**** BEGIN LOGGING AT Mon Jul 18 02:59:57 2011 Jul 18 08:18:12 freesmartphone.org: 03morphis 07msmcomm * rfdcc33211770 10/libmsmcomm/msmcomm/messagedisassembler.vala: libmsmcomm: register missing sound solicited and unsolicited message groups Jul 18 08:24:26 mickeyl: ping Jul 18 09:00:49 freesmartphone.org: 03morphis 07msmcomm * r7e9a4a972100 10/ (7 files in 6 dirs): libmsmcomm: implement name sound devices class and sub classes Jul 18 09:01:24 GarthPS: heyho Jul 18 09:05:27 morphis: hey hi Jul 18 09:05:37 GarthPS: you tried the image yesterday? Jul 18 09:05:57 morphis: not yet Jul 18 09:06:02 ok Jul 18 09:06:22 morphis: what do I need to check particularely ? Jul 18 09:06:44 suspend, pin request Jul 18 09:06:52 and making a call Jul 18 09:07:02 morning *yawn* Jul 18 09:07:03 but you will not hear anything yet when making the call Jul 18 09:07:09 mickey|office: good morning! Jul 18 09:07:36 morphis: i've seen you making large changes in cornucopia. excellent work, man Jul 18 09:07:46 mickey|office: thanks :) Jul 18 09:08:04 aurora takes me to a lot of issues in FSO we had Jul 18 09:08:15 especially with the usage stuff Jul 18 09:08:16 excellent. then it already serves its purpose Jul 18 09:08:33 yes Jul 18 09:09:07 and you are already can make a call with aurora :) Jul 18 09:09:08 good! :) (moin mickey|office) Jul 18 09:09:17 hi GarthPS Jul 18 09:09:19 awesome! Jul 18 09:09:20 I am doing the audio routing bits right now Jul 18 09:09:30 and suspend stuff seems to be solved too .... Jul 18 09:09:35 after a lot of work Jul 18 09:09:41 morphis: you dfid not finished your last phrase about making call Jul 18 09:10:00 morphis: oh Jul 18 09:10:07 sorry I am wrong :) Jul 18 09:10:10 :) Jul 18 09:10:14 miss reading :p Jul 18 09:10:35 GarthPS: do you have some time for working on a aurora-installer? Jul 18 09:11:17 mickey|office: we need to work on the presentation too Jul 18 09:11:29 and I will go by car too Jul 18 09:11:31 morphis: hmm not today I think. sorry to much to do Jul 18 09:11:54 as I have another appointment the next day Jul 18 09:12:04 morphis: I test the current image or are you building an other one (as the pre image is freesher than pre2) Jul 18 09:12:21 yes, I rebuilded it, but try it Jul 18 09:12:31 it's not any "real" fix included Jul 18 09:14:07 mickey|office, GarthPS: I found a very good UI design guideline we can use as base for aurora: http://www.developer.nokia.com/Resources/Library/Symbian_Design_Guidelines/index.html Jul 18 09:14:40 mickey|office: and I have some questions for you if you have some free time Jul 18 09:26:54 freesmartphone.org: 03morphis 07msmcomm * r8cea4baf7656 10/msmcomm-specs/src/sound.vala: msmcomm-specs: fix marshalling of new enumeration types for sound device Jul 18 09:36:04 morphis: ya, UI guidelines always good. the symbian ones are at least in use for a long while Jul 18 09:36:15 yes Jul 18 09:36:16 morphis: after lunch today would be good Jul 18 09:36:28 and they fit for our needs :) Jul 18 09:36:31 mickey|office: ok Jul 18 09:40:54 morphis: first it is hard to type "ok" for the pin screen. Jul 18 09:42:21 ok Jul 18 09:42:38 can you please open a bug report for all stuff you find? Jul 18 09:44:12 morphis: yep . plus look at this Jul 18 09:44:20 2011-07-18T08:16:33.581451Z [DEBUG] MsmcommDaemonHciModemChannel <>: MSMCOMM_LOW_LEVEL_MESSAGE_TYPE_RESPONSE_SIM_CALLBACK Jul 18 09:44:21 2011-07-18T08:16:33.603454Z [CRITICAL] msmcommd : GLib : _lambda21_: assertion `conn != NULL' failed Jul 18 09:44:23 .. Jul 18 09:44:31 2011-07-18T08:16:33.311126Z [INFO] QualcommPalmModem <>: SIM Auth status now FREE_SMARTPHONE_GSM_SIM_AUTH_STATUS_READY Jul 18 09:44:31 2011-07-18T08:16:33.590972Z [CRITICAL] fsogsmd : GLib : file phonebookhandler.c: line 666: uncaught error: GDBus.Error:org.freedesktop.DBus.Error.NoReply: Message did not receive a reply (timeout by message bus) (g-dbus-error-quark, 4) Jul 18 09:44:56 I don't know if it is linked.. Jul 18 09:44:58 ok, I need to rebuild the image for the palmpre2 tooo Jul 18 09:45:13 I had the same issue with the palmpre image Jul 18 09:45:22 morphis: ok so for now I don't repport this until I have it with the new image right ? Jul 18 09:45:24 it's fixed by rebuilding the msmcomm components Jul 18 09:45:28 yes Jul 18 09:45:34 but the pin button thing you can report Jul 18 09:45:38 and you can check suspend Jul 18 09:45:42 morphis: ok. ping me when it is builded Jul 18 09:45:46 yep Jul 18 09:45:47 will do Jul 18 09:46:26 morphis: actually with usb pluged the screen goes black but I think It does not suspend..am I right ? Jul 18 09:48:32 you can check this with logging in via ssh Jul 18 09:48:40 and take a look at dmesg and fsousaged.log Jul 18 10:25:09 morphis: I still no suspend : have fastpath and Resource CPU not suspended: not resuming Jul 18 10:28:49 GarthPS: respect, when usb is plugged in then the device will not suspend! Jul 18 10:29:31 morphis: yes but this is when usb is not pluged => the screen goes black and resume alone after few seconds Jul 18 10:31:07 can you give me the dmesg and fsousage.log files? Jul 18 10:34:20 morphis: http://pastebin.com/f77NBdS3 Jul 18 10:36:36 hm, misteriou Jul 18 10:36:36 s Jul 18 10:36:39 the suspend fails Jul 18 10:37:12 GarthPS: which kernel do you use? Jul 18 10:37:21 morphis: yours Jul 18 10:40:17 from the amethyst? Jul 18 10:41:17 morphis: yep Jul 18 10:41:40 ok Jul 18 12:04:37 hey there Jul 18 12:05:22 I'm having problems with compiling gcc with arm host, I get Jul 18 12:05:23 | /OE/shr-unstable/tmp/sysroots/x86_64-linux/usr/armv6/lib/gcc/arm-oe-linux-gnueabi/4.5.3/../../../../arm-oe-linux-gnueabi/bin/ld: error: .libs/selected_int_kind.o uses VFP register arguments, .libs/libgfortran.so.3.0.0 does not Jul 18 12:05:30 me too Jul 18 12:05:34 do I even need fortran support for shr? Jul 18 12:05:38 no Jul 18 12:05:49 is there an easy way to disable it? Jul 18 12:05:55 maybe Jul 18 12:06:13 look in the recipes of gcc Jul 18 12:06:21 looks pretty plain Jul 18 12:06:54 at least I haven't found any way to add a "--disable-fortran" or something of that kind flag Jul 18 12:07:09 grep for fortran Jul 18 12:07:20 you'll see variables with fortran inside Jul 18 12:07:28 mainly patches Jul 18 12:07:33 not only Jul 18 12:08:19 hmm Jul 18 12:08:41 in gcc-package-target.inc is a list of packages Jul 18 12:09:17 gcc-4.5.inc:FORTRAN = "" Jul 18 12:09:17 gcc-4.5.inc:FORTRAN_linux-gnueabi = ",fortran" Jul 18 12:09:22 looks promising Jul 18 12:12:30 also libgisi failed Jul 18 12:12:33 however I found http://permalink.gmane.org/gmane.comp.handhelds.openembedded/46249 Jul 18 12:12:59 As with the latest changes to libgisi the mentioned posix-ext is removed and it's using only posix as dependency. Will provide a patch to fix this later. Jul 18 12:13:23 so its known, where do I get the patch from? Jul 18 12:13:36 or is it sufficient to bump libgisi to the new SRCREV? Jul 18 13:25:26 marcellus1, libgisi isn't strictly necessary Jul 18 13:25:32 it's a lib for the n900 Jul 18 13:25:44 how did it fail? Jul 18 13:25:57 try to add autorev for fso and shr Jul 18 13:28:54 I bumped the SRC_URI and it compiled Jul 18 13:29:57 however gcc is a different deal, I disabled fortran, however bitbake builds the rest of the image while gcc is still compiling, that could take a while Jul 18 13:33:03 ok Jul 18 13:34:05 SHR: 03Martin.Jansa 07meta-smartphone * r8506c4ce9d2f 10/meta-shr/conf/distro/include/preferred-shr-versions.inc: SHR: prefer gcc 4.6% now when versioning was changed Jul 18 13:37:19 mickey|office: you now have some time? Jul 18 13:37:27 morphis: yes Jul 18 13:37:37 great Jul 18 13:38:00 first question is about sound handled by the modem Jul 18 13:38:16 we have two methods for this in our dbus api, setvolume and setmicrophonemute Jul 18 13:38:42 mickey|office: if you look at this http://git.freesmartphone.org/?p=msmcomm.git;a=blobdiff;f=libmsmcomm/msmcomm/soundmessage.vala;h=f9ebbb817546a42da1b66455029d5c90b52929d9;hp=a3e3e6e3bebf6f038bb54fee0ed8ee58f375691b;hb=7e9a4a972100ed27a2d387151858cc20dfc6a1a8;hpb=fdcc332117704fddb9b087e8a657933785581a12 Jul 18 13:39:13 I have to set the sound device of them modem when a call starts to device_class = HANDSET, device_sub_class = DEFAULT Jul 18 13:39:31 and reset this when the call ends Jul 18 13:39:33 marcellus1: it's known.. just pushed SRCREV bump Jul 18 13:39:36 where should I do this best? Jul 18 13:39:58 JaMa: you are back! how was your holiday? Jul 18 13:40:17 JaMa: i know, I posted the bugreport earlier Jul 18 13:40:31 JaMa: also I did bump it Jul 18 13:41:02 marcellus1: http://trac.shr-project.org/trac/ticket/1459 Jul 18 13:41:29 morphis: holiday was great.. I want to go back, but no more vacation days :/ Jul 18 13:41:41 morphis: two options: either in fsoaudiod as a plugin for the palm. or, if you consider it being a very lowlevel detail that has no configuration, directly in fsogsmd's custom callhandler for the pre Jul 18 13:42:31 mickey|office: ok, a fsoaudiod plugin is already what I thought about Jul 18 13:43:07 I need to play a little bit more with it as you can do a lot things with it, so seems to be better when it is in fsoaudiod Jul 18 13:43:16 mickey|office: ok, next question Jul 18 13:44:13 mickey|office: you maybe saw I implemented a self-healing functionality in the AbstractDBusResource class Jul 18 13:44:33 mickey|office: this means, when fsousaged dies (which happens quite often to me in the last time) all resource registrations are lost Jul 18 13:44:52 JaMa: is the patch in master? Jul 18 13:44:59 mickey|office: with the additional lines the resource will watch the bus of fsousaged to reappear Jul 18 13:45:09 mickey|office: and when it comes back it registers the resource again Jul 18 13:46:05 yes, saw that Jul 18 13:46:07 JaMa: hehe, my vacations started today but I have to do an exam on wednesday so "real" holidays will start after this Jul 18 13:46:19 mickey|office: whats your opinion about this? Jul 18 13:46:19 hi everyone! Jul 18 13:46:22 angelox|n900: heyho Jul 18 13:48:43 morphis: it's a good change short term. of course, fsousaged should never die, it needs to be scrutinized why it does for you. mid term, i'd rather change the API, since i think i got it backwards. Eventually all resources should be converted to shadow resources and I think fsousaged should always trigger the registration, as opposed to the client. Jul 18 13:48:53 s/client/resource provider/ Jul 18 13:48:53 mickey|office meant: morphis: it's a good change short term. of course, fsousaged should never die, it needs to be scrutinized why it does for you. mid term, i'd rather change the API, since i think i got it backwards. Eventually all resources should be converted to shadow re... Jul 18 13:50:33 however it may also be interesting to evaluate offloading some of that to a proper system component monitor, such as systemd or something along that lines Jul 18 13:51:02 mickey|office: can a dbus listener get the PID of of a sender process based on a message? Jul 18 13:51:33 DocScrutinizer: yes, that's possible Jul 18 13:51:35 marcellus1: yes.. Jul 18 13:51:44 mickey|office: thanks Jul 18 13:51:54 morphis: good for you :) Jul 18 13:52:05 mickey|office: yes, that should be the way in the long term, but for now it works Jul 18 13:52:42 mickey|office: the problem why fsousaged dies is because of the exception handling, most exception are not handled Jul 18 13:52:56 morphis: dbus ones? Jul 18 13:53:05 or internal/glib? Jul 18 13:53:29 most times glib/internal Jul 18 13:53:43 there are missing "throws GLib.Error" in the code Jul 18 13:53:46 ok, that needs to be fixed asap then Jul 18 13:53:49 but it's no that easy to add them Jul 18 13:54:04 needs some more time to get it right as I have currently Jul 18 13:54:34 but it's something we should fix in the near future to get a stable and reliable implementation Jul 18 13:55:49 mickey|office: next thing is our presentation at froscon Jul 18 13:56:23 yeah. we still have a bit of time, but it's getting close. i really try to get something to you during this week Jul 18 13:56:37 my time management sucks even more than before i was dad Jul 18 13:56:40 :) Jul 18 13:56:45 :) Jul 18 13:57:45 would be nice if you can send me your presentation style then I can start with putting my stuff on the slides Jul 18 13:58:02 right. i'll take care of that when i'm home Jul 18 13:58:06 playya_: ping Jul 18 13:58:11 OT: if you ever wonder why the hell you do not see pci ethernet card on the bus in one's hp "business" laptop, just plug it in the network and it'll appear :-O Jul 18 13:58:13 btw. Jul 18 13:58:22 did you see that connman starts to have support for tethering? Jul 18 13:58:27 no Jul 18 13:58:35 but thats nice Jul 18 13:58:36 we can probably obsolete the sharing part of fsonetworkd soon Jul 18 13:59:03 yeah they are already started with integrating some personal firewall stuff Jul 18 13:59:16 http://connman.net/blogs/sameo/2010/connman-064 Jul 18 13:59:29 shame that it took so long, but better late than never Jul 18 13:59:33 PaulFertser: heh Jul 18 13:59:42 PaulFertser: that's odd Jul 18 14:00:14 mickey|office: sounds like something we will integrate in the next aurora versions :) Jul 18 14:00:43 morphis: yes. proper holistic network handling is a very important task these days. Jul 18 14:01:17 yes, and it will take to much work and time to get it done on our own and connman seems to be in a very good shape these days Jul 18 14:01:35 so I should really polish the fsogsmd connman plugin in the near future :) Jul 18 14:01:43 yeah Jul 18 14:02:01 we also have that bluez plugin from Felix Huber Jul 18 14:02:09 that should be uplevelled to FSO2 and then submitted again Jul 18 14:02:11 yes, that is somewhere I never found it :) Jul 18 14:02:31 he sent me the last version some months ago, i'll try to dig it out Jul 18 14:06:05 ok Jul 18 14:06:54 morphis: Hi again,sorry,do you have some small time to help me to understand why my agents.py test didn't work? Jul 18 14:10:43 angelox|laptop: which one? Jul 18 14:11:39 morphis: the link is too big,can i write it here anyway? Jul 18 14:11:54 or pastebin the link (LOL) Jul 18 14:12:15 a link to what? Jul 18 14:12:39 mickey|office: can you activate your jabber? Jul 18 14:13:36 morphis: it's and diff of agents.py that i have added a test function,it's on git,if you want you can go into aurora-daemons on my branch and see my agents.py function getBrightness Jul 18 14:14:12 angelox|laptop: this one http://git.freesmartphone.org/?p=aurora.git;a=blobdiff;f=aurora-daemon/aurora/extensions/agents.py;h=b1e599d6005c1a261f65386d0b35e9be6010ef6e;hp=4cd516442ac8b46029f1fb932f10a6aa97ce9fed;hb=0ffc3a7f0925c68b352877cb1034375dc70ce824;hpb=4bccaf85569477cb234f18bbf09700ff5b17ccda Jul 18 14:14:13 ? Jul 18 14:14:44 yes,exactly :) Jul 18 14:15:19 what do you want to do in the one? Jul 18 14:15:33 from where comes the variable value? Jul 18 14:15:56 and why do you supply "0" as first parameter of getBrightness? Jul 18 14:17:08 because,since i'm bad in python and new in fso,i did copy setBrightness function :( sorry. if i remove that '0' it'll work? Jul 18 14:17:28 i tried read api but that shows only one parameter iirc Jul 18 14:18:42 yes Jul 18 14:19:08 the reply_handler and error_handler are python specific and not part of the api Jul 18 14:19:29 maybe this will help you: http://dbus.freedesktop.org/doc/dbus-python/doc/tutorial.html Jul 18 14:19:36 and removing the '0' will solve it Jul 18 14:20:26 wow,i was using that link but didn't see the final of it :) Jul 18 14:21:01 morphis: thank you,and sorry {dbus,python,fso}newbie question :) Jul 18 14:21:33 morphis: done Jul 18 14:22:14 angelox|laptop: no problem Jul 18 14:24:27 angelox|laptop: I started like you some years ago :) Jul 18 14:25:51 morphis: that helps me to be a bit more happy :) Jul 18 14:26:22 angelox|laptop: hehe, you can't ask enough questions Jul 18 14:39:15 * angelox|laptop discovered something cool,google translator 'listen' to what you talk and write on the screen,that helps he to improve his English Jul 18 15:05:03 SHR: 03Martin.Jansa 07shr-chroot * rb00b3d5d0acb 10/ (47 files in 15 dirs): bitbake upgrade Jul 18 15:05:15 SHR: 03Martin.Jansa 07shr-chroot * r5b49cc763d12 10/ (12710 files in 1356 dirs): system upgrade Jul 18 16:04:45 ~hail JaMa Jul 18 16:04:45 * apt bows down to JaMa and chants, "I'M NOT WORTHY!!" Jul 18 16:33:53 morphis_: with that change,simply calling "var test = dbus_DeviceDisplayAgent.getBrightness();" would work isn't ? Jul 18 16:35:07 in qml i mean Jul 18 18:09:18 freesmartphone.org: 03felix.huber 07zhone * rc697191016a3 10/src/zhone: Bugfix for dtmf exit button Jul 18 18:25:05 freesmartphone.org: 03felix.huber 07zhone * re596d1c867a7 10/ (4 files in 2 dirs): Add provider selection and more comfort functions Jul 18 18:53:25 hi mickeyl Jul 18 18:53:35 is it possible to compile mdbus2 statically? Jul 18 19:42:39 GNUtoo|laptop: I'm sure it is, but the problem will be the dependencies. You will have a hard time to compile glib2, dbus, etc. statically Jul 18 19:43:06 ah ok Jul 18 19:43:08 wow, Herr Huber still working with zhone Jul 18 19:43:12 amazing Jul 18 19:43:57 mickeyl, bluetoothd now starts automatically on om-gta02 with the rfkill plugin Jul 18 19:45:58 I guess compiling glib for android would be a hard time too Jul 18 19:47:19 then what's next for full om-gta02 bluetooth support? Jul 18 19:52:17 phone and applevel integration i guess Jul 18 19:52:32 phone integration, e.g. seamless connecting to known headsets/carkits Jul 18 19:52:46 applevel integration, e.g. tethering, vcard sharing, etc. Jul 18 19:57:27 ok Jul 18 19:57:41 but you should talk with mrmoku` Jul 18 19:57:53 for the audio routing Jul 18 20:04:37 no, at least not about audio routing :) Jul 18 20:06:11 so I still wonder how to route PCM to bluetooth Jul 18 20:06:29 I've no idea on how bluetooth work beside on how it's connected to the CODEC Jul 18 20:07:13 2011-05/#openmoko-cdevel.16.log:16-05-2011 18:36:31 < mickeyl!~mickey@openmoko/coreteam/mickey: morphis is the audio master for now Jul 18 20:07:14 hi larsc Jul 18 20:07:26 ok Jul 18 20:08:41 there was one config file with an important option... Jul 18 20:10:01 brb Jul 18 20:14:54 freesmartphone.org: 03felix.huber 07zhone * r87487d0defe2 10/src/zhone: Add mising comfort function code Jul 18 20:21:56 hi JaMa larsc Jul 18 20:24:41 where are larsc or JaMa Jul 18 20:24:43 ? Jul 18 20:29:09 GNUtoo|laptop: do you have the correct entry for SCORouting in /etc/bluetooth/audio? Jul 18 20:31:04 SCORouting=PCM Jul 18 20:31:05 that? Jul 18 20:33:02 yup Jul 18 20:34:05 do you have a clear idea on how it works Jul 18 20:34:40 at the hardware(like schematics etc...) and software level(like bluez etc...) and application level(like you run emtooth2 etc...) Jul 18 20:35:07 GNUtoo|laptop: i do have a clear idea about SCORouting Jul 18 20:35:12 nice Jul 18 20:35:30 first what is the difference between a2dp and SCO? Jul 18 20:35:48 GNUtoo|laptop: a2dp is unidirectional, stereo; SCO is bidirectional, mono. Jul 18 20:35:57 ah ok Jul 18 20:36:05 so I need A2DP for music Jul 18 20:36:08 and sco for calls Jul 18 20:36:11 GNUtoo|laptop: yes, exactly. Jul 18 20:36:16 so a2dp cannot arecord Jul 18 20:36:17 ok Jul 18 20:36:34 do you also appear to know how a2dp works? Jul 18 20:36:37 GNUtoo|laptop: alsa bluetooth plugin is automatically using SCO for input. Jul 18 20:36:57 because there is a gsmbluetooth scenario Jul 18 20:36:59 so I'll try it Jul 18 20:37:04 and I guess it will work Jul 18 20:37:07 for me a2dp works in anywhere,since i'm deaf of one ear :) Jul 18 20:37:09 GNUtoo|laptop: i think a2dp works as usual: alsa bluetooth plugin asks bluetoothd for that, and it travels over HCI (i.e. USB) on our hardware. Jul 18 20:37:28 but I'm unsure about a2dp, it works with .asoundrc but I've no idea if it can be routed with alsamixer Jul 18 20:37:33 GNUtoo|laptop: (a2dp) eso that's totally unrelated to ALSA states. Jul 18 20:37:47 ahhh ok Jul 18 20:38:01 so the following is the only way to make a2dp work: Jul 18 20:38:13 GNUtoo|laptop: a2dp sound source is always the SoC itself, you can in theory use "sox" or whatever to record something via SoC's DAI and then send over usb to BT chip. Jul 18 20:38:33 http://wiki.openmoko.org/wiki/User:GNUtoo Jul 18 20:39:19 is that the only way to get audio with bluetooth? Jul 18 20:39:25 if so how to handle the following: Jul 18 20:39:26 GNUtoo|laptop: that looks sane except that i wouldn't call it !default because i'd prefer mplayer/whatever to use the SoC's DAI for the output by default. Jul 18 20:39:36 ok Jul 18 20:39:41 but then how about: Jul 18 20:39:56 *plug plugin Jul 18 20:39:56 GNUtoo|laptop: so i'd add another alsa device (probably bluetooth) and used something like "mplayer -ao alsa:device=bluetooth". Jul 18 20:40:02 *dmix plugin Jul 18 20:40:10 ok Jul 18 20:40:22 basically I want everything to go on my bluetooth headset: Jul 18 20:40:24 *calls Jul 18 20:40:27 *ringing Jul 18 20:40:33 GNUtoo|laptop: you should treat a2dp basically as just an additional soundcard. Jul 18 20:40:34 *music, sound Jul 18 20:40:36 *games Jul 18 20:40:42 GNUtoo|laptop: now we come to the tricky part: SCO. Jul 18 20:41:26 GNUtoo|laptop: SCO in theory can be used in two ways: it can get data from/to HCI (SoC<->USB<->HCI) or from hardware PCM interface (that is connected to wm8753 on gta02). Jul 18 20:41:50 GNUtoo|laptop: our BT chips are configured (with an eeprom setting) to always use the second method. Jul 18 20:42:06 yes I saw the connection in both the schematics and the sound wiki article Jul 18 20:42:13 ok Jul 18 20:42:18 GNUtoo|laptop: unless the setting is changed and the chip is rebooted, it can't get/set audio data for SCO via HCI at all. Jul 18 20:42:29 ok Jul 18 20:43:26 GNUtoo|laptop: that particular bluez config switch makes bluetoothd aware that SCO is routed over PCM, and enables a corresponding dbus api call to activate SCO. The chip then expects the PCM data to flow over its interface. Jul 18 20:43:49 So you have to configure wm8753 accordingly etc etc. Jul 18 20:44:40 the voice dbus thing Jul 18 20:44:52 GNUtoo|laptop: iirc it was the Play() call. Jul 18 20:45:03 ok I didn't remember well Jul 18 20:45:05 indeed play Jul 18 20:45:08 it never worked Jul 18 20:45:16 maybe because I wasn't in call Jul 18 20:45:35 ah now it works Jul 18 20:45:58 GNUtoo|laptop: i can assure you it worked for me at one point of time with FSO (and a "compatible" headset). Jul 18 20:47:08 ok Jul 18 20:47:12 play works Jul 18 20:47:16 but I ear nothing Jul 18 20:47:19 I did that: Jul 18 20:47:37 mdbus2 -s org.bluez /org/bluez/255/hci0/dev_68_A3_C4_C5_AC_B3 org.bluez.Headset.Play Jul 18 20:47:39 then that: Jul 18 20:47:55 GNUtoo|laptop: there also established a bidirectional comminication channel between the headset and bluetoothd with AT-commands to attenuate volume/mic sensitivity and "unsolicited replies" for the buttons. Jul 18 20:47:56 mdbus2 -s org.freesmartphone.odeviced /org/freesmartphone/Device/Audio org.freesmartphone.Device.Audio.SetScenario gsmbluetooth Jul 18 20:48:16 ah that's why I get buttons stuff Jul 18 20:48:24 I tought it was hci Jul 18 20:48:58 GNUtoo|laptop: and also there used to be a kernel bug (and it probably still is) that made one use two additional amixer calls for SCO over PCM to work. Jul 18 20:49:10 GNUtoo|laptop: no, buttons work regardless of SCO routing. Jul 18 20:50:06 GNUtoo|laptop: probably you have to call Play and then stay on the dbus bus for as long as you want it to Play. Jul 18 20:50:15 ah ok Jul 18 20:50:17 GNUtoo|laptop: pong Jul 18 20:50:25 but rtt still high Jul 18 20:50:39 JaMa, ok do you have access to xf86-video-glamo Jul 18 20:50:46 I've a very small change/request Jul 18 20:50:48 GNUtoo|laptop: when Play() is active you should hear an ambient amp "noise" regardless of everything else. Jul 18 20:50:54 yes I did Jul 18 20:51:37 GNUtoo|laptop: no I don't, larsc or Weiss Jul 18 20:51:45 ok Jul 18 20:51:48 GNUtoo|laptop: and also you should at least once issue a special call to configure BT DAI to use corresponding (8000 rate, mono, etc) format that BT chip expects. Jul 18 20:51:51 JaMa, thanks for the infos Jul 18 20:52:00 ok Jul 18 20:52:39 GNUtoo|laptop: if you get some changes to that repo please include also recipes/xorg-driver/xf86-video-glamo/0001-glamo-drm-define-GLAMO_CMDQ_MAX_COUNT-instead-of-mag.patch Jul 18 20:52:56 GNUtoo|laptop: take a look here: http://git.freesmartphone.org/?p=framework.git;a=blob;f=framework/subsystems/ophoned/headset.py;h=c59173e1e452ed07f7828d46657c31737f2bbea8;hb=HEAD Jul 18 20:53:28 JaMa, the changes are sys node API change for qvga Jul 18 20:53:51 PaulFertser, ok thanks a lot Jul 18 20:54:56 GNUtoo|laptop: np; i sometimes feel i know quite a lot about quite useless stuff. Everybody's busy with "fancier" hardware and i'm stuck with gta02... Jul 18 20:56:04 PaulFertser: i used one Motorola A1200,400Mhz CPU,without GPS,Wifi,3G... :) Jul 18 20:56:17 angelox|laptop: i remember :) Jul 18 20:57:13 GNUtoo|laptop: please also see https://github.com/radekp/gta02-gsm-bt-fix/blob/b8bbf0c9c2f28e144fe550f37ac0f22984bafca8/gta02-gsm-bt-fix.c Jul 18 20:57:32 useless??:) Jul 18 20:58:04 In fact https://github.com/radekp/gta02-gsm-bt-fix/blob/b8bbf0c9c2f28e144fe550f37ac0f22984bafca8/common.h Jul 18 20:58:34 tmzt: yes, it's basically useless for me because my gta02 still works "suboptimally" to say the least and i'm unable to do anything about it... Jul 18 20:59:02 I meant useless knowledge Jul 18 20:59:09 gta02-gsm-bt-fix.c -> what does it do? Jul 18 20:59:42 ahhh now I understand better with the header Jul 18 20:59:42 GNUtoo|laptop: that's the stuff qtmoko folks use to make SCO work for GSM voice calls. See the corresponding .h file (my second link) for the actual code. Jul 18 20:59:57 Yep, it's clearly a sick idea to write code in the header :/ Jul 18 21:01:27 tmzt: so did i. What sense in knowing something if it doesn't help to make it work... The smartphones world is still so sick it makes me wonder if i get a device that simply works _ever_... Jul 18 21:02:51 ah, quite true. unfortunately I've learned to live with android and my g2/desire z Jul 18 21:03:10 ported X to android but don't use it, helped with the 'rooting' but don't use that either Jul 18 21:03:57 If my FR stops working i'll go back to using featurephones (the old nokia 1100 my friend found in the mud still works just fine). Jul 18 21:04:10 Android is quite strange to me,principally the busybox part :\ Jul 18 21:04:58 always nicer to have debian chroot Jul 18 21:05:12 PaulFertser: That's why i have about 5 phones here(one of them is a very-fantastically-old CDMA phone) :) Jul 18 21:05:26 angelox|laptop: how many phones do you carry with you? ;) Jul 18 21:05:39 wow Jul 18 21:05:47 it works with radek's program Jul 18 21:05:50 GNUtoo|laptop: anything else you might want to ask me before i fall asleep? Jul 18 21:06:15 GNUtoo|laptop: it's Gabrys's program in fact, i tought him the theory, he wrote the code :) Jul 18 21:06:22 for a2dp, I've issues with one on the plugins Jul 18 21:06:27 I want 2 plugins: Jul 18 21:06:34 *plug Jul 18 21:06:38 *dmix Jul 18 21:06:53 since it seem to support mmap it should work with dmix but it refuses to do so Jul 18 21:08:20 [AO_ALSA] alsa-lib: pcm_direct.c:1490:(_snd_pcm_direct_get_slave_ipc_offset) Invalid type 'bluetooth' for slave PCM Jul 18 21:08:23 that's what it says Jul 18 21:08:30 that is quite problematic Jul 18 21:08:46 since it wouldn't be able to ring on the bluetooth headset while playing music Jul 18 21:13:56 Interesting alsa stuff, some higher level abstraction, DocScrutinizer: http://answerpot.com/showthread.php?1006883-alsa-lib%3A+UCM+-+Use+Case+Manager Jul 18 21:14:03 PaulFertser: ahh, falling asleep is a very good idea :-) Jul 18 21:14:26 mrmoku: i had too much of a nap during the daytime, so probably it won't work as easy as it should. Jul 18 21:14:35 :) Jul 18 21:14:45 * mrmoku not so :D Jul 18 21:14:48 gnight all Jul 18 21:14:56 good night Jul 18 21:15:01 mrmoku: night Jul 18 21:15:06 mrmoku: gnight Jul 18 21:15:14 good night Jul 18 21:15:30 i have about 4 hours before go to sleep Jul 18 21:15:32 :) Jul 18 21:16:50 GNUtoo|laptop: 1022 if (snd_pcm_type(spcm) != SND_PCM_TYPE_HW) { Jul 18 21:16:54 1023 SNDERR("dmix plugin can be only connected to hw plugin"); Jul 18 21:17:11 yes usual thing Jul 18 21:17:18 but what can I do to workarround? Jul 18 21:17:28 using pulseaudio is not a good idea Jul 18 21:17:41 ( DocScrutinizer would kill me if I do ) Jul 18 21:17:54 I'm here, I'm here Jul 18 21:18:03 hahaha Jul 18 21:18:42 ooh, missed your post PaulFertser Jul 18 21:20:09 sorry,i don't undestand about Audio things,why pulseaudio is bad? Jul 18 21:20:18