hmmm, I don't remember Jan 07 15:40:38 hmm i saw something about hw ecc... but can remember what it was Jan 07 15:40:43 mrmoku, altough I remember what I told you Jan 07 15:40:49 but I don't remember if: Jan 07 15:40:56 *the correct fix was to link Jan 07 15:41:01 *the correct fix was to delete Jan 07 15:41:04 let me look Jan 07 15:41:25 GNUtoo|laptop, could you post to ML if you find out please? Jan 07 15:41:25 :) Jan 07 15:42:38 GNUtoo|laptop: btw. I will try your gdbserver way now... as phonefsod segfaults on the FR but not on my laptop Jan 07 15:42:51 and it segfaults directly when entering the mainloop Jan 07 15:43:09 everything happening there has a logger as first thing in theory Jan 07 15:43:16 so there should be something in the log... but is not Jan 07 15:47:08 ok Jan 07 15:47:11 I can try it for you Jan 07 15:47:19 that would be great Jan 07 15:47:30 as I have to rebuild phonefsod from scratch first Jan 07 15:47:30 else howto can be found on elinux.org (improved a lot by me) Jan 07 15:47:41 yeah, have that open in my browser :) Jan 07 15:48:00 btw. the cross gdb is not in tmp/cross Jan 07 15:48:03 at least for me Jan 07 15:48:18 yes things have changed Jan 07 15:48:23 update the howto then Jan 07 15:48:36 need an account? Jan 07 15:49:05 not sure Jan 07 15:49:20 if so I will have to create one :/ Jan 07 15:49:35 will update that then when my build is populated Jan 07 15:52:50 radekp: btw patches silently not accepted by lars are not more ugly than many pieces of kernel code. At least all they are working, while we have udc, ar6000 and other stuff which is actually much less working. Jan 07 15:53:47 gena2x: at least there are only a few of them, i will post mail about them when i am done with porting & testing Jan 07 15:54:34 radekp: why not... Jan 07 15:55:34 * pabs3 interested to not ath6kl got into 2.6.37 staging Jan 07 15:56:47 as far as i read, it doesnt support ar6k Jan 07 15:57:12 only 6k1 Jan 07 15:57:13 mrmoku, Program exited normally. Jan 07 15:57:15 as I said Jan 07 15:57:26 I'll try to ddb the childs Jan 07 15:57:37 GNUtoo|laptop: hmm... it does segfault for me Jan 07 15:57:48 GNUtoo|laptop: try with -d 1 Jan 07 15:57:49 it segfaults for me too but only when using -d Jan 07 15:57:53 ahh Jan 07 15:58:10 without -d 1 it segfaults too Jan 07 15:58:16 but not the main program Jan 07 15:58:18 yeah Jan 07 15:58:22 only the childs Jan 07 15:58:25 so I'll try -d 1 Jan 07 15:58:27 for a start Jan 07 15:58:32 ok Jan 07 15:58:37 else I've to findout what to type to follow childs Jan 07 15:59:24 http://pastebin.com/NXRmPdyL Jan 07 16:00:03 I'll install libgio-dbg Jan 07 16:00:49 http://pastebin.com/vdnxr7HB Jan 07 16:02:45 GNUtoo|laptop: hmm... why isn't fso_dimit in the bt? Jan 07 16:02:57 btw. nice bt :) Jan 07 16:03:01 * mrmoku wants that too Jan 07 16:03:21 do you need more infos? Jan 07 16:03:45 * mrmoku takes a look at deactivate_screensaver Jan 07 16:03:52 ok Jan 07 16:04:17 or just comment some stuff in phonefsod.conf Jan 07 16:04:19 I'll tyr that Jan 07 16:10:17 GNUtoo|laptop: do you have a debug phonefsod.log? Jan 07 16:10:41 yes Jan 07 16:10:48 hmm it might be interesting to try Atheros AR6003 support (non mac80211) (ATH6K_LEGACY) Jan 07 16:11:53 http://pastebin.com/Ygi2V9n0 Jan 07 16:13:01 ahh, that is on htcdream even Jan 07 16:13:06 does not even get that far for me Jan 07 16:13:18 you could try one thing Jan 07 16:13:20 in phonefsod.c Jan 07 16:13:34 g_timeout_add(0, fso_startup, NULL); Jan 07 16:13:41 change the 0 to 10 Jan 07 16:13:48 ok Jan 07 16:13:54 htcdream rebuilt from scratch Jan 07 16:14:02 I've also n900 not rebuilt from scratch Jan 07 16:14:27 btw image creation fails because of missing cough....libnl1 Jan 07 16:14:34 heh Jan 07 16:14:37 the netlink fun Jan 07 16:14:39 yes Jan 07 16:17:08 http://pastebin.com/Cyh9Zws1 Jan 07 16:18:08 hmmm... g_timeout_add was seconds or milliseconds Jan 07 16:19:06 miliseconds I think Jan 07 16:19:21 oh, yes Jan 07 16:19:29 so better make the 10 a 10000 :P Jan 07 16:19:29 sorry Jan 07 16:19:30 * pespin 's having problems building from scratch with python-numeric Jan 07 16:19:51 else it would have been g_timeout_add_seconds Jan 07 16:20:02 yeah, I always forget that :/ Jan 07 16:20:35 I change to g_timeout_add_seconds Jan 07 16:20:37 it's prettier Jan 07 16:20:43 ok Jan 07 16:20:54 is it possible to boot from SD card on the n900? Jan 07 16:20:56 it works Jan 07 16:21:11 playya_: sure Jan 07 16:21:18 kewl Jan 07 16:21:45 my father needs to refresh his contract and gets a cheap n900 :) Jan 07 16:22:06 http://pastebin.com/pqXKuc5Z Jan 07 16:22:25 playya_: nice :) Jan 07 16:22:42 GNUtoo|laptop: great... that looks a lot nicer :D Jan 07 16:23:01 so it's a race with the phoneui proxy not yet being there Jan 07 16:23:04 http://pastebin.com/7Ds8WQ3p Jan 07 16:23:10 and still not crashed Jan 07 16:23:32 you should grep for g_timeout_add Jan 07 16:23:39 to see if there aren't any other bugs Jan 07 16:23:54 and change the ones that are there to seconfs Jan 07 16:23:55 well... it's not the actual problem Jan 07 16:23:58 ah? Jan 07 16:24:21 the problem is that when starting up fso it wants to disable screen blanking Jan 07 16:24:25 for the startup phase Jan 07 16:24:32 it needs a phoneui proxy for that Jan 07 16:25:16 have to think a minute about the correct fix Jan 07 16:25:26 GNUtoo|laptop: thx very much :) Jan 07 16:25:41 thanks to you Jan 07 16:25:46 you do most of the n900 work Jan 07 16:26:00 I'm rather a traitor(replicant) lol Jan 07 16:26:05 :P Jan 07 16:26:28 altough I finish and relase that thing(replicant) and then I'll try to be back Jan 07 16:26:44 if I'm not too busy with other stuff Jan 07 16:26:58 (like university, work or bug device) Jan 07 16:27:13 yeah, too many things to do for everybody of us :/ Jan 07 16:27:46 it's rather that I can't promise anything, as usual Jan 07 16:27:55 else I would have to do what I promised Jan 07 16:28:21 which could be hard depending on the future situation Jan 07 16:28:25 as I can't tell future Jan 07 16:28:29 I don't promiss Jan 07 16:28:34 we should do some recruitment plan for SHR :P http://www.hetemeel.com/unclesam/1.jpg Jan 07 16:28:54 pespin, for me you need to change the image a bit Jan 07 16:29:08 GNUtoo|laptop, yeah jeje Jan 07 16:29:18 http://www.fsf.org/associate/gsam_iwantgnu.jpg Jan 07 16:29:29 change that for we need GNU Jan 07 16:29:32 :) Jan 07 16:29:44 btw, I might be doing some presentation on SHR in two months or so Jan 07 16:29:51 nice Jan 07 16:30:06 mickeyl, do we have any plan to present something at fosdem? Jan 07 16:30:30 in a special meeting we are prepairing in the catalan LoCO team oriented to free mobile OS Jan 07 16:30:57 *ubuntu catalan LoCo Jan 07 16:31:11 wow Jan 07 16:31:42 nah don't expect a big meeting hehe Jan 07 16:32:03 if it can bring more people it would be great Jan 07 16:32:08 but hope I will be interesting and good for SHR. We may get some help and spread the word Jan 07 16:32:12 I wasn't thinking about big meeting Jan 07 16:32:27 just some focussed small meeting that get more people involved Jan 07 16:32:47 GNUtoo|laptop, people are not so good speaking about thecnical knowledge, but can be great as users and spread SHR Jan 07 16:33:48 I hope I'll get people attention. I'll be competing about lots of Android guys :P Jan 07 16:34:03 ok lol Jan 07 16:34:20 if you need android training I can train you to speak against android Jan 07 16:34:29 do you know if there are some presentation material already done? Jan 07 16:34:44 about SHR? yes at the SHR meeging or fosdem Jan 07 16:34:46 GNUtoo|laptop, yeah, would be great, I'll ask you for help when I start thinking about it :) Jan 07 16:35:01 GNUtoo|laptop, do you know where can I find it? Jan 07 16:36:52 pespin, shr website I think Jan 07 16:37:01 ok thanks, I'll search there :) Jan 07 16:37:32 btw, I hope I'll be in fosdem this year ;) Jan 07 16:37:42 http://shr.bearstech.com/ Jan 07 16:37:51 has some stuff Jan 07 16:37:53 for instance: Jan 07 16:37:54 http://shr.bearstech.com/fsoshrudcon/ Jan 07 16:39:11 thanks :) Jan 07 16:40:32 SHR: 03mok 07phonefsod * rd760ef427882 10/src/ (4 files): delay the fso startup sequence if phoneuid is not yet on the bus Jan 07 17:03:11 it turns out i can only be there on sunday Jan 07 17:04:04 nikolaus schaller and me have applied for a common talk in the embedded room, if that gets accepted and is on a sunday, we can spend some slides on the state of FSO / SHR on hardware Jan 07 17:07:30 ok nice Jan 07 17:09:16 so, two tasks for the weekend: Jan 07 17:09:22 1.) gpsd plugin for tdld Jan 07 17:09:33 2.) correct timeouts for fsogsmd Jan 07 17:09:53 sounds like a nice plan :-) Jan 07 17:10:03 and maybe finally fix the netlink mess Jan 07 17:10:07 oh yesd Jan 07 17:10:12 perhaps we can do that together Jan 07 17:10:22 yup Jan 07 17:10:56 yesd is the new name of the ssh daemon( the daemon that say yes at all what you type) ? Jan 07 17:11:02 hahaha Jan 07 17:11:05 righto Jan 07 17:11:37 i'm afraid i rather need a nod though Jan 07 17:11:57 so now that phone works again what are mrmoku 's plan? Jan 07 17:12:31 * GNUtoo|laptop wants to discuss going back to 2.6.28 Jan 07 17:12:46 on nokia900 Jan 07 17:13:42 GNUtoo|laptop: does not work yet :/ Jan 07 17:13:51 hmm where are the vala bindings for E maintained now? Jan 07 17:13:56 GNUtoo|laptop: could you build current phonefsod and check if it works on the dream? Jan 07 17:14:08 because i remember there's a tree in fso.org and in E svn repo Jan 07 17:14:15 but don't remember which in the mainstream now Jan 07 17:14:30 mrmoku, ah it works fine on the dream, I did a test call, I'll test more Jan 07 17:14:32 pespin: i maintain them upstream, there's no longer a repo on fso Jan 07 17:14:39 mickeyl, ok thanks :) Jan 07 17:14:57 GNUtoo|laptop: did you build what I pushed? Jan 07 17:15:05 pespin: unfortunately E progresses very quick Jan 07 17:15:09 i need to find time to catch up Jan 07 17:15:15 but i won't do that until i need it for my own UI work Jan 07 17:15:18 mrmoku, yes but I did manual stuff so I need to restest Jan 07 17:15:18 so patches appreciated :) Jan 07 17:15:53 GNUtoo|laptop: if it works for you on dream then we have a different problem on gta02 :/ Jan 07 17:15:59 ah ok Jan 07 17:16:14 * GNUtoo|laptop dreams of the time when it works on nokia900 Jan 07 17:16:56 find us a couple of C devels Jan 07 17:17:00 then we can get it quickly Jan 07 17:17:52 (or rewrite the damn stuff in Vala ;) Jan 07 17:17:59 mrmoku, it say phoneui is not on the bus Jan 07 17:18:03 if I find some problem I try looking into it :) Jan 07 17:18:15 phoneuid is launched Jan 07 17:18:15 GNUtoo|laptop: correct Jan 07 17:18:17 pespin: ok, cool. i think it's mainly missing API and some refactorerd things Jan 07 17:18:28 shr-settings Jan 07 17:18:32 tons of new widgets since i last looked Jan 07 17:18:49 say to launch phonefsod Jan 07 17:18:51 mickeyl, I tried edje/vala some time ago but I couldn't get it working. Jan 07 17:19:09 but never used edje out of vala, so i don't know where's the problem :P Jan 07 17:19:11 GNUtoo|laptop: hmm... so it segfaulted? Jan 07 17:19:11 pespin: hmm.. take a look at zhone Jan 07 17:19:20 don't know, Jan 07 17:19:23 zhone2 in the aurora repository talks a bit to edje Jan 07 17:19:29 mickeyl, oh ok great :) Jan 07 17:19:29 I launch phonefsod manually and it works it seems Jan 07 17:20:13 I'm now suscribed to wind Jan 07 17:20:18 my local operator Jan 07 17:20:25 GNUtoo|laptop, btw, I'm using about 5 Gb of music in my FR, and only thing works really slow in enjoy is Album related things. Jan 07 17:20:26 I had to launch phonefsod manually Jan 07 17:20:37 pespin, nice Jan 07 17:20:39 it takes forever to show all albums, or albums of one artist Jan 07 17:20:55 but it shows All songs quite fast Jan 07 17:20:55 pespin, that album stuff should be fixed Jan 07 17:21:07 taking albums from the net each time is not a good option Jan 07 17:21:29 GNUtoo|laptop, yeah, I thought it was some problem with internet fetching too :) Jan 07 17:21:44 I plan adding fso support on enjoy too. Jan 07 17:21:50 mickeyl, btw before gdbus migration you could use multiple plugins for 1 resource Jan 07 17:21:52 but before I'd like to have emtooth2 working :) Jan 07 17:21:55 now you can't anymore Jan 07 17:22:14 GNUtoo|laptop: that was by accidant Jan 07 17:22:21 ah ok Jan 07 17:22:22 nice Jan 07 17:22:28 it's fixed now? Jan 07 17:22:46 i mean it's not allowed to have multiple plugins per resource Jan 07 17:22:52 it should've never worked Jan 07 17:22:58 ouch Jan 07 17:23:00 now it doesn't - which is correct Jan 07 17:23:01 cough Jan 07 17:23:12 so I should write a plugin per phone? Jan 07 17:23:23 lol? src/main.c:5:18: fatal error: glib.h: El fitxer o directori no existeix <- file not found Jan 07 17:23:23 copying code? Jan 07 17:23:35 why copying? either things are standard interfaces or they're device specific Jan 07 17:23:44 if they're standard interfaces, one plugin for all Jan 07 17:23:48 what about a combinaison of 3 things: Jan 07 17:23:50 ifconfig Jan 07 17:23:52 if they're device specific, you can't get away with multiple plugins anyways Jan 07 17:23:54 rfkill Jan 07 17:23:56 and modprobe Jan 07 17:24:06 for those things we need a miniscripting Jan 07 17:24:12 or factor it out into a powersupply_ Jan 07 17:24:14 yes but how do I achieve that? Jan 07 17:24:14 err Jan 07 17:24:18 powercontrol_ Jan 07 17:24:25 I wait for you to come up with something? Jan 07 17:24:30 I can wait Jan 07 17:24:45 give me two or even better three different examples Jan 07 17:24:48 then i think about a solution Jan 07 17:24:56 (please write it up somewhere) Jan 07 17:25:01 example means? Jan 07 17:25:05 machines? Jan 07 17:25:18 yes, concrete usage patterns we need to solve Jan 07 17:25:24 ok I'll gather that Jan 07 17:25:26 like (on machine bla you need to foo then bar then blub) Jan 07 17:25:33 I'll ask morphis too Jan 07 17:25:34 thanks Jan 07 17:25:40 that'll help to find the proper balance Jan 07 17:25:44 and JaMa had the same issue then Jan 07 17:25:51 s/then// Jan 07 17:25:51 GNUtoo|laptop meant: and JaMa had the same issue Jan 07 17:27:12 GNUtoo|laptop: looks like gdb is badly broken on armv4t :( Jan 07 17:27:33 mrmoku, ah? pastebin Jan 07 17:27:34 (gdb) bt Jan 07 17:27:34 #0 0x4016923c in ?? () Jan 07 17:27:34 #1 0x40169134 in ?? () Jan 07 17:27:34 Backtrace stopped: previous frame identical to this frame (corrupt stack?) Jan 07 17:27:36 ok Jan 07 17:27:42 is all I get Jan 07 17:27:44 you need -dbg Jan 07 17:27:49 -dbg is installed Jan 07 17:27:57 ok pastebin the full gdb trace Jan 07 17:28:06 starting from ./gdb Jan 07 17:28:25 s#./gdb#gdm program# Jan 07 17:28:46 gena2x: https://github.com/radekp/linux-2.6/tree/qtmoko-v32-base Jan 07 17:29:15 hmm Jan 07 17:29:18 GNUtoo|laptop: http://shr.pastebin.com/YNTxHEjb Jan 07 17:29:19 ~lart gpsd Jan 07 17:29:19 * apt squishes gpsd like a bug Jan 07 17:29:21 head doesn't build Jan 07 17:29:37 gena2x: this is first attempt with my "ugly" patches (that arent sometimes that ugly at all ;-) Jan 07 17:30:04 mrmoku, you didn't follow the howto Jan 07 17:30:14 you should at least set sysroot Jan 07 17:30:17 gena2x: btw does not boot for me if usb cable is plugged in Jan 07 17:30:33 mrmoku, basically do that: Jan 07 17:30:35 GNUtoo|laptop: yeah, quickly fired it without reading as the build finished :P Jan 07 17:30:55 sudo sshfs the openmoko filesystem Jan 07 17:31:03 then sysroot to that sshfs dir Jan 07 17:31:11 set sysroot /path/to/sshfs/ Jan 07 17:31:18 and uses as usual Jan 07 17:31:37 ok Jan 07 17:32:05 btw my keyboard is not illuminated anymore on htcdream Jan 07 17:32:08 strange.... Jan 07 17:32:15 maybe it's related to all this stuff Jan 07 17:36:23 mickeyl, uhm mdbus2 fails build here Jan 07 17:36:32 that's bad Jan 07 17:36:33 how? Jan 07 17:36:35 I have vala-11.2 Jan 07 17:37:08 you need 11.3 or our version Jan 07 17:37:21 dbus.vapi was broken for some time Jan 07 17:37:22 ouch Jan 07 17:37:34 ok I'll compile vala-git Jan 07 17:37:39 yes, or that Jan 07 17:38:13 if there's sufficient interest, i can do a release for mdbus2 Jan 07 17:38:22 it's pretty complete Jan 07 17:38:35 radekp: i am not surpised (about usb cable) Jan 07 17:39:03 radekp: anyway, if base doesn't boot, with pathes it will not boot too :( Jan 07 17:39:17 freesmartphone.org: 03mickey 07cornucopia * rd29509c0f0a1 10/fsotdld/ (4 files in 3 dirs): fsotdld: add build skeleton for new gpsd plugin Jan 07 17:39:25 mickeyl, as you wish. it's no problem at all for me to fetch it from git :) Jan 07 17:39:30 GNUtoo|laptop: have to prepare the table and have dinner first :/ Jan 07 17:39:40 bbl Jan 07 17:39:47 pespin: ok. some debian folks have problems with git versions, so i sometimes cut releases Jan 07 17:39:49 gena2x: 2.6.37 booted for me but a lot of stuff is missing Jan 07 17:39:54 oh 11.3 is out Jan 07 17:39:58 vala Jan 07 17:39:59 radekp: please, just don't call em 'ugly', i like definition something like 'non-kosher' :) Jan 07 17:40:14 I thought you meant git with 11.3 :P Jan 07 17:40:18 ok Jan 07 17:40:48 no, it's out, although there are some good fixes already in post0.11.3 git Jan 07 17:40:49 :) Jan 07 17:40:53 hmm biggest problem is that it does not work with tslib Jan 07 17:41:11 lindi-: this is normal for lars branch, he doesn't like lot of stuff Jan 07 17:41:48 i get "selected device is not a touchscreen I understand" Jan 07 17:41:48 gena2x: heh :) Jan 07 17:42:34 radekp: do we want to start next round of same patches? Jan 07 17:42:46 already had it with 2.6.34, but it got fixed with this commit: https://github.com/radekp/linux-2.6/commit/98ce762ebaac40cdd896095a6e3fca53678d0317 Jan 07 17:43:03 radekp: until lars reivent them all like wsod fis :) Jan 07 17:43:09 *fix Jan 07 17:43:11 radekp: ah Jan 07 17:43:16 radekp: I know how to fix it! :) Jan 07 17:43:20 anarsoul: hi, cool :) Jan 07 17:43:32 radekp: pressure fix doen't work? Jan 07 17:43:35 actually, it's tslib fault Jan 07 17:43:42 gena2x: pressure fix is necessary aswell Jan 07 17:44:01 radekp: http://downloads.tuxfamily.org/linuxrx1950/patches/rx1950.7jan/0005-Revert-input-version-as-tslib-compiled-against-older.patch Jan 07 17:44:02 anarsoul: more fixes :) Jan 07 17:44:12 gena2x: i really want 2.6.37, with a bit luck we can catch upstream for the first time Jan 07 17:44:15 gena2x: it's tslib fault, it was compiled with older kernel Jan 07 17:44:24 real fix is recompile tslib with 2.6.37 headers Jan 07 17:45:09 radekp: why do you think we'll catch more that with .34? Jan 07 17:45:15 *than Jan 07 17:45:20 anarsoul: ahh oki, but you patch seems easier :) Jan 07 17:45:23 * anarsoul is too lazy to recompile userland Jan 07 17:45:37 * radekp tries Jan 07 17:45:38 anarsoul: nice, thanks for info Jan 07 17:45:49 you're welcome :) Jan 07 17:46:01 radekp: what do you mean by 'close to upstream'? Jan 07 17:46:03 gena2x: 2.6.37 is released 1 day ago, 2.6.34 is not even supported now Jan 07 17:46:16 ah, this... then ok :) Jan 07 17:46:34 2.6.32 is supported, afaik it's LTS kernel Jan 07 17:47:33 but still the the final solution for me is to get all gta02 stuff in vanilla Jan 07 17:49:15 would be nice. but as far as we want to do some development too, we'll always lag behind. Jan 07 17:51:59 it's not big problem having more versions supported, i am not going to delete my 2.6.34 branch of course :) Jan 07 17:52:27 i think main problem is bugs. current state of upstreaming look good alredy Jan 07 17:53:10 afaik glamo is not upstream Jan 07 17:53:43 but of course, if someone wishes to improve upstreaming skills - thats good :) Jan 07 17:54:49 gena2x: upstreaming would motivate people Jan 07 17:55:03 gena2x: and would make it possible to have the kernel in debian Jan 07 18:00:51 gena2x: btw where do you have latest uboot? Jan 07 18:00:59 lindi-: really, no disgreement or subject to talk here, upstreaming is good of course :) Jan 07 18:01:16 gena2x: i am answering questin on ML and want him to point to your page :) Jan 07 18:01:28 anyone tried uboot upstreaming? Jan 07 18:01:30 radekp: i am keeping link on http://wiki.openmoko.org/wiki/U-boot-gena2x Jan 07 18:01:44 gena2x: nice thanks Jan 07 18:01:55 radekp: also updating link on debian-on-freerunner and in qtmoko wiki Jan 07 18:03:01 pabs3: i tried to read u-boot ml about laforge iteraction with upstream Jan 07 18:03:18 pabs3: upstream didn't like openmoko patches Jan 07 18:03:34 pabs3: then years later they reinvented some of them Jan 07 18:05:12 pabs3: but u-boot ml is huge really Jan 07 18:05:29 anarsoul: it works nice, thanks a lot! Jan 07 18:06:38 * pabs3 finds http://elinux.org/Merge_DFU_support_into_mainline_U-Boot Jan 07 18:07:32 pabs3: "Their U-Boot appears to be in good shape" :) Jan 07 18:07:32 (about our u-boot) Jan 07 18:10:59 pabs3: ok, i am not going to search u-boot ml again. but remarkable mail were that (from some of u-boot core devels) 'openmoko doesn't commit their pathes back' while laforge were greatly surprised by that and really i read may be 20 mails with all kind of rejected patches. just real u-boot ml since beginning. Jan 07 18:11:25 *read Jan 07 18:12:42 ya, the non-cooperation with u-boot was quite frustrating Jan 07 18:12:48 we tried for a long time Jan 07 18:12:51 but then just gave up Jan 07 18:13:43 later on product management really complained about the engineering budget that went into something "as trivial as a bootloader" Jan 07 18:14:04 yeahh, i guessed that :) Jan 07 18:15:13 pabs3: but anyway, i just downloaded whole ml, merged it and grep for 'laforge' :) Jan 07 18:16:27 lot of things seems to work in 2.6.37 now Jan 07 18:17:06 usb networking, wifi, bluetooth, touchscreen, gsm calls, bluetooths are ok on the first sight Jan 07 18:17:52 heh, if only all this were able to boot here... Jan 07 18:17:54 hmm 5mA in suspend? i guess that must be some error Jan 07 18:18:20 where did you got 5mA? Jan 07 18:18:21 gena2x: have you tried my branch and my config? Jan 07 18:18:30 radekp: do you already have some qtmoko-2.6.37 branch to test? Jan 07 18:18:37 gena2x: in NeoConrol - it reads current_now Jan 07 18:18:42 lindi-: yop Jan 07 18:19:05 ok, congrats, first bug :) Jan 07 18:19:12 'git fetch radekp-linux-2.6'? Jan 07 18:19:13 gena2x: but i guess the battery driver gives wrong numbers - or we would have 2x longer battery life :) Jan 07 18:19:17 '* [new branch] qtmoko-v32 -> radekp-linux-2.6/qtmoko-v32' Jan 07 18:19:36 radekp: is that 5mA with gsm? Jan 07 18:20:10 lindi-: https://github.com/radekp/linux-2.6/tree/qtmoko-v32-base Jan 07 18:20:22 lindi-: cant push now the latest touchscreen fix Jan 07 18:20:30 lindi-: yes with gsm Jan 07 18:20:40 lindi-: at least it should be Jan 07 18:20:41 radekp: gsm is bit bursty Jan 07 18:20:47 radekp: better measure without gsm Jan 07 18:22:49 radekp: your tree has no changes related to ubi or mtd Jan 07 18:23:01 radekp: so no point to try it until it will boot Jan 07 18:23:30 radekp: also i do not really like you 'doesn't boot with plugged usb' :) Jan 07 18:23:34 gena2x: it should be in .config Jan 07 18:23:37 *your Jan 07 18:23:59 mickeyl, I'm having problems with vala+ItemList. In fact, I had this trouble some time ago but no I was relooking at the problem, but I don't kn ow how to overpass it. Jan 07 18:24:00 just clone it and run make Jan 07 18:24:03 it should boot Jan 07 18:24:12 even ubifs Jan 07 18:24:25 the thing is that when you append an ItemList, you have a *data to set there whatever you want (in C) Jan 07 18:24:38 ok, it's relatively fast try :) Jan 07 18:24:50 but afaik, vala uses that data pointer for internal things, so it is not available Jan 07 18:25:29 then, when it is clicked, in the callback, I can get the label of the item, but I have no way to get the Object which the button is referring too Jan 07 18:25:46 s/too/to Jan 07 18:26:14 but first lets answer u-boot question in ML Jan 07 18:26:14 mickeyl, any idea on what to do? Jan 07 18:27:04 I could do a HashTable with but it's quite dirty :S Jan 07 18:27:08 gena2x: i have to try booting without my patches, i think it can be one them i havent noticed it with openmoko's version Jan 07 18:27:53 no-no-no, really, may be config - i just used my .34 one Jan 07 18:28:17 with quick-running menuconfig, may be something went wrong Jan 07 18:29:04 radekp: qtmoko-v32 has the same issue 'FIC Neo1973 GSM Phone (GTA01 Hardware) (MACH_NEO1973_GTA01) [N/y/?] (NEW)' Jan 07 18:30:29 lindi-: which config do you use? Jan 07 18:30:37 radekp: gta02_defconfig Jan 07 18:30:56 radekp: I should use this .config? Jan 07 18:31:02 lindi-: yes please Jan 07 18:31:15 having different config files in different branches makes automatic building difficult :) Jan 07 18:31:50 i can overwrite also the gta02_defconfig, but i cant promise i will not forget it always Jan 07 18:31:59 I just need to add a symlink Jan 07 18:41:35 hi, I've a friend that has a question, he wants to have a shell over GSM, what's the best way to do that? can the freerunner or other phones do DATA calls? and how to piggy-back a shell on that? Jan 07 18:41:57 GNUtoo|laptop: usually data calls are more expensive than gprs. why does he not want to use gprs? Jan 07 18:42:09 ah I didn't know Jan 07 18:42:13 GNUtoo|laptop: ask that first Jan 07 18:42:23 he wanted a cheaper way Jan 07 18:42:29 like he calls Jan 07 18:42:35 and the remote phone answer Jan 07 18:42:39 and give him a shell Jan 07 18:42:40 GNUtoo|laptop: but usually the calls are more expensive than gprs Jan 07 18:42:42 ok Jan 07 18:42:46 I'll tell him Jan 07 18:42:48 thanks a lot Jan 07 18:42:55 GNUtoo|laptop: depends on where he lives though Jan 07 18:43:03 italy Jan 07 18:43:16 and he has already 3g connection Jan 07 18:43:19 I'll ask him Jan 07 18:43:48 basically the idea is to fix his vpn or system trough an alternate connection Jan 07 18:44:38 GNUtoo|laptop: I have a callback system for that. if my FR receives a call from a configured number it doesn't answer. instead it starts a gprs connection ands lets me login via ssh Jan 07 18:44:46 GNUtoo|laptop: handy if I ever forget my phone somewhere Jan 07 18:45:03 ok wow Jan 07 18:45:34 he said that there are offers where you can call a number of your choice very cheap(2E) Jan 07 18:45:51 but you can call the number as much as you want Jan 07 18:45:52 I used to do periodic calls every two hours to squeeze out bugs Jan 07 18:46:02 ok Jan 07 18:46:05 but it turned out that my phone company was more unreliable than my FR Jan 07 18:46:10 lol Jan 07 18:46:15 radekp: ah, that guy is using NOR u-boot :) Jan 07 18:46:19 their voip service was just opened that year Jan 07 18:46:47 GNUtoo|laptop: ok if you have such numbers then it might be worth the effort Jan 07 18:46:52 ok Jan 07 18:47:00 do you have an idea on where to start Jan 07 18:47:00 ? Jan 07 18:47:06 GNUtoo|laptop: I don't have those so I can't really test cheaply Jan 07 18:47:22 I'd pay for every minute :) Jan 07 18:47:24 thanks a lot Jan 07 18:47:28 me too Jan 07 18:47:42 thoses only work with a fixed number Jan 07 18:47:46 gena2x: yup, in fact it's nice that NOR now does not work with qtmoko :) Jan 07 18:47:47 you choose a special number Jan 07 18:47:52 GNUtoo|laptop: just call a modem Jan 07 18:47:52 such as your house number Jan 07 18:47:56 radekp: :) Jan 07 18:47:56 ok Jan 07 18:47:59 but how to do taht Jan 07 18:48:05 I call it as DATA Jan 07 18:48:07 and then? Jan 07 18:48:16 GNUtoo|laptop: then calypso feeds you bytes Jan 07 18:48:24 can I piggy-back a serial connection on it? Jan 07 18:48:32 on top of the bytes Jan 07 18:48:33 GNUtoo|laptop: calypso is connected via a serial port Jan 07 18:48:36 eys Jan 07 18:48:39 *yes Jan 07 18:48:39 no need to piggypack Jan 07 18:48:43 /dev/ttysomething Jan 07 18:48:49 but that's for AT Jan 07 18:48:51 gena2x: i was using NOR uboot for long time too, and havent found any issues with it, but then someone told it's bugged so i switched to qi completely Jan 07 18:48:53 ahh got it Jan 07 18:48:56 GNUtoo|laptop: you can multiplex those Jan 07 18:49:05 GNUtoo|laptop: have you ever used a normal modem by hand? ;) Jan 07 18:49:07 multiplexing by the multiplexing daemon...in userspace Jan 07 18:49:08 yes Jan 07 18:49:17 I just didn't remember how it worked Jan 07 18:49:26 gena2x: but it's nice that you keep uboot up, at least we have 2 choices which is nice Jan 07 18:49:26 I just didn't connect the pieceds Jan 07 18:49:28 *pieces Jan 07 18:49:33 radekp: i just compose long essay why nand u-boot may fail in that way and how it is impossible to get error he have, then noticed his 'AUX+power' :) Jan 07 18:49:33 got it now Jan 07 18:49:50 gena2x: :) Jan 07 18:51:51 btw I usually control modem not for data Jan 07 18:51:55 but for calls etc... Jan 07 18:52:04 I never digged into gprs/3g Jan 07 18:57:35 GNUtoo|laptop: still no go Jan 07 18:57:50 ah? Jan 07 18:57:53 (gdb) set substitute-path / /home/mok/shrbuild/OE-chroot/OE/tmp/sysroots/armv4t-oe-linux-gnueabi/ Jan 07 18:57:59 hmmm Jan 07 18:58:03 that's for sources Jan 07 18:58:13 set sysroot Jan 07 18:58:28 else I do it for you Jan 07 18:58:33 but I need to eat at a point Jan 07 18:58:59 I also need to build for om-gta02 Jan 07 18:59:03 mickeyl, I got it working but in a very hackish way Jan 07 19:00:17 (gdb) bt Jan 07 19:00:17 #0 0x4016923c in __real_g_dbus_proxy_call () from /mnt/neo/usr/lib/libgio-2.0.so.0 Jan 07 19:00:20 #1 0x40031274 in __real_phoneui_idle_screen_call_deactivate_screensaver () from /mnt/neo/usr/lib/libshr-glib.so.0 Jan 07 19:00:23 #2 0x0000cc5e in ?? () Jan 07 19:00:26 #3 0x0000cc5e in ?? () Jan 07 19:00:28 hmm Jan 07 19:01:24 * mrmoku rereads the howto Jan 07 19:01:30 GNUtoo|laptop: buon apettito :) Jan 07 19:03:41 GNUtoo|laptop: ahh... now :-) Jan 07 19:03:44 GNUtoo|laptop: thanks very much Jan 07 19:03:56 * GNUtoo|laptop fired a build Jan 07 19:10:29 pespin: sorry, was away. how did you do it? Jan 07 19:11:15 pespin: libeflvala was planned to add some high level stuff, until that is ready we need to do some quirks Jan 07 19:11:49 SHR: 03mok 07phonefsod * r45a198ba2f25 10/src/phonefsod-fso.c: phonefsod-fso: don't activate/deactivate screensaver if phoneuid is not there Jan 07 19:12:38 item = this.li.append(label, null, null, opener.go); Jan 07 19:12:39 opener.item = item; // this is needed to unselect ListItem to be able to press it again. Jan 07 19:12:42 mickeyl, ^ Jan 07 19:13:04 hmm, right Jan 07 19:13:10 opener.go is an objectwhich I was already using before and that holds some info about each item Jan 07 19:13:36 mickeyl, thing is that the callback from the append() doesn't work the same way as in the "clicked" list smartcallback Jan 07 19:13:58 it doesn't get triggered each time you press an item. Jan 07 19:14:19 you have to press A, and if you want to press A another time, you have to press B before Jan 07 19:14:24 this is the problem I had before Jan 07 19:14:41 so what I did, is get the item reference in the opener object Jan 07 19:14:52 and inside the go() method I add: Jan 07 19:15:11 this.item.selected_set(false); Jan 07 19:15:32 to unselect it and then next time you press the ListItem, the go method gets called another time :P Jan 07 19:15:57 hmm Jan 07 19:16:07 that sounds like a problem in EFL though Jan 07 19:16:21 mickeyl, I asked about i before in #edevelop Jan 07 19:16:38 you can read response ther eif you want :) Jan 07 19:17:22 radekp: fyi i tried reflashing the nor u-boot without a debug board and it worked ok. Jan 07 19:17:22 ok, i see Jan 07 19:17:40 resuming... Is this expected behaviour? yes, for now Jan 07 19:17:44 *nod* :/ Jan 07 19:17:48 ugly, ugly Jan 07 19:17:57 is this genlist already or one of the older ones? Jan 07 19:18:07 yeah. efl + vala is sometimes more messy than using it in C directly hehe Jan 07 19:18:20 yep. we'll fix that soon Jan 07 19:18:29 just needs some nice classes that hold additional state Jan 07 19:18:34 mickeyl, I'm using List, not GenList ;) Jan 07 19:19:08 right, List. GenList is only necessary once you have _many_ objects, but it has its virtues thanks to a better separation of model and view Jan 07 19:19:21 (if one can say something like that for EFL...) Jan 07 19:20:35 I don't feel like using GenList, as you don't have ususally hundreds of bluetooth devices around, and the UI and expected behaviour is quite simpler Jan 07 19:20:57 just press -> get associated object to item -> do something with the object Jan 07 19:22:01 yep Jan 07 19:22:45 mickeyl, look at GenListItem hehe: Jan 07 19:22:45 // FIXME: Caution: Do we really have access to the data field or is it used by Vala? Jan 07 19:22:46 public void* data_get(); Jan 07 19:22:46 public void data_set( void* data ); Jan 07 19:24:39 yes, some gray areas Jan 07 19:40:46 GNUtoo|laptop: yay, works now on gta02 too :) Jan 07 19:44:17 congrats mrmoku Jan 07 19:44:24 what was wrong? Jan 07 19:44:48 everything is working now? Jan 07 19:46:20 mickeyl: async proxy race for phoneui Jan 07 19:46:24 ah, ok Jan 07 19:46:29 the proxy was requested but not there yet Jan 07 19:46:45 and then came the idle notfier signal causing phonefsod to use the phoneui proxy Jan 07 19:46:49 to disable the screensaver Jan 07 19:47:00 and boom Jan 07 19:47:03 yup Jan 07 19:47:04 :P Jan 07 19:47:22 so... last segfaulting thing is fsodeviced Jan 07 19:47:32 GNUtoo|laptop: not sure if _everything_ works now Jan 07 19:47:46 but at least it asks for pin and registers again Jan 07 19:47:52 contacts and messages seem to work Jan 07 19:48:01 did not send a message or make a call yet though Jan 07 19:48:14 * mrmoku tries GPRS Jan 07 19:48:37 ok Jan 07 19:48:55 i'll try as soon as I can Jan 07 19:51:04 hmm... fsogsmd segfaulted Jan 07 19:54:19 PaulFertser: that sounds scary :) Jan 07 19:55:41 hmm... segfaulted again Jan 07 19:55:47 mickeyl: something is fishy with fsogsmd ;) Jan 07 19:55:50 nah Jan 07 19:55:53 never Jan 07 19:56:17 can you get me a BT? Jan 07 19:56:44 radekp: unless you ruin both nand and nor it's safe. Jan 07 19:57:14 yup that's right Jan 07 19:58:24 mickeyl: will try, yeah Jan 07 19:58:43 now that I know how to use gdbserver :-) Jan 07 19:59:11 something you can teach me eventually ;) Jan 07 19:59:36 mickeyl: find me an N900, then I can apply my C devel skills to it ;-) Jan 07 20:01:35 mickeyl: thanks to GNUtoo|laptop Jan 07 20:01:49 mickeyl: http://elinux.org/GDB Jan 07 20:04:22 antrik: hmm ;) Jan 07 20:04:38 lindi-: hm. wait, does the fact that nor bootloader will be used if you resume with aux-pressed also mean that if you boot from nor and resume without aux pressed, nand bootloader will be used to init hardware? Jan 07 20:05:55 I hate bluez -> SIGNAL: Remote device removed(/org/bluez/21021/hci0/dev_00_00_00_00_00_00) Jan 07 20:05:56 lindi-: ah forgot, this is not question, just thinking aloud :) Jan 07 20:15:01 gena2x: yes Jan 07 20:15:12 gena2x: you should not touch aux on resume Jan 07 20:15:15 freesmartphone.org: 03mickey 07cornucopia * r49aebdfb49b9 10/fsotdld/ (5 files in 3 dirs): fsotdld: start with libgps.vapi; scheduled for upstream when completed Jan 07 20:26:48 mickeyl: http://shr.pastebin.com/SJ6ukBJm Jan 07 20:27:51 happens when calling PDP.GetContextStatus Jan 07 20:28:15 mrmoku, I bet you're missing some .debug libs Jan 07 20:28:28 GNUtoo|laptop: I installed fsogsmd-dbg Jan 07 20:28:29 I see many in Jan 07 20:28:31 but not at Jan 07 20:28:41 just the sources are not there yet Jan 07 20:28:43 still building Jan 07 20:28:47 ok Jan 07 20:30:47 mrmoku, should I build nokia900 or om-gta02? Jan 07 20:32:01 mickeyl: MDBUS2> org.freesmartphone.ogsmd /org/freesmartphone/GSM/Device org.freesmartphone.GSM.PDP.GetContextStatus Jan 07 20:32:04 org.freedesktop.DBus.Error.NoReply: Message did not receive a reply (timeout by message bus) Jan 07 20:32:17 GNUtoo|laptop: dunno Jan 07 20:33:33 mrmoku, ok, you don't need me to debug on om-gta02 anymore then? Jan 07 20:34:12 GNUtoo|laptop: thank you I'm able to finally get good backtraces :) Jan 07 20:34:19 ok Jan 07 20:36:27 mickeyl: MDBUS2> org.freesmartphone.ogsmd /org/freesmartphone/GSM/Device org.freesmartphone.GSM.PDP.SetCredentials Jan 07 20:36:32 makes it go boom too Jan 07 20:36:44 I just rebuilt for htcdream Jan 07 20:36:53 and phonefsod still has issues Jan 07 20:39:25 [libphone-ui] CRITICAL: ALSA: No control named [...] Jan 07 20:39:27 hmmm Jan 07 20:39:29 what's that? Jan 07 20:39:33 CRITICAL Jan 07 20:47:02 GNUtoo|laptop, uhm I think emtooth2 is now more robust than emtooth. I need to do a recipe but I'm not able to build my OE tree. Do you want to try building it and trying it in FR? :) Jan 07 20:47:10 and pushing recipe to OE if it works Jan 07 20:47:41 pespin, hi, if I remember well you've got no commit access.... Jan 07 20:48:08 GNUtoo|laptop, you are right :) that's your part too :P Jan 07 20:48:25 yes I'd like to try it but the only tree that is build is for the htcdream, which doesn't have a bluetooth plugin yet Jan 07 20:48:48 so send the recipe to a mailing list I'm on Jan 07 20:48:51 and I'll look Jan 07 20:48:55 if I don't look Jan 07 20:48:56 ping me Jan 07 20:48:58 on irc Jan 07 20:49:35 ok :) Jan 07 20:54:25 GNUtoo|laptop: wrong config in libphone-ui.conf Jan 07 20:54:38 ok Jan 07 20:55:32 GNUtoo|laptop: how to unmount a sshfs mount of a rebooted device? :P Jan 07 20:55:38 umount -l Jan 07 20:56:14 ok Jan 07 20:57:56 GNUtoo|laptop: does PDP.GetDeviceStatus work for you on dream? Jan 07 20:58:06 err Jan 07 20:58:09 GetContextStatus Jan 07 21:00:34 hello Jan 07 21:01:22 do we have alsactl compatible state files in SHR (unstable) ? Jan 07 21:01:47 EsEhAr: no Jan 07 21:01:58 :( Jan 07 21:02:21 where to get correct state files for playback and mic recording ? Jan 07 21:02:52 statefiles are in /etc/freesmartphone/conf/GTA02/alsa/default Jan 07 21:03:07 not alsactl compatible though Jan 07 21:03:49 how to load them ? Jan 07 21:04:07 you can load a alsa statefile with alsactl and then save it with Jan 07 21:04:08 http://git.freesmartphone.org/?p=specs.git;a=blob_plain;f=html/org.freesmartphone.Device.Audio.html;hb=HEAD#SaveScenario Jan 07 21:05:20 alsa state file : /etc/freesmartphone/conf/GTA02/alsa/default ? Jan 07 21:05:38 no Jan 07 21:05:41 that is a directory Jan 07 21:05:45 in there are the statefiles Jan 07 21:05:50 inside this Jan 07 21:06:01 but are they alsactl compatible ? Jan 07 21:06:03 no Jan 07 21:06:21 then where to get one ? Jan 07 21:06:50 root@om-gta02 /etc/freesmartphone/conf/openmoko_gta/alsa-2.6.34 # ls Jan 07 21:06:50 capturehandset gsmbluetooth gsmhandset gsmheadset gsmspeaker headset stereoout voiphandset Jan 07 21:07:14 you can load those with SetScenario Jan 07 21:07:24 oh Jan 07 21:07:34 that's greate Jan 07 21:07:43 mdbus2 is your friend :-) Jan 07 21:08:18 actually I am new to SHR Jan 07 21:09:22 can I use setScenario on terminal ? Jan 07 21:09:37 or any command for that ? Jan 07 21:10:03 it's dbus Jan 07 21:10:15 do you have mdbus2 installed? Jan 07 21:10:25 yes Jan 07 21:10:32 then do mdbus2 -si Jan 07 21:10:38 ok Jan 07 21:10:46 gives you an interactive shell for dbus commands with autocompletion Jan 07 21:11:15 org.freesmartphone.odeviced /org/freesmartphone/Device/Audio org.freesmartphone.Device.Audio.SetScenario capturehandset Jan 07 21:11:18 for example Jan 07 21:11:19 yes Jan 07 21:14:45 yes :) Jan 07 21:14:47 its done Jan 07 21:14:55 now how to exit ? Jan 07 21:15:48 GNUtoo|laptop, sent :) Jan 07 21:16:27 pespin, ok Jan 07 21:20:21 EsEhAr: Ctrl+D Jan 07 21:20:59 mrmoku:thanks Jan 07 21:21:17 yw Jan 07 21:22:40 so I should set the respected scenario before recording and playing ? Jan 07 21:24:50 for things like calls they will be automatically set by fso Jan 07 21:25:02 if you need a specific scenario, then yes that it this way Jan 07 21:25:09 s/that/set/ Jan 07 21:25:11 mrmoku meant: if you need a specific scenario, then yes set it this way Jan 07 21:27:07 no UI for that? Jan 07 21:28:14 antrik, for recording? Jan 07 21:28:34 for switching scenarios Jan 07 21:28:55 antrik, hmmm maybe in shr-settings Jan 07 21:29:07 hmm no I think those are profiles Jan 07 21:29:12 maybe I'll debug midori Jan 07 21:29:17 midori was a great browser Jan 07 21:29:24 but it segfaults too often Jan 07 21:29:25 hmm but we could add it to shr-settings, that could be great Jan 07 21:29:31 it was a fast browser Jan 07 21:29:39 eve is great but is very slow Jan 07 21:29:48 and is made for finger devices Jan 07 21:29:55 like with capacitive touchscreens Jan 07 21:30:36 hmm I should write down some kind of todo list for myself u.u Jan 07 21:30:50 yes,maybe online? Jan 07 21:30:52 in a wiki? Jan 07 21:33:46 GNUtoo|laptop, I think there was one somewhere in the wiki some months ago Jan 07 21:33:56 ok Jan 07 21:34:15 that would be great Jan 07 21:34:16 TODO: fix the TODO program on the SHR/FSO devices Jan 07 21:34:18 lol Jan 07 21:34:24 xDDDD Jan 07 21:34:36 and it's real: tasks doesn't compile anymore Jan 07 21:34:38 since ages Jan 07 21:35:03 uhm I thought that shr-image uses tasks? Jan 07 21:39:31 btw we could use trac too :P Jan 07 21:45:13 ah ok, trac tasks :) Jan 07 21:45:21 I thought you were speaking about OE Jan 07 21:47:58 GNUtoo|laptop, btw, ou can change to rev 103 instead of 100 if you want. It has some string typo fixed and some more info popup messages for errors Jan 07 21:48:55 pespin, ok Jan 07 21:51:12 freesmartphone.org: 03morphis 07next * rc60463cb5d42 10msmcomm/libmsmcomm-next/specs/structures/ (call.xml misc.xml state.xml): libmsmcomm-next: add missing xml structure definitions Jan 07 21:51:13 freesmartphone.org: 03morphis 07next * rd1cb76b26e2d 10msmcomm/libmsmcomm-next/ (5 files in 3 dirs): libmsmcomm-next: implement operation mode unsolicited response Jan 07 22:07:44 aplay -D hw /usr/share/sounds/notify_message.wav giving error Jan 07 22:07:52 channel count not available Jan 07 22:08:38 set_params:1059:channels count not available Jan 07 22:08:59 although I flashed new shr-unstable Jan 07 22:09:04 and kernel Jan 07 22:56:28 lindi-: gena2x: send mail about new kernel, i have also rebased the problematic merges in one commit, so new branch is here: https://github.com/radekp/linux-2.6/commits/qtmoko-v32.1 Jan 07 22:56:54 still missing accelerometers patch, but i leave it for tomorrow Jan 07 22:57:34 rest is described in the mail... Jan 07 23:32:17 mrmoku, uhm should I close http://shr-project.org/trac/ticket/1188 ? (suppousing I'm able to do it). Usb host is working in FR. Jan 07 23:32:33 we should clean the bugtracker a bit jeje Jan 07 23:36:49 http://shr-project.org/trac/ticket/1228 <- hwclock is working too **** ENDING LOGGING AT Sat Jan 08 02:59:57 2011