**** BEGIN LOGGING AT Sat Oct 29 02:59:56 2011 Oct 29 07:51:11 freesmartphone.org: 03morphis 07cornucopia * r7e450b768f30 10/libfsobasics/fsobasics/utilities.vala: Oct 29 07:51:11 freesmartphone.org: libfsobasics: add simple method to convert a glib list into an array Oct 29 07:51:11 freesmartphone.org: Signed-off-by: Simon Busch Oct 29 07:51:14 freesmartphone.org: 03morphis 07cornucopia * r789c0b866ecc 10/fsogsmd/src/plugins/modem_samsung/ (Makefile.am callhandler.vala mediators_call.vala plugin.vala): Oct 29 07:51:14 freesmartphone.org: fsogsmd: modem_samsung: implement most stuff for call handling; still untested ... Oct 29 07:51:14 freesmartphone.org: I implemented this in the train so I didn't have the possiblity to test everything; I just Oct 29 07:51:14 freesmartphone.org: verified that everything compiles fine and integrates in the FSO architecture. Oct 29 07:51:15 freesmartphone.org: Signed-off-by: Simon Busch Oct 29 07:51:15 freesmartphone.org: 03morphis 07cornucopia * r7d404b62909c 10/fsogsmd/src/plugins/modem_samsung/unsolicited.vala: Oct 29 07:51:16 freesmartphone.org: fsogsmd: modem_samsung: reorder debug statements for network registration urc message Oct 29 07:51:16 freesmartphone.org: Signed-off-by: Simon Busch Oct 29 07:51:17 freesmartphone.org: 03morphis 07cornucopia * r377c0d580cff 10/fsogsmd/src/plugins/modem_samsung/ (channel.vala mediators_device.vala): Oct 29 07:51:46 freesmartphone.org: fsogsmd: modem_samsung: request baseband version information in the correct mediator Oct 29 07:51:46 freesmartphone.org: Signed-off-by: Simon Busch Oct 29 08:27:46 JaMaOff: your elfe fix is in... time to bump :) Oct 29 08:28:00 I know and I have it here Oct 29 08:28:02 with more stuff Oct 29 08:28:06 :) Oct 29 08:28:11 do you want it now? :) Oct 29 08:28:16 what do we do with dbus... revert that commit? Oct 29 08:28:27 I have work around for that Oct 29 08:28:33 calling chowns in postinst Oct 29 08:28:35 as it was before Oct 29 08:28:36 ok Oct 29 08:28:44 good Oct 29 08:28:58 * mrmoku wants all of it :-D Oct 29 08:28:59 waiting for image build to test it all Oct 29 08:29:02 ok Oct 29 08:30:50 build finished, pushing Oct 29 08:33:35 done Oct 29 08:33:46 SHR: 03Martin.Jansa 07libphone-ui-shr * rcf8c0d8a4eec 10/src/ (7 files in 2 dirs): replace elm_layout_content_set with elm_object_content_part_set to adapt to API change r64403 Oct 29 08:41:25 * mrmoku updating Oct 29 08:45:17 JaMaOff: for gta04 I would like to use gpsd instead of fso-gpsd Oct 29 08:45:31 JaMaOff: we have a provider set which I can override for gta04 Oct 29 08:45:50 but still it (fso-gpsd) is hardcoded in task-minimal Oct 29 08:46:09 replacing it there with gpsd is what we want? Oct 29 08:51:54 iirc task-minimal is not machine arch Oct 29 08:52:01 so it won't work very well Oct 29 08:52:25 even if we have a provider for gpsd? Oct 29 08:52:32 what is it for then? Oct 29 08:52:47 ahh... no I understand Oct 29 08:52:51 +w Oct 29 08:53:14 the override for gta04 won't work, right? Oct 29 08:53:50 hmm... how do we solve that then? Oct 29 08:54:05 recommend the correct one in machine.conf and remove it from task-minimal? Oct 29 08:55:05 SHR: 03Martin.Jansa 07libphone-ui-shr * r5dc22704dba9 10/src/util/ui-utils.c: replace elm_notify_content_set with elm_object_content_set to adapt to API change r64502 Oct 29 08:55:34 mrmoku: that would work Oct 29 08:56:04 and even allow for machines without gps :-) Oct 29 08:56:07 mrmoku: ahh minimal is MACHINE_ARCH Oct 29 08:56:12 so feel free to override it there Oct 29 08:56:15 hehe, ok Oct 29 08:56:25 and include gpsd instead of fso-gpsd there Oct 29 08:57:00 and you can use something like VIRTUAL_RUNTIME_gpsd provider Oct 29 08:57:28 ahh, ok Oct 29 08:58:15 mrmoku: how do you plan to do AGPS style stuff with gpsd? Oct 29 08:58:40 lindi-: no plans... my hope is the gps on gta04 is fast enough without :P Oct 29 08:59:12 SHR: 03Martin.Jansa 07libphone-ui-shr * r2e8275d2fffd 10/src/view/sim-manager-view.c: replace elm_frame_content_set with elm_object_content_set to adapt to API change r64502 Oct 29 08:59:27 can't remember the chip Oct 29 09:02:19 lindi-: wd2g004 Oct 29 09:02:26 err Oct 29 09:02:33 w2sg004 Oct 29 09:02:46 heh Oct 29 09:02:51 * mrmoku still fills sick :P Oct 29 09:02:55 lindi-: w2sg0004 Oct 29 09:02:58 now its correct :) Oct 29 09:03:36 mrmoku: they say cold start is 36 seconds Oct 29 09:03:57 yup, I could live with that Oct 29 09:04:00 so it's about the same as with gta02 Oct 29 09:36:09 freesmartphone.org: 03morphis 07cornucopia * rfa70f283558f 10/fsogsmd/src/plugins/modem_samsung/mediators_call.vala: Oct 29 09:36:09 freesmartphone.org: fsogsmd: modem_samsung: fix compilation error; needs latest libsamsung-ipc Oct 29 09:36:09 freesmartphone.org: Signed-off-by: Simon Busch Oct 29 09:41:46 hi morphis Oct 29 09:52:21 mrmoku: the dbus fix is not enough.. Oct 29 09:52:34 mrmoku: ${D} is expanded in postinst ;/ Oct 29 09:52:52 hmm :/ Oct 29 10:24:47 mrmoku: updated patch should work now Oct 29 10:25:03 JaMaOff: ok... needs manual rebuild of dbus? Oct 29 10:25:08 yes Oct 29 10:25:13 good, thx Oct 29 10:25:54 weird.. if you use ${D} in postinst in recipe it's expanded.. so you have to be carefull and use only $D Oct 29 10:30:45 hehe Oct 29 10:32:22 GNUtoo: could you please remove distro/machine overrides from mplayer recipes and add patch headers for mplayer and fbreader? Oct 29 10:34:14 add patch headers? Oct 29 10:34:38 Upstream-Status etc Oct 29 10:36:12 that's a lot of work Oct 29 10:36:36 I'll first remove distro/machine overrides Oct 29 10:38:57 time for breakfast/lunch.. bbl Oct 29 10:39:05 there are too much patches for mplayer..... Oct 29 10:50:10 GNUtoo: well without it, it won't be accepted Oct 29 10:57:47 I know Oct 29 10:58:01 but maybe distributing the workload a bit could be done Oct 29 10:58:05 and then it could be accepted Oct 29 11:01:20 * JaMaOff spent 5,5 hours on OE today.. and didn't even eat.. so really bbl Oct 29 11:14:45 zub: I see... you got your elm -> elementary renaming in... breaks ffalarms :P Oct 29 12:05:28 mrmoku, yeah, we love breaking things :) Oct 29 12:05:41 mrmoku, which error did you get with ffalarms? Oct 29 12:07:32 morphis, hi Oct 29 12:07:36 I've serial on nexus S Oct 29 12:47:35 GNUtoo: more changes requested from your patches on oe-devel ML, please have a look Oct 29 12:48:22 ok Oct 29 12:52:11 ok I forgott to add the commit id, but I copied other commits messages..... Oct 29 13:07:42 pespin: elm.vapi not found error Oct 29 13:07:52 JaMaOff: possible that ffalarms needs SRCREV bumping? Oct 29 13:08:31 hmm.. Oct 29 13:09:50 JaMaOff: is Lukasz fine with us fixing stuff in his repo directly? Oct 29 13:10:02 | ffalarms.vala:1436.22-1436.24: error: Argument 1: Cannot pass value to reference or output parameter Oct 29 13:10:05 | lst.item_append(itc, a, null, Elm.GenlistItemFlags.NONE, null); Oct 29 13:10:08 this needs a ref Oct 29 13:14:57 heyho Oct 29 13:30:00 freesmartphone.org: 03angelo 07aurora * ra3ffb0290378 10/aurora-daemon/src/ (Makefile.am fbevents.cpp fbevents.h): aurora-daemon: add skeleton for FBEvents class Oct 29 13:40:39 SHR: 03mok 07meta-smartphone * r4c20b615aec9 10/meta-shr/recipes-shr/3rdparty/ffalarms_git.bb: ffalarms_git.bb: bump SRCREV for libeflvala changes Oct 29 13:53:23 GNUtoo: ping Oct 29 13:54:29 morphis, pong Oct 29 13:54:35 btw I've that: Oct 29 13:54:37 ExpansionError: Failure expanding variable SUMMARY, expression was ${SUMMARY} - Debugging files which triggered exception Exception: variable SUMMARY references itself! Oct 29 13:54:41 with task-x11 Oct 29 13:54:43 anyway Oct 29 13:54:46 morphis, I've serial Oct 29 13:55:02 the serial port is /dev/s3c2410_serial2 Oct 29 13:55:18 knowing that I've serial and that some drivers are mainline Oct 29 13:55:23 is it worth to go mainline? Oct 29 13:55:32 GNUtoo: great Oct 29 13:55:58 GNUtoo: I don't know how much stuff we need from the samsung kernel to get mainline kernel working on the device Oct 29 13:56:11 for instance the modem is not mainline Oct 29 13:56:19 GNUtoo: do we have wifi support in mainline kernel? Oct 29 13:56:38 however looking at http://wiki.freesmartphone.org/index.php/Hardware/Nexus_S/MainlineKernelStatus shows that it's possible to go mainline Oct 29 13:56:39 yes Oct 29 13:56:42 there is Oct 29 13:56:46 or asked the other way round: how much is it to get a minimal mainline-based kernel working on the device? Oct 29 13:56:54 ah minimal, very easy Oct 29 13:57:13 minimal means serial mux and board init Oct 29 13:57:18 however we have an issue Oct 29 13:57:24 I didn't tought about that before Oct 29 13:57:28 it's called device tree Oct 29 13:57:58 I must ask what's the mainline policy with reguards to device tree Oct 29 13:57:59 I already heard about it but never looked into the details Oct 29 13:58:06 to see if they still accept board files Oct 29 13:58:19 however we could have everything but the board file Oct 29 13:58:55 ok, I see two things: first we should not start our work with the aim to get all board stuff into mainline kernel Oct 29 13:58:56 basically it's a big issue knowing that: Oct 29 13:59:03 *we can only chainload bootloaders Oct 29 13:59:24 s/*// Oct 29 13:59:25 GNUtoo meant: we can only chainload bootloaders Oct 29 13:59:27 if we start with mainline work: we should start to get everything working and then look if time is left to get it mainline Oct 29 14:00:01 morphis, what do you mean? everything is not working because of the bad kernel Oct 29 14:00:22 yeah, a mean a minimal subset we need to use the device Oct 29 14:00:27 else we would had wifi and good alsa Oct 29 14:00:31 this means: modem, wifi, alsa Oct 29 14:00:36 and power Oct 29 14:00:42 I don't understand Oct 29 14:00:54 ok, let me start to explain Oct 29 14:00:59 ok Oct 29 14:01:12 we have a kernel which does support all features we want Oct 29 14:01:19 no Oct 29 14:01:26 no? Oct 29 14:01:28 alsa is bad and wifi too Oct 29 14:01:36 wifi is unusable with that kernel Oct 29 14:01:40 yeah, we don't want bad alsa and bad wifi Oct 29 14:01:41 that's why I want to go mainline Oct 29 14:01:46 yes Oct 29 14:01:54 that I meant with "all features we want" Oct 29 14:02:47 GNUtoo: so if we're going to mainline kernel Oct 29 14:03:29 we should just use the mainline kernel as base for our work and try to port everything need (which is not already in mainline kernel) to the new kernel (like board support and modem) Oct 29 14:03:40 s/need/needed/ Oct 29 14:03:42 morphis meant: we should just use the mainline kernel as base for our work and try to port everything needed (which is not already in mainline kernel) to the new kernel (like board support and modem) Oct 29 14:03:45 yes Oct 29 14:04:06 if we have done that and the kernel is usable and there is time left we can start working on getting our stuff into the mainline kernel Oct 29 14:05:04 hmmm Oct 29 14:05:11 then we'll have to rebase in order to push Oct 29 14:05:16 oops Oct 29 14:05:21 as long as we want to push Oct 29 14:05:21 then we'll have to rebase in order to send upstream Oct 29 14:05:39 yes but it shouldn't be our aim for the first time to get something upstream Oct 29 14:05:55 as we need to concentrate on our problems and not on problems with getting stuff upstream Oct 29 14:06:19 well it depend on what you want to do with nexus S Oct 29 14:06:35 if you want something quick and dirty that doesn't last.... Oct 29 14:06:47 mainlining is useless in that case Oct 29 14:06:51 I want a usable device in a reasonable time Oct 29 14:06:53 else if you want something that lasts.... Oct 29 14:07:08 you can mainline while doing the stuff Oct 29 14:07:18 it's just git-send-email and responding to some comments Oct 29 14:07:19 as otherwise we sit here with working on getting the device ready in a year Oct 29 14:07:21 it's not that hard Oct 29 14:07:26 ok Oct 29 14:08:05 because then you'll have some issues with an old kernel Oct 29 14:08:14 how much time do we need to get a basic mainline based kernel working? Oct 29 14:08:16 and the device will become not maintainable Oct 29 14:08:20 ok Oct 29 14:08:22 just serial Oct 29 14:08:27 board file with serial Oct 29 14:08:30 that's the minimum Oct 29 14:09:08 not maintainable means that you'll get compiler issues for compiling an old kernel, make will change, udev which we don't use won't work anymore on it, etc... Oct 29 14:09:29 and maybe some stuff like glib will depend on newer syscalls Oct 29 14:09:29 yes, you are right Oct 29 14:09:30 etc... Oct 29 14:09:44 so we should start with getting board file with serial working Oct 29 14:09:56 yes Oct 29 14:10:13 and I should find out with reguards to the device tree Oct 29 14:10:46 yes, I already started to read this http://devicetree.org/Device_Tree_Usage Oct 29 14:11:01 ok Oct 29 14:11:13 seems some nice but very new feature Oct 29 14:11:23 the device is 1GHz,512M of ram, 32bit framebuffer with a lot of stuff mainline Oct 29 14:11:30 so it's worth Oct 29 14:11:46 ok Oct 29 14:12:00 should we use linux-next or linux as base? Oct 29 14:12:42 not the linus tree Oct 29 14:12:47 or linux-next Oct 29 14:12:55 or the specific tree you aim for inclusion Oct 29 14:13:04 for instance wireless tree if you work on wireless Oct 29 14:13:05 etc... Oct 29 14:14:24 ok, but how do we structure our work? Oct 29 14:14:33 we need some common branch to build from Oct 29 14:14:46 yes Oct 29 14:14:56 we could build on linux-next Oct 29 14:14:56 which branch should be the base for this? Oct 29 14:15:00 ok Oct 29 14:15:07 and send the patches rebasing on various trees Oct 29 14:15:08 then I will setup one Oct 29 14:15:09 as usual Oct 29 14:15:13 yes Oct 29 14:15:18 I'll take care of that Oct 29 14:15:23 but we will setup our working tree based on linux-next? Oct 29 14:15:29 yes Oct 29 14:15:32 that's the idea Oct 29 14:15:45 basically kernel tree is split up with tons of branches Oct 29 14:15:48 because it works better Oct 29 14:16:04 GNUtoo: can you please setup a samsung-crespo/upstream branch in the fso linux-2.6.git repository? Oct 29 14:16:12 yeah I know Oct 29 14:16:18 I'll do when I'll have to push the first patch Oct 29 14:16:22 I just want to define our working space and structure Oct 29 14:16:29 GNUtoo: ok Oct 29 14:17:00 GNUtoo: should I get serial working too and finish modem call support working and then help you? Oct 29 14:17:15 finish modem first Oct 29 14:17:20 I can make serial work Oct 29 14:17:24 I already did it for bug1 Oct 29 14:17:29 but I had uboot tough Oct 29 14:17:33 so I could have dmesg Oct 29 14:18:05 1)finish to make serial work and findout how to get the kernel messages Oct 29 14:18:15 getting them trough uboot is impossible tough Oct 29 14:18:17 you didn't have serial working yet? Oct 29 14:18:18 so it'll be harder Oct 29 14:18:22 I have Oct 29 14:18:25 but look at that: Oct 29 14:18:26 but? Oct 29 14:18:41 Uncompressing Linux... done, booting the kernel. Oct 29 14:18:43 then nothing Oct 29 14:18:52 ah ok Oct 29 14:18:53 I must fix that Oct 29 14:18:59 maybe kernel commandline? Oct 29 14:19:00 trough rootfs and cmdline Oct 29 14:19:08 I started to do it Oct 29 14:19:13 but I had that error: Oct 29 14:19:32 ExpansionError: Failure expanding variable SUMMARY, expression was ${SUMMARY} - Debugging files which triggered exception Exception: variable SUMMARY references itself! Oct 29 14:23:56 hm Oct 29 14:24:02 what did you add to the command line? Oct 29 14:25:06 let me look Oct 29 14:25:45 console=s3c2410_serial2 Oct 29 14:25:57 since building an image didn't work I did it manually Oct 29 14:26:05 ok Oct 29 14:26:09 (added to /etc/securetty and /etc/inittab) Oct 29 14:26:41 you removed "no_console_suspend"? Oct 29 14:27:00 no Oct 29 14:28:30 GNUtoo: maybe paulk-laptop has it working? Oct 29 14:28:59 morphis, no, I stopped when I got the bootloader messages Oct 29 14:29:04 he didn't get that far Oct 29 14:29:10 he just wanted to dump the bootloader Oct 29 14:29:16 *bootloader messages Oct 29 14:29:31 ok Oct 29 14:30:28 console=s3c2410_serial2,115200 Oct 29 14:30:31 still nothing Oct 29 14:30:50 hm Oct 29 14:31:31 hmmm Oct 29 14:31:34 maybe I know why Oct 29 14:31:43 [ 0.102801] console [ram-1] enabled Oct 29 14:31:46 ramconsole Oct 29 14:31:48 etc... Oct 29 14:35:03 can we disable it? Oct 29 14:35:25 btw do you used a om debg board? Oct 29 14:35:34 yes Oct 29 14:35:37 I've 2 om debug boards Oct 29 14:35:43 one has a broken cable tough Oct 29 14:35:57 you must put dedicate one finger for holding the cable to make it work Oct 29 14:36:04 the other works flawlessly Oct 29 14:37:23 still nothing Oct 29 14:37:30 maybe it would work on mainline better Oct 29 14:39:27 maybe Oct 29 14:49:18 hm I have here a tiny micro usb connector I bought in the electronic shop Oct 29 14:49:28 it has five pins on the back Oct 29 14:49:42 yes Oct 29 14:49:46 three in a row on the top and two in a row on the bottom Oct 29 14:49:50 yes Oct 29 14:49:58 which pin is which? the datasheet is unclear in this point Oct 29 14:50:02 I used the same kind of micro usb Oct 29 14:50:06 you have colors Oct 29 14:50:09 no_console_suspend Oct 29 14:50:14 s/no_console_suspend// Oct 29 14:50:14 morphis meant: Oct 29 14:50:34 http://trac.osuosl.org/trac/replicant/wiki/NexusSBootloader Oct 29 14:50:34 GNUtoo: there are no colors Oct 29 14:50:38 ah ok Oct 29 14:50:43 just plain metal Oct 29 14:50:53 http://www.produktinfo.conrad.com/datenblaetter/725000-749999/733923-da-01-en-MICRO_USB_STECKER_TYP_B_5POL.pdf Oct 29 14:50:53 ok Oct 29 14:50:55 thats the datasheet Oct 29 14:50:59 I removed the colors tough Oct 29 14:51:03 I changed the cable Oct 29 14:51:06 for a network cable Oct 29 14:51:15 because there was one pin that was left unconnected Oct 29 14:51:25 so that would have been messy otherwise Oct 29 14:51:27 anyway Oct 29 14:51:32 I've noted the colors Oct 29 14:51:39 but you'll easily find the pinout Oct 29 14:51:49 s/noted/wrote down Oct 29 14:52:16 hm Oct 29 14:52:50 so I need to cut another usb cable just to get the color <-> pin relation Oct 29 14:52:54 maybe we should support only devices that can have mainline kernels? Oct 29 14:53:00 no Oct 29 14:53:06 you just need to search the web Oct 29 14:53:47 I can give you the colors but the way it is will be hard to explain Oct 29 14:53:52 like this: http://pinoutsguide.com/PortableDevices/micro_usb_pinout.shtml Oct 29 14:53:59 but it's only the front of the connector Oct 29 14:54:06 yes Oct 29 14:54:12 GNUtoo: about only mainline kernels: we shoudl discuss this at FSOSHRCON Oct 29 14:54:17 indeed Oct 29 14:54:22 because we could have: Oct 29 14:54:49 nokia900, om-gta02, crespo only Oct 29 14:55:15 nexusone has some mainline parts Oct 29 14:55:20 but it's not that worth Oct 29 14:55:36 hm Oct 29 14:55:43 GNUtoo: https://upload.wikimedia.org/wikipedia/commons/2/27/Micro-USB.jpg Oct 29 14:55:54 should the smaller numbers mean it's the bottom row? Oct 29 14:55:55 basically the goal is to have devices that once they are complete require minimal work Oct 29 14:56:16 morphis, no idea Oct 29 14:56:27 maybe ask in the #electronics channel? Oct 29 14:56:46 morphis, do you have a DMM? Oct 29 14:57:00 https://meego.gitorious.org/meego-device-adaptation/n900_kernel Oct 29 14:57:01 for n900 Oct 29 14:57:35 GNUtoo: DMM? Oct 29 14:57:42 digital multimeter Oct 29 14:57:45 no Oct 29 14:57:50 ahh too bad Oct 29 14:57:53 stop Oct 29 14:57:57 there are cheap ones Oct 29 14:58:06 my bad I have one Oct 29 14:58:08 because there is a beep mode on the DMM Oct 29 14:58:10 ok lol Oct 29 14:58:15 put it in beep mode Oct 29 14:58:25 it's the mode where when there is a conduction it beeps Oct 29 14:58:39 that way you'll find out what is connected to what Oct 29 15:02:15 yes Oct 29 15:08:12 where do I have to set the voltage I want to use? doesn't usb supply 5V as default? Oct 29 15:10:13 usb is always 5 V Oct 29 15:11:51 morphis, you need a voltage shifter or an already shifted usb<->serial adapter Oct 29 15:13:57 I have an om debug board Oct 29 15:14:34 ok Oct 29 15:14:39 maybe you can use that Oct 29 15:14:59 I used a sparkfun serial adapter tough Oct 29 15:16:26 ok Oct 29 15:27:20 maybe DocScrutinizer would know for the debugboard Oct 29 15:28:39 umm, wazzup? Oct 29 15:29:15 DocScrutinizer, what's the voltage of the serial pins of the debug board Oct 29 15:29:22 the part where there is written UART Oct 29 15:29:29 duh, LV I think Oct 29 15:29:51 sth between 1.8 and max 3.3 I guess Oct 29 15:30:47 ok Oct 29 15:31:06 DocScrutinizer, morphis is trying to make a serial cable for the nexus S Oct 29 15:31:09 check the debubo schematics, should tell a lot about it Oct 29 15:31:15 ok Oct 29 15:31:57 I tend to use some "data cable" for arbitrary phone for that purpose Oct 29 15:32:33 they come phone-LV<->RS232 and phone-LV<->USB flavour Oct 29 15:32:59 personally I got a sparkfun usb<->serial adapter so I used that Oct 29 15:33:08 usually cost less than 20 bucks Oct 29 15:33:10 it came with a bugduino.... Oct 29 15:36:49 DocScrutinizer: thanks for the comment Oct 29 15:36:57 so I will do that not today Oct 29 15:36:59 np Oct 29 15:37:12 always a pleasure Oct 29 15:37:21 ping me if you need any help on that Oct 29 15:37:39 ok, thanks Oct 29 16:03:58 GNUtoo: I will forward this and continue working on modem suport Oct 29 16:09:08 forward? Oct 29 16:22:28 morphis, check that http://wiki.freesmartphone.org/index.php/HardwareComparison#Developer_Status Oct 29 16:22:52 it's better not to come to FSOSHRCON without serious data to decide what to do Oct 29 16:23:02 so I did that and the people page Oct 29 16:23:10 the people page was done long ago tough Oct 29 16:27:26 GNUtoo: looks good Oct 29 16:27:55 should the "bad(special protocol) " stay there for the nexuss? Oct 29 16:28:04 as we have the protocol mostly reverse engineered? Oct 29 16:29:22 just tell it works then Oct 29 16:29:30 but it's bad from maintenance point of view Oct 29 16:29:35 for instance n900 was broken Oct 29 16:29:40 it doesn't work anymore Oct 29 16:29:47 and it's hard to dig inside and fix it Oct 29 16:31:04 maybe we should add a line under it for adding a status Oct 29 16:33:13 we should Oct 29 16:33:28 as bad indicates more or less that it is not working which is wrong Oct 29 16:33:44 it's just not documented for the public Oct 29 16:36:02 ok doing it Oct 29 16:39:42 ok Oct 29 16:49:28 ok done Oct 29 16:52:03 thanks Oct 29 16:53:19 wanna ad a col "N9" ? Oct 29 16:53:45 unfortunately I don't know anything about the n9 Oct 29 16:53:57 ask me, I know a bit Oct 29 16:54:14 N9 and N900 not that much different Oct 29 16:54:29 from an OS maintenance POV Oct 29 16:54:33 first: is there some security mecanism preventing you from playing with the device(load kernel, change rootfs etc....) Oct 29 16:54:42 no Oct 29 16:55:24 what is the network protocol of the modem? Oct 29 16:55:31 there's a security mechanism protecting the original OS Oct 29 16:55:37 ok Oct 29 16:55:44 same cellmo prot as on N900 Oct 29 16:55:47 basically Oct 29 16:55:55 the problem is then who has one Oct 29 16:56:06 and the n900 is not that well supported Oct 29 16:56:06 not me, yet Oct 29 16:56:08 ok Oct 29 16:56:31 maybe I should try the 3.0 kernel for the n900 Oct 29 16:56:45 but I guess we should fix the network first Oct 29 16:56:50 it doesn't register anymore Oct 29 16:56:53 mrmoku, ping Oct 29 16:56:59 GSM? Oct 29 16:57:02 yes Oct 29 16:57:05 :-/ Oct 29 16:58:38 I could look at it now Oct 29 16:58:43 but I don't know a lot about it Oct 29 16:58:48 beside the phonet-utils stuff Oct 29 17:05:02 where's RUI btw Oct 29 17:05:17 because it's really really annoying Oct 29 17:05:41 ah now it registers Oct 29 17:05:42 nice Oct 29 17:08:46 RUI!!!!!!! Oct 29 17:08:51 sigh Oct 29 17:11:14 I've also to fix what JaMaOff asked me to fix Oct 29 17:18:55 GNUtoo: pong Oct 29 17:20:10 mrmoku, sorry I had plane mode on for n900 Oct 29 17:20:17 never the less it still require some work Oct 29 17:20:24 also do you still have a working gta02? Oct 29 17:20:56 yeah, but not with the recent stuff Oct 29 17:21:15 around 2-3 weeks old Oct 29 17:21:15 ah so you didn't loose it with the gta04 transformation Oct 29 17:21:24 no,no Oct 29 17:21:34 I still have it untouched :) Oct 29 17:21:43 ok nice Oct 29 17:30:58 mrmoku, btw what are you working on? Oct 29 17:31:14 GNUtoo: waiting for the new SD I ordered :P Oct 29 17:31:26 GNUtoo: btw. what model is the one giving you problems? Oct 29 17:32:19 GNUtoo: other than that right now I'm updating my laptop with current efl stuff Oct 29 17:32:25 ok Oct 29 17:32:37 I've some microsd that are broken Oct 29 17:32:39 I want to start hacking on some telepathy client Oct 29 17:32:40 I don't know why tough Oct 29 17:32:42 ok Oct 29 17:32:50 I'm not sure it's a good idea Oct 29 17:32:56 I bought one SanDisk 4GB SDHC Oct 29 17:33:03 our devices support are not good at all Oct 29 17:33:09 and it does not work correctly... neither in n900, nor gta04 Oct 29 17:33:12 nor my laptop Oct 29 17:33:12 none of our phones are usable Oct 29 17:33:16 all give errors in dmesg Oct 29 17:33:35 so a telepathy client is quite useless Oct 29 17:33:43 nah, I need that for my laptop too Oct 29 17:33:44 let's applications developers like pespin_ do it Oct 29 17:33:59 for instance sound during call in the n900 is horrible Oct 29 17:34:13 writei is almost good but not there yet Oct 29 17:34:15 yeah, for that I need the new SD Oct 29 17:34:17 but readi..... Oct 29 17:34:33 ok Oct 29 17:34:52 hmm... I could try to use the NAND of the gta02 and use it's SD meanwhile... Oct 29 17:34:56 that might be an idea Oct 29 17:34:56 ok Oct 29 17:35:09 though dfu-util was not working correctly the last time a tried Oct 29 17:35:13 that's quite some time ago Oct 29 17:35:18 and there is active development going on Oct 29 17:35:22 it works fine Oct 29 17:35:22 so it might have changed Oct 29 17:35:27 good Oct 29 17:35:28 you just need to do that first: Oct 29 17:35:30 nand scrub Oct 29 17:35:33 oops Oct 29 17:35:35 nand bad Oct 29 17:35:36 nand scrub Oct 29 17:35:40 nand createbbt Oct 29 17:35:42 dynpart Oct 29 17:35:45 and then reflash Oct 29 17:35:51 where should I do that? Oct 29 17:36:51 in uboot Oct 29 17:37:03 hmm... the one in NOR works? Oct 29 17:37:16 because I'm using Qi Oct 29 17:37:32 yes Oct 29 17:37:36 use the one in NOR Oct 29 17:37:39 I use QI too Oct 29 17:37:45 ok Oct 29 17:38:05 and then connect to u-boot via usb? Oct 29 17:38:11 * mrmoku never did that on gta02 Oct 29 17:38:23 yes Oct 29 17:38:30 picocom -b 115200 /dev/ttyACM0 Oct 29 17:39:17 ahh, works :) Oct 29 17:39:41 hmm nand scrub does not Oct 29 17:40:34 GNUtoo: the message nand scrub gives is quite scary :-P Oct 29 17:40:58 what does it give? Oct 29 17:41:10 NAND scrub: device 0 whole chip Oct 29 17:41:10 Warning: scrub option will erase all factory set bad blocks! There is no reliable way to recover them. Use this command only for testing purposes if you are sure of what you are doing! Oct 29 17:41:21 yes that's normal Oct 29 17:41:28 ask DocScrutinizer if you're unsure Oct 29 17:41:31 ok Oct 29 17:41:37 but basically it erases nand Oct 29 17:41:40 the whole nand Oct 29 17:41:44 but you still have NOR Oct 29 17:42:08 also it'll recreate the bbt Oct 29 17:42:12 with createbbt Oct 29 17:43:16 hmm... its normal it does not show any progress after pressing y ? Oct 29 17:44:06 no Oct 29 17:44:27 there is a lag tough Oct 29 17:44:37 more than a min? Oct 29 17:44:56 I press y Oct 29 17:44:59 and nothing happens Oct 29 17:45:10 pressing any other key gives me scrub aborted Oct 29 17:45:14 * mrmoku will be patient Oct 29 17:45:27 restart from scratch Oct 29 17:45:30 nand bad Oct 29 17:45:35 nand scrub Oct 29 17:45:38 nand createbbt Oct 29 17:45:41 dynpart Oct 29 17:45:58 then exit picocom Oct 29 17:46:02 and flash with dfu-util Oct 29 17:46:36 GNUtoo: gah it turned off Oct 29 17:47:22 brb Oct 29 17:48:44 GNUtoo: you already had success for serial on nexuss? Oct 29 17:50:44 morphis, yes serial works in hardware, but I'm working on n900 right now Oct 29 17:50:57 I means I've still to activate it in software Oct 29 17:52:19 mrmoku: yes it's probably not AUTOREV (ffalarms) Oct 29 17:52:32 and I guess that update for newer API is fine to be pushed by us Oct 29 17:55:53 GNUtoo: ok Oct 29 17:56:08 JaMaOff: dbus activation seems broken with latest dbus here Oct 29 17:56:38 JaMaOff: look at this: Oct 29 17:56:39 MDBUS2> org.freesmartphone.ogsmd Oct 29 17:56:39 [ERR]: Failed to execute program /usr/libexec/dbus-daemon-launch-helper: Success Oct 29 17:57:58 did you upgrade dbus today? Oct 29 17:58:08 you need -r2 Oct 29 17:58:30 morphis, I think JaMaOff has a fix Oct 29 17:58:38 JaMaOff: a "chmod +x /usr/libexec/dbus-daemon-launch-helper" fixes this Oct 29 17:59:09 I have: 1.4.12-r2 Oct 29 18:03:30 GNUtoo: is there some way to stop autosuspend? Oct 29 18:03:55 GNUtoo: ahh... y ... does the trick :P Oct 29 18:04:42 GNUtoo: ok, all done successfully Oct 29 18:05:02 morphis: wrong fix :-P Oct 29 18:05:15 mrmoku: my one or JaMa's one? Oct 29 18:05:15 morphis: it needs messagesbus as group owner Oct 29 18:05:23 morphis: yours Oct 29 18:05:26 ok Oct 29 18:05:42 makes your running everything as root phone unsafe :-P Oct 29 18:05:50 :) Oct 29 18:06:01 * mrmoku off to eat some maroni before dealing with dfu-util :) Oct 29 18:18:24 morphis: I've updated -r2 once today, maybe you were unlucky to pull it with older (but still -r2 version) Oct 29 18:18:39 morphis: and manual clean + reinstall should fix it Oct 29 18:19:06 JaMaOff: I am on the way doing this :) Oct 29 18:43:28 there is a lot of strange things in n900 fsoaudiod plugin Oct 29 18:44:04 count=320, avail=0 is what I've in the logs sometimes(not all the time), there is a lot of them Oct 29 18:44:24 then it does if (avail < count) Oct 29 18:44:30 and throw an error Oct 29 18:44:59 ah ok the error is in the log Oct 29 18:45:01 ok Oct 29 18:45:04 not to stderr Oct 29 19:13:50 mrmoku, did you succeed? Oct 29 19:17:02 sigh Oct 29 19:17:11 I broke writei locally Oct 29 19:17:20 I mean readi Oct 29 19:24:46 I'm too tired and I'm nervous again Oct 29 19:32:44 mrmoku, you are welcome to collaborate in etalk if you want :) I know there are still some ugly things in the code though, which I'd like to remove or improve but I have not much time right now :S Oct 29 19:34:58 but at least minimal working code with conversations and user list is there Oct 29 19:37:01 mrmoku, piing Oct 29 19:37:28 pespin_, could you take a look at rui's shr_elm_softkey? Oct 29 19:37:31 and fix it? Oct 29 19:37:54 GNUtoo, I'm afraid I have no knowledge on that, I only know elementary atm ;) Oct 29 19:37:56 because it's unbereable Oct 29 19:38:02 ah? Oct 29 19:38:08 in what is it written? Oct 29 19:38:15 GNUtoo: back Oct 29 19:38:27 I mean, I didn't play with all that app/windows switch messing and I have no idea on how does it work Oct 29 19:38:33 ok Oct 29 19:38:43 and I have enough work to do. Oct 29 19:38:51 GNUtoo, better ping Rui when he appears hehe Oct 29 19:38:52 * mrmoku downloading current dfu-util and qi Oct 29 19:38:59 but he doesn't appear Oct 29 19:39:04 that's the problem Oct 29 19:39:16 I could get nervous at any moment because of that Oct 29 19:39:24 for instance if it break during a call Oct 29 19:39:35 GNUtoo, what about begging to captainigloo for a hack to use elfe window switch without compositing? :P Oct 29 19:39:51 it's for gta02 Oct 29 19:40:04 I've to kill and re-launch shr-elm-softkey all the time Oct 29 19:40:12 yeah I know the problem Oct 29 19:40:18 during the day I don't count anymore the: Oct 29 19:40:24 killall -9 shr_elm_softkey Oct 29 19:40:31 shr_elm_softkey & Oct 29 19:40:34 lol Oct 29 19:40:50 I hope it's fixed soon too Oct 29 19:40:52 at the end I'll abandon using and hacking on fso+shr if that's not fixed Oct 29 19:41:03 because it's unbearable Oct 29 19:41:31 sometimes it even breaks right after launching it Oct 29 19:42:16 GNUtoo, I'll try looking at the code but I don't promise anything Oct 29 19:42:22 ok thanks a lot Oct 29 19:42:30 I could try to help you also Oct 29 19:42:33 where's the code? Oct 29 19:42:33 anyway, I don't have an OE build env due to the libc problem Oct 29 19:42:38 no idea Oct 29 19:42:41 what libc issue? Oct 29 19:43:07 git and wget throw a libc_nsquery assertion when using shr-chroot Oct 29 19:43:27 isn't that an already known problem that morphis had with debian lenny? Oct 29 19:43:43 really? Oct 29 19:43:54 it's a question Oct 29 19:44:00 no idea u.u Oct 29 19:44:14 then ask JaMaOff Oct 29 19:44:17 he might know Oct 29 19:44:21 I told him Oct 29 19:44:31 but he can't reproduce Oct 29 19:44:52 ok Oct 29 19:44:58 what about building without the chroot? Oct 29 19:45:46 pff I'd prefer not to, I don't want to get own building problems which others can reproduce (which is funny because that's happening with the chroot) Oct 29 19:45:53 GNUtoo: hmm... dfu-util -l does not show me the partitions... that is bad? Oct 29 19:46:02 no Oct 29 19:46:17 ahh, because in the om wiki list shows them Oct 29 19:46:20 picocom -b 115200 /dev/ttySAC0 Oct 29 19:46:24 dynpart Oct 29 19:46:29 and wait like 5s Oct 29 19:46:30 and retry Oct 29 19:47:11 ok flashing qi worked :) Oct 29 19:47:21 yeah, I did the dynpart Oct 29 19:47:25 ok Oct 29 19:48:19 pespin_, it's part of shr-e-gadgets Oct 29 19:48:22 * mrmoku downloads kernel and rootfs Oct 29 19:49:23 pespin_: did you take a look at /etc/nsswitch.conf? Oct 29 19:50:23 pespin_: you could try to add '78.40.125.6 build.shr-project.org' to your hosts Oct 29 19:50:43 mrmoku, I'll try Oct 29 19:53:48 mrmoku, I don't know what's that /etc/nsswitch.conf for Oct 29 19:54:12 mrmoku: it breaks everything that uses elementary, but it was wrong :-/ Oct 29 19:54:41 and fix = --pkg elementary instead of --pkg elm Oct 29 19:54:55 pespin_: to tell the order in which way for example host names are resolved Oct 29 19:54:56 zub, fortunately, only like 5 projects are using valaèlementary, so no much problem :P Oct 29 19:55:25 and there are some more breakages on the horizon, as elementary did some renaming and deprecation -> eventually libeflvala should reflect it too Oct 29 19:55:44 ah ok,knowing that the file has some sense hehe Oct 29 19:56:38 mrmoku: I'm also thinking of throwing in some generics, to decrease the amount of ugly void* in the vapi Oct 29 19:56:47 * mrmoku forgot how slow flashing a rootfs with dfu-util is :/ Oct 29 19:56:51 but till ~wed I'm basically n/a Oct 29 19:57:05 zub: feel free to throw whatever you want :-) Oct 29 19:57:45 pespin_: other question would be if you can clone simple projects from the shr git Oct 29 19:57:54 or does that fail too? Oct 29 19:58:52 bbiab Oct 29 20:00:11 mrmoku, http://paste.pocoo.org/show/500151/ Oct 29 20:03:55 pespin_: hmm... I thought it is the initial cloning of the chroot that creates you problems? Oct 29 20:04:13 mrmoku, no, the weird thing is taht it happens INSIDE the chroot Oct 29 20:04:26 but you or JaMa don't get this error :S Oct 29 20:04:39 heh Oct 29 20:06:48 the only semirelated weird thing on my system is that I'm using glibc-2.14-5 instead of glibc-2.14-6 which come with arch because last one has dns resolution problems here Oct 29 20:08:18 and our chroot has? Oct 29 20:09:06 2.14 Oct 29 20:09:14 with some gentoo patches Oct 29 20:10:12 hmm Oct 29 20:10:24 https://bugs.archlinux.org/task/26007?dev=2008 Oct 29 20:10:42 but that's in my host, not inside chroot Oct 29 20:10:54 pespin_: try to change nameservers Oct 29 20:11:03 (in /etc/resolv.conf) Oct 29 20:11:11 in chroot? Oct 29 20:11:12 https://bugzilla.redhat.com/show_bug.cgi?id=730856 Oct 29 20:11:14 pespin_: yup Oct 29 20:11:18 ok I'll try Oct 29 20:11:25 I have to go now though, pub is calling me ;) Oct 29 20:11:48 python-2.7 seems fixed... finally :) Oct 29 20:12:35 pespin_: have fun ;) Oct 29 20:13:06 hp != hp2 <- that was the assertion failing here as showed in the redhat bugzilla Oct 29 20:14:54 pespin_: yup... that's what I googled for :) Oct 29 20:15:05 GNUtoo: yay, seems to work... booting without sd Oct 29 20:16:41 ok Oct 29 20:25:12 GNUtoo: hmm... last image from buildhost does not look good though Oct 29 20:25:23 GNUtoo: I get your usb0 here and immediately gone Oct 29 20:25:44 and display turns black when the e desktop should popup Oct 29 20:26:12 GNUtoo: what do you want me to look at first... gta02 or n900? Oct 29 20:26:46 mrmoku, it's normal Oct 29 20:26:55 it's because udev doesn't start Oct 29 20:26:58 oops Oct 29 20:27:04 I meant fsousaged Oct 29 20:27:05 I hope so ;) Oct 29 20:27:06 with dbus Oct 29 20:27:07 ahh Oct 29 20:27:08 ok Oct 29 20:27:18 n900 audio would be nice Oct 29 20:27:20 hmm... with an SD that is easy to fix Oct 29 20:27:28 but with NAND not so easy :P Oct 29 20:27:37 since you have the shell on Xorg it is possible to fix Oct 29 20:27:51 shell on Xorg? Oct 29 20:27:57 no Xorg at all Oct 29 20:28:00 just a black screen Oct 29 20:28:54 ok, will build an image here then and take a look at n900 meanwhile Oct 29 20:33:09 mrmoku, ok Oct 29 20:54:20 ou guys having real halloween fun here, eh? :-D Oct 29 20:54:36 ;) Oct 29 20:54:37 nand scrub, wow :-P Oct 29 20:55:05 apart from the u-boot partition being 'unnamed partition' all seems to work :P Oct 29 20:55:29 well, mrdparts seems to miss some parameter or sth? Oct 29 20:55:51 no idea... and I don't care as long as it works :) Oct 29 20:56:04 errr dynpart Oct 29 20:56:39 well, the partitioning of nand is a bit of black magic even to me Oct 29 20:57:54 anyway for the furure I'd suggest to learn how to get the list of bad blocks out of NAND prior to scrub Oct 29 20:58:32 yes nand bad Oct 29 20:58:51 I did nand bad Oct 29 20:58:56 scrub is like HDD lowlevel formating Oct 29 21:00:02 you might want to keep the bad blocks that acted up in the past, not mark all as good and then decide which are bad on a short write test Oct 29 21:01:01 hmm Oct 29 21:01:06 too late :/ Oct 29 21:01:10 though nothing really bad will happen, except some blocks may give error on read and will get marked as bad on any future reflash Oct 29 21:01:22 ok, not too bad then Oct 29 21:01:44 of course "give error on read" means your system blows chunks and reflash pending Oct 29 21:01:46 nand is a temporary solution anyway until the new sd arrives Oct 29 21:02:55 generally you want to keep the bad guys in prison, as long as those bad guys don't trigger feelings of humanity Oct 29 21:03:25 I.E. one time bad block, ever a bad block Oct 29 21:03:40 :) Oct 29 21:17:39 NO MERCY (ZZ Top) Oct 29 21:25:35 hmmm I'm looking at shr_elm_softkey Oct 29 21:41:25 GNUtoo: ok, n900 boots and even registers Oct 29 21:41:33 has some timing problem on boot Oct 29 21:41:43 and does not bring the pin dialog Oct 29 21:42:05 the usual stop phonefsod; kill fsogsmd; start phonefsod fixes that Oct 29 21:42:12 that's the first thing I will look into Oct 29 21:42:29 tomorrow though... now my body demands some sleep :) Oct 29 21:42:40 gnight all Oct 29 21:43:21 mrmoku, ok Oct 29 21:43:37 good night Oct 29 21:43:49 I think the most important part is the sound quality during calls Oct 29 21:43:53 and that shr_elm_softkey Oct 29 21:54:00 I found something Oct 29 21:54:17 gdbserver<->gdb shr_elm_softkey Oct 29 21:54:25 ctrl + c at some point Oct 29 21:54:31 click on the top bar Oct 29 21:54:37 it reproduces the crash Oct 29 21:54:41 then c Oct 29 21:54:45 and it continue Oct 29 21:54:49 re-click Oct 29 21:54:59 and the widget appears as normal Oct 29 21:55:02 so as I tought Oct 29 21:55:16 s/so as I tought// Oct 29 21:55:16 GNUtoo meant: Oct 29 21:55:29 combine that with the fact that it uses 100% CPU.... Oct 29 22:03:13 JaMaOff, still there? **** ENDING LOGGING AT Sun Oct 30 02:59:56 2011