**** BEGIN LOGGING AT Sat Nov 05 02:59:57 2011 Nov 05 07:54:18 i m working on ARM board ...... I want to integrate external eeprom AT24C04C ... through userspace.... I detected SDA, SCL port with no external peripherals.... so I have to use that slot .... can anyone help... Nov 05 11:56:15 good morning all Nov 05 12:16:17 heyho Nov 05 12:23:10 morphis: heyho Nov 05 12:26:16 JaMa|Wrk: you got my last mail? Nov 05 12:26:40 angelox|laptop: ping Nov 05 12:30:03 SHR: 03morphis 07meta-smartphone * r10822ffa50c0 10/meta-fso/recipes-freesmartphone/freesmartphone/libfso-qt_git.bb: meta-fso: libfso-qt: only depend on native version of qfsodbusxml2cpp utility Nov 05 12:30:14 SHR: 03morphis 07meta-smartphone * rf6d700acb497 10/meta-fso/recipes-freesmartphone/freesmartphone/ (qfsodbusxml2cpp-native_git.bb qfsodbusxml2cpp_git.bb): meta-fso: qfsodbusxml2cpp: we only have and need a native version of this utility Nov 05 12:30:32 mrmoku: yup trying now Nov 05 12:30:37 morphis: ^^ Nov 05 12:39:57 morphis: still cannot find qmake Nov 05 12:46:13 morphis: hey! Nov 05 12:47:38 JaMa|Wrk: hm Nov 05 12:53:07 morphis: saw your aurora work! fantastic (though i got confused in directories renames) :) Nov 05 12:57:24 morphis, v2.6.35 stock fails to work Nov 05 13:02:51 GNUtoo: thats sad Nov 05 13:03:01 angelox|laptop: yeah I changed a lot Nov 05 13:04:00 adding .fixup doesn't change a thing Nov 05 13:04:27 ah wait a sec Nov 05 13:04:31 fixup is wrong for 35 Nov 05 13:07:45 still nothing Nov 05 13:07:46 sigh Nov 05 13:08:02 I'll look in git log Nov 05 13:08:54 HERRING: Makefile: Change zreload address configuration. Nov 05 13:08:56 what's that? Nov 05 13:09:15 ahhh Nov 05 13:09:29 in arch/arm/mach-s5pv210/Makefile.boot: Nov 05 13:09:34 - zreladdr-y := 0x20008000 Nov 05 13:09:49 +ifdef CONFIG_MACH_S5PC110_CRESPO Nov 05 13:09:49 +zreladdr-y := 0x30008000 Nov 05 13:09:49 +params_phys-y := 0x30000100 Nov 05 13:09:57 that's my problem Nov 05 13:10:16 ok Nov 05 13:10:23 I already about something like this Nov 05 13:10:29 all address are changed Nov 05 13:10:35 so it's the cause Nov 05 13:10:36 and is it booting? Nov 05 13:10:41 wait a sec Nov 05 13:10:45 I need to import the patch Nov 05 13:10:57 but it would be the only explanation Nov 05 13:11:07 yes Nov 05 13:11:24 I'll try to import on master Nov 05 13:12:28 ok Nov 05 13:20:31 I'll eat Nov 05 13:21:31 ok I have to reboot Nov 05 14:03:12 moo Nov 05 14:14:08 JaMa|Wrk: ok, I have the issue now too Nov 05 14:15:22 JaMa|Wrk: I will try to fix it Nov 05 16:32:15 GNUtoo: ping Nov 05 16:33:01 morphis, pong Nov 05 16:36:07 GNUtoo: you got the kernel working? Nov 05 16:37:03 no Nov 05 16:38:46 GNUtoo: but you tried with the changes you mentioned today? Nov 05 16:39:01 yes Nov 05 16:39:09 I'll read more logs Nov 05 16:39:25 http://www.pastie.org/2815585 Nov 05 16:39:33 maybe the machine is not herring Nov 05 16:39:36 that's my next guess Nov 05 16:39:41 maybe it's smdkc110 Nov 05 16:39:53 because at the beginning it was smdkc110 Nov 05 16:39:54 not herring? Nov 05 16:40:06 so maybe the bootloader expects another MACH_ID Nov 05 16:40:39 ah ok Nov 05 16:40:48 that can be a problem, yes Nov 05 16:40:54 after that guess I'll have to read a ton of logs Nov 05 16:41:01 (if that guess fails) Nov 05 16:41:48 and at the beginning it was CONFIG_CRESPO Nov 05 16:42:28 I'll try it now if you want Nov 05 16:43:51 only if you have time Nov 05 16:44:13 still the same Nov 05 16:44:21 I start to desesperate Nov 05 16:46:25 don't give up Nov 05 16:46:46 all this kind of tasks needs time Nov 05 16:46:47 and more than one try Nov 05 16:46:47 it will work Nov 05 16:46:57 there some little bit wrong and we need to find it in the haystack Nov 05 16:47:07 can you push your efforts on some branch on fso.git? Nov 05 16:47:25 yes Nov 05 16:47:28 I'll do that Nov 05 16:47:34 but expect rebases Nov 05 16:47:39 no problem Nov 05 16:47:49 just I can follow your work easily Nov 05 16:47:54 ok Nov 05 16:47:59 as it's better than to do deal with patches Nov 05 16:48:02 else I can keep you informed Nov 05 16:48:07 no problem Nov 05 16:48:21 as long as I don't work on that task it's not a problem for me Nov 05 16:48:32 ok Nov 05 16:48:48 note that with the perfect hardware I already succedded at doing that Nov 05 16:48:51 see that patch: Nov 05 16:49:01 GNUtoo: if I have some time tomorrow I will review your changes Nov 05 16:49:07 http://git.kernel.org/?p=linux/kernel/git/next/linux-next.git;a=blob;f=arch/arm/mach-mx3/mach-bug.c;h=d137d7078ee90c8a94f843910e4e7151376ec325;hb=ea7ee4cae4db6630b388f39c3abe4f86f67211f9 Nov 05 16:49:09 s/patch/file Nov 05 16:49:24 that's from me and it's in mainline Nov 05 16:49:32 yeah thats kind of smart :) Nov 05 16:49:52 the problem here is the bootloader that get in my way Nov 05 16:49:59 maybe I should try to port uboot? Nov 05 16:50:12 like a quick and dirty chainloaded uboot? Nov 05 16:50:25 so I would have md to debug Nov 05 16:50:43 but I guess I'll just have the exact same issue with uboot porting Nov 05 16:51:55 I think so Nov 05 16:51:59 stay with the kernel Nov 05 16:53:20 I detest beeing blind Nov 05 16:53:41 what should I do now? JTAG and brick the device while soldering the jtag behind the camera? Nov 05 16:54:03 maybe I should continue working on it as usual Nov 05 16:54:08 try to findout what's wrong Nov 05 16:54:12 from the first commit Nov 05 16:54:16 in the working tree Nov 05 16:54:24 and read tons of commits Nov 05 16:54:38 morphis, in the meantime the n900 takes dust Nov 05 16:55:18 and I already know the response....we'll discuss that in the fso+shr meeting in germany Nov 05 16:55:21 but that's in december Nov 05 16:55:22 GNUtoo: yes, do it that way its more safe and I don't want you to loose your device Nov 05 16:55:31 it takes dust? Nov 05 16:55:42 it's a french expression sorry Nov 05 16:55:49 I mean it's not worked on Nov 05 16:55:52 no one works on n900 Nov 05 16:55:54 oh Nov 05 16:55:56 why not? Nov 05 16:56:01 ah ok Nov 05 16:56:17 because the only person wanting to work on it is busy with porting a new kernel to the nexus S Nov 05 16:56:18 as JaMa|Wrk is working on OE and mrmoku on gta04? Nov 05 16:56:23 :) Nov 05 16:56:38 yes but it has huge issues with alsa Nov 05 16:56:47 not the kind you're used too tough Nov 05 16:56:49 GNUtoo: yeah that will be one point we need to discuss in december Nov 05 16:56:52 easier ones Nov 05 16:56:55 ok Nov 05 16:57:10 basically me or mrmoku need to fix the userspace program that plays audio from the modem Nov 05 16:59:26 morphis, the problem is : what do I do between now and december? Nov 05 16:59:39 I guess you want me to work on the nexus S Nov 05 17:00:07 yes Nov 05 17:00:21 I want the nexus s in a really good shape until december Nov 05 17:00:31 for it means: getting aurora working Nov 05 17:00:46 with support for call and pdp Nov 05 17:01:01 so you can use aurora on the nexus to make calls and browse the web Nov 05 17:01:17 ok Nov 05 17:01:37 in the other hand the n900 might be older and has a gps attached to the modem but....: Nov 05 17:01:48 *good kernel, a lot of it is mainline Nov 05 17:01:53 *modem will hit mainline soon Nov 05 17:01:56 *standard kernel Nov 05 17:02:00 *known modem Nov 05 17:02:13 we have ofono+headers Nov 05 17:02:22 *known gps but not implemented Nov 05 17:02:47 it just lack some people working on it Nov 05 17:08:01 hm Nov 05 17:08:17 hard questions but still part of the problem we have as community Nov 05 17:08:26 and about the problem we want to talk in december Nov 05 17:08:34 whats our focus? Nov 05 17:08:53 which devices we want to target? should everyone target the device he likes most? Nov 05 17:10:04 I made a lot of tables in the wiki page Nov 05 17:10:08 it will help decide that Nov 05 17:10:13 who is a dev Nov 05 17:10:19 what skill the devs have Nov 05 17:10:24 what device they have Nov 05 17:10:45 and for the device : how hard is it to maintain it Nov 05 17:10:49 what's problematic Nov 05 17:10:51 etc... Nov 05 17:11:05 because if for instance many people have one device Nov 05 17:11:15 it's good because each dev have different skills Nov 05 17:11:25 for instance JaMa is good in openembedded Nov 05 17:11:28 you're good in modem Nov 05 17:11:35 some people are good in device integration Nov 05 17:11:38 and we need all parts Nov 05 17:11:41 else we loose time Nov 05 17:11:48 and do things badly Nov 05 17:12:14 for instance you will make errors in openembedded and fail to know how to integrate foo in SHR Nov 05 17:12:27 maybe that's also why you started aurora Nov 05 17:12:48 and I would not know for instance how to reverse a modem that is not AT Nov 05 17:17:23 mrmoku, hi Nov 05 17:17:29 mrmoku, what are you working on? Nov 05 17:18:23 morphis, ^^^ Nov 05 17:18:26 GNUtoo: yeah I aw that and it's awesome Nov 05 17:18:42 and a good base for a community decsision we should make in december Nov 05 17:18:49 ok Nov 05 17:18:58 I don't think we should decide now Nov 05 17:19:16 ok I just wanted to sub-decide now on n900 vs nexus S Nov 05 17:19:18 we have to prepare for the future and think what we want and what we can archive with a relative amount of work Nov 05 17:19:26 I tell you: work on nexus S Nov 05 17:19:38 but it's my personal opinion Nov 05 17:19:42 yes but what are the good arguments to make me work on it? Nov 05 17:19:48 it's an android device Nov 05 17:20:06 there are two people working on it rather than one Nov 05 17:20:26 I know that if I stop working on it it'll fail Nov 05 17:20:38 so maybe the solution is that mrmoku fix that audio thing on n900 Nov 05 17:21:08 but the work to do on nexus S is not fun at all Nov 05 17:21:12 it's rather very frustrating Nov 05 17:21:14 would be an option Nov 05 17:21:14 (for me) Nov 05 17:21:22 so it's risky Nov 05 17:22:11 and the kernel work may be very long once it boots Nov 05 17:22:14 for example we need the boot image too Nov 05 17:22:19 ok Nov 05 17:22:21 that's easy Nov 05 17:22:37 and that would be something I would like to see you do it Nov 05 17:23:01 I've just to disable some stuff in the kernel to findout why it reboots at some point in dmesg when using the ttySAC2 Nov 05 17:25:04 morphis, basically it would be to package dcordes initramfs script and depend on it and mkbootimg the whole thing(like it's done with dream+nexus one) Nov 05 17:35:24 GNUtoo: yes Nov 05 17:35:27 I have to leave Nov 05 17:35:29 bye Nov 05 18:14:41 GNUtoo: I got my new SD and preparing it for the gta02 now Nov 05 18:14:57 ah ok nice Nov 05 18:15:29 GNUtoo: I need a working daily phone to be able to work on the n900 Nov 05 18:15:36 ok Nov 05 18:15:44 GNUtoo: btw. Slyon will work on gta04 Nov 05 18:15:51 nice Nov 05 18:15:54 and he said SMS work out of the box :) Nov 05 18:16:16 biggest problem will be alsa there too :/ Nov 05 18:16:55 ah? Nov 05 18:16:57 why? Nov 05 18:17:54 multiple sound cards Nov 05 18:17:59 nothing handles that yet Nov 05 18:18:05 really? Nov 05 18:18:14 what about multi component in ASOC Nov 05 18:18:23 ? Nov 05 18:18:53 at some point they decided to allow multiple CODECS or components in alsa SOC Nov 05 18:19:02 let me find the commit Nov 05 18:19:25 yeah, but they (goldelico) went for multiple cards Nov 05 18:20:37 f0fba2ad1b6b53d5360125c41953b7afcd6deff0 Nov 05 18:20:43 ASoC: multi-component - ASoC Multi-Component Support Nov 05 18:22:02 I don't know how it's exposed in userspace tough Nov 05 18:22:13 I know they have modem + normal card Nov 05 18:22:42 altough it's easy Nov 05 18:22:46 arecord -Dhw:1 Nov 05 18:23:13 yeah... it just needs integration in fso world Nov 05 18:23:29 yes maybe trough the same kind of thing than the n900 Nov 05 18:23:41 so that's also a good reason to fix n900 audio Nov 05 18:23:45 during calls Nov 05 18:24:04 readi sound is clearly unacceptable currently Nov 05 18:24:16 ok Nov 05 18:24:16 someone could accept writei if he's very tolerent Nov 05 18:24:18 but not readi Nov 05 18:26:48 * mrmoku booting gta02 Nov 05 18:34:24 hmm... e segfaults :/ Nov 05 19:01:38 JaMa|Wrk: hmm... libXau is missing and the input config module of e does not load due to that Nov 05 19:05:02 JaMa|Wrk: strange... it's *not* missing... just e thinks so Nov 05 19:05:57
libXau.so*6: cannot open shared object file: No such file or directory
Nov 05 20:06:52 238 ? Ss 0:00 /usr/sbin/fso-gpsd -S localhost gpsd -P /var/run/fso-gpsd.pid Nov 05 20:06:55 243 ? S wrong too Nov 05 20:19:04 ??? Nov 05 20:23:05 GNUtoo-N900: having both (fso-gpsd + gpsd) installed and started is wrong Nov 05 20:29:32 ok Nov 05 20:30:02 anyway gps ismt implemented yet Nov 05 20:30:11 on gta02? Nov 05 20:30:18 on n900 Nov 05 20:30:31 yeah, but that is gta02 Nov 05 20:30:41 ok Nov 05 20:30:55 you need gps i guess then Nov 05 20:32:06 no, no Nov 05 20:32:16 just trying latest image Nov 05 20:32:22 to get a daily phone Nov 05 20:32:31 ok Nov 05 20:34:04 hello, can anybody give me a hint where "free_smartphone_usage_shutdown" is declared ? Nov 05 20:34:30 nschle85: fsousaged Nov 05 20:34:37 ahh Nov 05 20:34:42 now I got it Nov 05 20:35:01 nschle85: which bindings? Nov 05 20:35:28 i dont know i think gtk Nov 05 20:35:43 but i need the implementation Nov 05 20:36:16 free_smartphone_usage_shutdown is a dbus wrapper calling a method of fsousaged Nov 05 20:37:24 so if you want to know what actually happens you have to look at fsousaged Nov 05 20:37:39 brb Nov 05 20:38:01 ok thanks Nov 05 20:50:05 nschle85: trying to fix the 'damn thing does not shut down' but? :-) Nov 05 20:50:31 yes, but understanding fso is hard Nov 05 20:50:39 a lot of components ... Nov 05 20:51:28 the keys components are fsodeviced and fsousaged Nov 05 20:51:35 bye Nov 05 20:52:02 but where is the dbus wrapper code ? Nov 05 20:53:55 nschle85: guess I know what the problem is Nov 05 20:53:57 Posix.system( "shutdown -h now" ); Nov 05 20:54:12 PATH not being correct for fsousaged Nov 05 20:55:03 it needs absolute pathes ? Nov 05 20:55:26 we better fix that PATH problem Nov 05 20:55:43 that probably hits us in other places as well Nov 05 20:56:00 like it already did for ifconfig in some place Nov 05 20:56:29 but the command seems not to work 100 % Nov 05 20:56:48 the screen backlight is switched iff Nov 05 20:57:05 off but i can see that the phone stands in an open console Nov 05 20:57:21 nschle85: http://git.freesmartphone.org/?p=cornucopia.git;a=blob;f=fsousaged/src/plugins/dbus_service/commands.vala;h=7a47e2c5d5b702a3a8cf94267be634a57d3373a7;hb=HEAD#l224 Nov 05 20:57:46 it first disables all resources Nov 05 20:58:09 that's why fsogsmd switches off Nov 05 20:58:19 backlight on n900 maybe too... dunno Nov 05 20:58:27 but the actual shutdown then does not work Nov 05 20:59:43 on GTA02 on old oe dev i had alwys the problem that i had to shutdown twice Nov 05 21:00:03 if [ "$EUID" = "0" ] || [ "$USER" = "root" ] ; then PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:${ROOTPATH}" Nov 05 21:00:06 else PATH="/usr/local/bin:/usr/bin:/bin:${PATH}" Nov 05 21:00:08 nschle85: that was something different Nov 05 21:00:11 fi Nov 05 21:00:37 nschle85: you could try if removing that if in /etc/profile makes it shutdown Nov 05 21:00:45 to verify PATH is actually the problem Nov 05 21:00:50 have to reboot before trying though Nov 05 21:01:01 PATH="/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:${ROOTPATH}" Nov 05 21:01:04 just this one Nov 05 21:01:11 ok Nov 05 21:03:52 is fsousaged not executed as root ? Nov 05 21:04:06 i cannot look, my phone is booting Nov 05 21:04:24 hmm... it is started via dbus-activation Nov 05 21:04:37 so the env might be cleaned somehow Nov 05 21:04:45 and EUID and USER not set correctly Nov 05 21:09:32 mrmoku: no that does not help Nov 05 21:09:46 nschle85: hmm... thinking about it... maybe making it an absolute path is maybe the correct fix Nov 05 21:09:52 it should not depend on environment Nov 05 21:10:00 nschle85: yeah, ok Nov 05 21:11:05 mrmoku: ok make it configurable :-) Nov 05 21:12:26 nschle85: ok, will fix that tomorrow. One way or the other :) Nov 05 21:12:38 now I'm on my way to bed Nov 05 21:12:46 mrmoku: ok, bye Nov 05 21:13:07 nschle85: gnight Nov 05 21:13:28 nschle85: btw. last night message arrived Nov 05 21:13:31 bye Nov 05 22:15:31 PaulFertser: sorry to ping you but did you have time to look at the qi stuff? ;) I'd like to repartition to only have one partition but I'll do that only when debian has a qi that can still support two kernel options in that setup :) **** ENDING LOGGING AT Sun Nov 06 02:59:57 2011