**** BEGIN LOGGING AT Thu Feb 11 02:59:57 2010 Feb 11 03:12:53 sorry to scare you guys Feb 11 04:10:56 tmzt : Hi Feb 11 04:43:13 hello Feb 11 06:23:41 morning Feb 11 06:24:36 yeah, morning Feb 11 06:27:20 mrmoku: another black screen on wakeup Feb 11 06:27:40 2010-02-11T06:25:03.093476Z [DEBUG] KernelIdleNotifier : onState transitioning from s1 to s0 Feb 11 06:27:40 2010-02-11T06:25:03.112186Z [DEBUG] Kernel26Display : Brightness set to 0 Feb 11 06:29:08 Q-Master, sounds like the phonefso brightness bug Feb 11 06:29:19 TAsn: definitely Feb 11 06:31:14 anyhow, gotta go to work, ciao. Feb 11 06:39:12 Q-Master: hmm Feb 11 06:39:44 Q-Master: have to go too now... do you some phonefsod.log around that time? to see what's happening? Feb 11 06:39:49 will take a look then later Feb 11 06:39:56 anyhow... bbl Feb 11 06:45:30 Wonka: remove that file as well as the same with trailing .md5 Feb 11 06:47:25 mrmoku|away: phonefsod.log is 0 sized. log level WARNING there. Feb 11 06:48:03 TAsn : hi Feb 11 06:49:15 is the SPI driver implementation on openmoko is hardware way or GPIO based software protocol way ? Feb 11 06:54:17 satish_: spi_s3c24xx_gpio.c Feb 11 06:55:54 PaulFertser : is the file avaiable with SHR testing image / Feb 11 06:56:29 satish_: what? It's the kernel driver source file. Feb 11 06:56:36 satish_: apparently bitbanging. Feb 11 06:57:11 PaulFertser : i have only the below files Feb 11 06:57:13 root@satish-desktop:/home/satish/shr/shr-testing# find . -name *spi*.c ./tmp/staging/om-gta02-oe-linux-gnueabi/kernel/arch/arm/mach-pxa/spitz_pm.c ./tmp/staging/om-gta02-oe-linux-gnueabi/kernel/arch/arm/mach-pxa/spitz.c ./tmp/staging/om-gta02-oe-linux-gnueabi/kernel/arch/arm/plat-s3c24xx/spi-bus0-gpe11_12_13.c ./tmp/staging/om-gta02-oe-linux-gnueabi/kernel/arch/arm/plat-s3c24xx/spi-bus1-gpg5_6_7.c ./tmp/staging/om-gta01-o Feb 11 06:57:41 PaulFertser : In drivers i don't see any SPI driver stuff Feb 11 06:58:01 satish_: are you sure you're on andy-tracking branch? Feb 11 06:58:31 i have downloaded from SHR -Testing image Feb 11 06:58:47 let me post u the site addr Feb 11 07:00:11 satish_: it makes no sense, i use git.openmoko.org anyway. Feb 11 07:00:49 http://build.shr-project.org/shr-testing/ Feb 11 07:01:01 i have downloaded Makefile from here Feb 11 07:01:06 and used gi Feb 11 07:01:08 and used git Feb 11 07:02:05 wget http://shr.bearstech.com/Makefile Feb 11 07:03:38 satish_: drivers/spi/spi_s3c24xx_gpio.c Feb 11 07:05:44 PaulFertser : i have only media and crypto folders under dirver Feb 11 07:05:47 root@satish-desktop:/home/satish/shr/shr-testing# ls tmp/staging/om-gta02-oe-linux-gnueabi/kernel/drivers/ crypto media Feb 11 07:06:32 satish_: because you're looking into staging instead of the sources folder. Feb 11 07:07:26 PaulFertser ; what is the source path Feb 11 07:07:29 root@satish-desktop:/home/satish/shr/shr-testing# ls bitbake conf downloads exp Makefile openembedded setup-env tmp Feb 11 07:08:47 PaulFertser : how to get the source code downloaded for kernel using OE framework / Feb 11 07:08:55 satish_: it's you using OE, not me. Do something like bitback -c prepare linux-image or whatever. Feb 11 07:08:56 i do see only recipies Feb 11 07:09:25 and download folders Feb 11 07:09:42 satish_: you should ask bitbake to unpack the sources for you Feb 11 07:13:23 PaulFertser : bitbake is executing by downloading the tar.gz files into download dir Feb 11 07:13:39 satish_: bitbake -c blablabla Feb 11 07:36:24 ERROR: Please set the 'PERSISTENT_DIR' or 'CACHE' variable. Feb 11 07:36:30 any idea ? Feb 11 07:40:31 . ./setup-env Feb 11 07:40:37 and reetry? Feb 11 07:43:15 mrmoku|away: maybe this is the cause why buildhost is rebooted so often http://pastebin.ca/1793583 Feb 11 07:47:59 is it possible to tell a bitbake recipe to always use the most recent version of a svn checkout? Feb 11 07:48:55 blindcoder: SRCREV = "${AUTOREV}" Feb 11 07:50:11 JaMa: ah, great. I have tagged fltkspacetrader 1.0-stable now to be able to further work on trunk and still backport bugfixes (twice already...) without having to update the recipe in shr every other day Feb 11 07:51:08 blindcoder: btw if there would be SRCREV_pn-fltkspacetrader in sane-srcrevs.inc (which is not IIRC), then you have to set AUTOREV there as it has higher priority Feb 11 07:52:55 JaMa: noted, thanks Feb 11 07:53:58 blindcoder: and I don't mind pushing SRCREV update every other day :) with AUTOREV you have to swear that you won't commit something which won't compile :) Feb 11 08:00:15 Weiss: that segfault yesterday (bad bt :/) http://pastebin.ca/1793598 Feb 11 08:02:09 JaMa: hmm.. did you update the DRI2 extension package when you updated X.org? Feb 11 08:03:26 Weiss: you mean dri*proto? because xserver-xorg-extension-dri2 is built with xserver and thus updated with it Feb 11 08:04:44 Weiss: please note, then now I'm using kms enabled DDX with kms enabled kernel with mesa-dri with only swrast compiled (maybe the fallback mechanism doesn't work as before).. Feb 11 08:05:45 Weiss: because before we used to fallback to swrast when dri wasn't available and now I have it available but dri driver in mesa is missing Feb 11 08:06:18 I mean fallback between glamo mesa-dir and swrast Feb 11 08:54:48 blindcoder: sorry for confuse you (I expected that you want AUTOREV for local development), but SRCREV = "${AUTOREV}" IN recipe is not so good as some stable rev in recipe and then SRCREV_pn-fltkspacetrader = "${AUTOREV} in some distro config like shr-autorev.inc Feb 11 08:55:28 blindcoder: because in 1 case, upstream svn repository is checked for latest revision with every build (for every distribution using OE) Feb 11 08:55:57 blindcoder: ie Angstrom doesn't use AUTOREV at all (as it's dangerous - build can fail even without any change in OE repo) Feb 11 08:56:10 blindcoder: can you update your last patch? Feb 11 08:59:08 ьс Feb 11 09:07:45 Weiss: rebase of mesa/glamo to master was really pretty straightforward only small conflict in configure.ac and configs/linux-dri Feb 11 09:09:36 raster: does there exists sth. like a notification area for ilume? Where one can show icons for missed calls or messages? If not, what would be the best method to use in e to trigger events in such an area? Feb 11 09:10:07 Heinervdm: illume2 has them Feb 11 09:10:09 notification windows Feb 11 09:10:24 click top bar to show/hide them Feb 11 09:10:36 technically there should be something saying u have something there to look at Feb 11 09:10:41 but currently there isnt Feb 11 09:11:53 raster: ok, then i will wait for illume2 Feb 11 09:14:44 spaetz : Hi Feb 11 09:15:47 spaetz : after . ./setup-env also, got the same error Feb 11 09:16:03 ERROR: Please set the 'PERSISTENT_DIR' or 'CACHE' variable. Feb 11 09:16:14 JaMa: ah, okay, I will do that Feb 11 09:16:40 JaMa : any clue on above error Feb 11 09:18:04 JaMa: so I'll just continue with SRCREV="110" in this case. Even so, I expect updates to come in less frequent now :) Feb 11 09:18:20 satish_: you can set CACHE in local.conf Feb 11 09:19:37 i have set to CACHE="" Feb 11 09:19:48 in local.conf Feb 11 09:20:03 blindcoder: yes.. use latest (tested) revision in recipe and then you can add AUTOREV to distro config if you expect small updates from time to time Feb 11 09:20:20 satish_: set it to a real dir Feb 11 09:20:46 in the sense which dir ? any tmp dir ? Feb 11 09:21:06 satish_: it should be tmp/cache Feb 11 09:21:09 blindcoder: but we're using as little autorevs as possible, because it's pity to have whole image build failing because some minor bug in latest commit.. but for "extra" recipes just in feeds I'm ok with autorev Feb 11 09:21:40 Heinervdm, satish_: IIRC if you remove that line, it will be set to sane default tmp/cache Feb 11 09:22:10 satish_: don't confuse CCACHE="" and CACHE Feb 11 09:22:38 i have both CACHE and CCACHE Feb 11 09:22:41 and more precisely default would be something like ${TMPDIR}/cache Feb 11 09:23:07 satish_: CCACHE is ok if you don't want to use ccache (few issues while rebuilding), but empty CACHE is WRONG Feb 11 09:23:20 CCACHE = "" CACHE = "" Feb 11 09:23:25 in my local.conf Feb 11 09:23:27 CACHE = "${TMPDIR}/cache${@['', '/' + str(bb.data.getVar('MACHINE', d, 1))][bool(bb.data.getVar('MACHINE', d, 1))]}" Feb 11 09:23:38 satish_: REMOVE CACHE = "" Feb 11 09:23:57 JaMa : in local.conf Feb 11 09:24:02 yes Feb 11 09:24:30 SRCPV cannot work without it and also persistent cache for latest revs needs it Feb 11 09:25:02 JaMa : that error got resolved Feb 11 09:25:21 JaMa : how to get SHR -Testing image kernel source code using bitbake files ? Feb 11 09:25:45 bitbake -c fetch -b recipes/linux/linux-openmoko-shr-devel_git.bb Feb 11 09:34:32 JaMa ; In which dir the source code will fetched ? Feb 11 09:34:47 tmp/work/armv4*/linux* Feb 11 09:35:21 satish_: if you're also interested in patches and defconfig use -c configure Feb 11 09:38:16 JaMa ; i do see only header files Feb 11 09:39:33 tmp/work/armv4t-oe-linux-gnueabi/linux-libc-headers-2.6.31-r0/ Feb 11 09:41:51 tmp/work/om-gta02-oe-linux-gnueabi/linux-openmoko-shr-devel-2.6.29-oe11+gitr0+b90406de472c1aa5371ab593a2bb79136d5de658-r6/temp/ Feb 11 09:42:38 yes the second one is right.. and you can see that while running that bitbake cmd.. Feb 11 09:43:38 JaMa : in the second also don't have the src directory Feb 11 09:43:50 because its named git Feb 11 09:44:02 and it's 1 level above temp Feb 11 09:44:23 [SHR-U] I did an opkg-update this morning and now my display is.. strange. The SHR home screen takes about 0.75x0.75 of the top-left corner, and the PIN screen takes about 75% (in both directions) on top of it. Any ideas? Feb 11 09:44:35 satish@satish-desktop:~/shr/shr-testing$ ls -l tmp/work/om-gta02-oe-linux-gnueabi/linux-openmoko-shr-devel-2.6.29-oe11+gitr0+b90406de472c1aa5371ab593a2bb79136d5de658-r6/ total 4 drwxr-xr-x 2 satish satish 4096 Feb 11 15:02 temp Feb 11 09:44:44 i have only temp in that ... Feb 11 09:44:58 i don't git in my /tmp/work Feb 11 09:45:21 satish@satish-desktop:~/shr/shr-testing$ find tmp/work/ -name git satish@satish-desktop:~/shr/shr-testing$ Feb 11 09:46:53 satish@satish-desktop:~/shr/shr-testing$ bitbake -c fetch -b recipes/linux/linux-openmoko-shr-devel_git.bb NOTE: Preparing runqueue NOTE: Executing runqueue NOTE: Tasks Summary: Attempted 2 tasks of which 2 didn't need to be rerun and 0 failed. Feb 11 09:48:08 satish_: -c clean first Feb 11 09:51:57 satish_: or use -f Feb 11 09:52:55 atish@satish-desktop:~/shr/shr-testing$ bitbake -c -f fetch -b recipes/linux/linux-openmoko-shr-devel_git.bb NOTE: Preparing runqueue ERROR: Task do_-f does not exist for target linux-openmoko-shr-devel Feb 11 09:53:54 i have done bitbake -c clean world Feb 11 09:54:16 satish_: you have extra -f in there Feb 11 09:55:01 bitbake -f fetch -b recipes/linux/linux-openmoko-shr-devel_git.bb Feb 11 09:55:03 is working Feb 11 09:57:47 Weiss: ah, rebase was easy in git, but then it fails to compile http://tinderbox.openembedded.net/packages/474887/ Feb 11 09:58:15 satish_: lol please read at least bitbake -h Feb 11 09:58:42 ok Feb 11 09:58:43 http://bitbake.berlios.de/manual/ch04s02.html#id870080 Feb 11 09:58:43 bitbake -f -c fetch, would be much better Feb 11 09:59:14 then you can see that -f is option without value and -c is followed by cmd Feb 11 09:59:26 yes Feb 11 09:59:28 :) Feb 11 09:59:38 so -c -f fetch and -f fetch is both non-sense :) Feb 11 09:59:57 yes Feb 11 10:00:15 well the later will do something.. (try to build package "fetch" and then that kernel recipe) Feb 11 10:00:24 bitbake -f -c fetch -b recipes/linux/linux-openmoko-shr-devel_git.bb Feb 11 10:00:50 satish_: expecting you're in openembedded directory Feb 11 10:00:53 is this recipe is correct for kernel ? Feb 11 10:01:04 satish_: if you're somewhere else then update path in -b Feb 11 10:01:11 satish_: yes it is Feb 11 10:01:39 atish@satish-desktop:~/shr/shr-testing/openembedded$ pwd /home/satish/shr/shr-testing/openembedded Feb 11 10:02:14 satish@satish-desktop:~/shr/shr-testing/openembedded$ bitbake -f -c fetch -b recipes/linux/linux-openmoko-shr-devel_git.bb NOTE: Preparing runqueue NOTE: Executing runqueue NOTE: Running task 1 of 2 (ID: 0, /home/satish/shr/shr-testing/openembedded/recipes/linux/linux-openmoko-shr-devel_git.bb, do_setscene) NOTE: Running task 2 of 2 (ID: 1, /home/satish/shr/shr-testing/openembedded/recipes/linux/linux-openmoko-shr-devel_g Feb 11 10:03:46 looks success Feb 11 10:03:48 but Feb 11 10:03:52 satish@satish-desktop:~/shr/shr-testing/openembedded$ ls ../tmp/work/om-gta02-oe-linux-gnueabi/linux-openmoko-shr-devel-2.6.29-oe11+gitr0+b90406de472c1aa5371ab593a2bb79136d5de658-r6/ temp Feb 11 10:05:32 satish_: ls ../tmp/work/om-gta02-oe-linux-gnueabi/linux-openmoko-shr-devel-2.6.29-oe11+gitr* Feb 11 10:06:29 that has only temp dir Feb 11 10:06:32 nothing else Feb 11 10:07:28 isn't there ie *gitr1* directory? Feb 11 10:08:03 you should see output where it's unpacked from that bitbake cmd (aah use -c unpack instead -c fetch :)) Feb 11 10:08:03 yes.... not git dir Feb 11 10:09:09 NOW got it Feb 11 10:09:36 satish_: I'm talking about *gitr* not git see ../tmp/work/om-gta02-oe-linux-gnueabi/linux-openmoko-shr-devel-2.6.29-oe11+gitr1+a15608f241a40b41fed5bffe511355c2067c4e88 Feb 11 10:10:24 in git i have the below Feb 11 10:10:29 git$ ls COPYING Kbuild README block dfu-kern fs ipc makerecovery remote_install_sdcard security virt CREDITS MAINTAINERS REPORTING-BUGS build drivers include kernel mm samples sound Documentation Makefile arch crypto firmware init lib net scripts usr Feb 11 10:24:32 JaMa : how can i make my spi co processor device detected by freerunner phone Feb 11 10:24:56 JaMa : i don't see /dev/spi Feb 11 10:25:01 on phone Feb 11 10:28:34 i want to update spi driver in kernel (that we have downloaded just now ...)and use the same to build and run on phone Feb 11 10:28:47 Weiss: to compile after rebase http://jama.homelinux.org/org.openembedded.shr.images/0001-glamo-remove-Private-postfix-from-__DRI-wrt-http-cgi.patch Feb 11 10:29:16 Q-Master: heh, WARNING is of not much use to solve it :P Feb 11 10:29:17 blindcoder: those patches for fltkspacetrader pushed Feb 11 10:29:20 blindcoder: thanks Feb 11 10:30:01 mickey|zzZZzz: that libeflvala_svn you pushed is not available in EFL_SRCREV we have in oe.dev.. testing newer revision now.. Feb 11 10:30:06 JaMa: ouch :( Feb 11 10:30:16 JaMa JaMa ; how can i use my local fetch & unpacked kernel to build with modifications instead of getting from git repository ? Feb 11 10:30:49 i mean what is the bitbake cmd to pick local files instead of git repository Feb 11 10:30:51 satish_: see conf/devellocal.inc Feb 11 10:31:12 mrmoku: not so big problem as long as we have older git based libeflvala in feeds :) Feb 11 10:31:40 mrmoku: but EFL bump needs few recipes update as some parts are moved from trunk/OLD to OLD and the same fro BROKEN Feb 11 10:31:54 JaMa : satish@satish-desktop:~/shr/shr-testing$ ls conf/ auto.conf local-builds.inc local.conf site.conf topdir.conf Feb 11 10:32:05 i don't have conf/devellocal.in Feb 11 10:32:26 satish_: conf/local-builds.inc Feb 11 10:32:30 i have local-build.in Feb 11 10:32:34 ok Feb 11 10:34:09 Heinervdm : do i need to add this kernel here ? Feb 11 10:34:17 sorry I'm not using shr Makefile so I remeberred the name wrong and used different here Feb 11 10:34:18 yes Feb 11 10:38:13 JaMa: I was referring to the reboots Feb 11 10:38:38 mrmoku: ah ic Feb 11 10:39:14 mrmoku: but I'm building there right now without reboot :) Feb 11 10:39:30 :) Feb 11 10:46:54 Heinervdm : i see libphone related stuff Feb 11 10:47:01 in loca-build.inc Feb 11 10:47:10 satish_: that are examples Feb 11 10:47:18 for kernel Feb 11 10:48:21 uImage ...what is the lib name ? Feb 11 10:49:41 linux-libc-headers ? Feb 11 10:56:23 JaMa: cool.. nicely done Feb 11 10:56:33 (central IT killed my desktop, so I can't check properly :S) Feb 11 10:57:38 to communicate with SPI co processor device from freerunner phone ............. Feb 11 10:57:44 i have to update SPI driver in kernel source code and build the local kernel Feb 11 10:58:02 and load on to the phone ....after boot i will get /dev/spi device nodes Feb 11 10:58:15 and able to program using IOCTLs Feb 11 10:58:33 to open and read , write and spi message transfer Feb 11 10:58:54 i feel the above procedure is correct Feb 11 10:59:08 any other procedures or ways to do the same / Feb 11 10:59:09 good evening everyone Feb 11 11:00:22 JaMa and Heinervdm and mrmoko and PaulFertser : can u guide me .... Feb 11 11:00:53 is the above procedure is ok to go ahead Feb 11 11:01:00 or any other simple way / Feb 11 11:02:29 JaMa : ? Feb 11 11:02:40 Heinervdm : ? Feb 11 11:06:00 satish_: i have no idea about IOCTL Feb 11 11:06:01 satish_: do you mean you want to have a userspace API for spi to use your device with userspace driver? Feb 11 11:06:11 satish_: kinda libusb-like? Feb 11 11:06:39 to build form a local directory, remove the # infront of the require local-builds line in local.conf Feb 11 11:06:40 PaulFertser : i want to communicate with co processor Feb 11 11:06:56 so i felt ...i need to write user space API's IOCTLs Feb 11 11:06:58 and set the variables for linux-openmoko-shr-devel and remove all others Feb 11 11:07:54 satish_: do you want to write a kernel or userspace driver? Feb 11 11:08:08 Heinervdm : ok...i will do it.....any suggesstions from ur side to my question to communicate with SPI device Feb 11 11:08:41 satish_: no, i have no knowlege about kernel programming Feb 11 11:08:49 PaulFertser : any thing is ok ...but want to communicate with the SPI device in the simplest way Feb 11 11:09:13 <[Rui]> hi Feb 11 11:09:29 [Rui]: hi :) Feb 11 11:09:30 PaulFertser ; as possible and able to do it as early as possible....i have demo in the weekend Feb 11 11:09:38 <[Rui]> just upgraded shr, and it seems something's wrong as pin dialog doesn't show up! Feb 11 11:10:23 satish_: read this Documentation/spi/spidev ? Feb 11 11:10:29 Weiss: it's not complete :/ glamo_screen.c:105: error: too few arguments to function 'driCreateConfigs' I'll check Feb 11 11:10:50 PaulFertser ; doesn't matter ...wheather it is user space or kernel space for my app ....but able to communicate with the device Feb 11 11:11:09 PaulFertser ; i have read little but i don't see /dev/spi on my phone Feb 11 11:11:18 satish_: have you actually read the document in question? Feb 11 11:13:03 PaulFertser ; i read it Feb 11 11:13:57 satish_: so what's not clear about it? Feb 11 11:14:41 PaulFertser ; i ran mdev on phone Feb 11 11:15:00 PaulFertser : but didn't get the spi related device on phone Feb 11 11:15:16 if i can able to get that...i could have tried using Ioctls Feb 11 11:15:43 then i started looking into the kernel spi driver as i don't see /dev/spi.... Feb 11 11:15:56 satish_: have you modified spi_board_info ?! Feb 11 11:16:02 and it is mentioned in the spidev as don't modifiy using hand under /dev Feb 11 11:16:09 yes Feb 11 11:16:29 spi_board_info in kernel source ? Feb 11 11:16:42 or mach-omgta02.c Feb 11 11:17:31 satish_: were you running a modified kernel when you couldn't see /dev/spidev* nodes? Feb 11 11:18:37 no Feb 11 11:18:47 i have used the shr testing image kernel Feb 11 11:18:49 uImage Feb 11 11:18:53 tmp/staging/om-gta02-oe-linux-gnueabi/kernel/arch/arm/mach-s3c2442/mach-gta02.c Feb 11 11:19:12 in this the spi_board_info Feb 11 11:19:20 u r talking about ? Feb 11 11:19:26 <[Rui]> 2010-02-11T11:19:04.775406Z [ERROR] UsageController <8 R>: Resource GSM can't be enabled: Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken.. Trying to disable instead Feb 11 11:20:33 PaulFertser : i just update the table in mach-omgta02.c....and don't know how to build and load on to phone Feb 11 11:21:26 satish_: so no wonder you didn't see the device node... Feb 11 11:22:16 satish_: the spi table is for spi clients. you want a master... Feb 11 11:22:45 ok... Feb 11 11:22:55 how to build this and load on to the phone Feb 11 11:23:57 my co processor is a slave device and want to make freerunner as master and communicate Feb 11 11:26:13 <[Rui]> help, I got no gsm :) Feb 11 11:28:13 PaulFertser : how to enable that /dev/spi0.... Feb 11 11:30:53 satish_: i'm sorry to repeat it but what is not clear in the documentation? Feb 11 11:32:09 * pespin just discovered a while ago that the university canteen is a good place to develop a bluetooth app, as it's full of bt devices xD Feb 11 11:33:05 Weiss: the second and still not last issue after rebase http://jama.homelinux.org/org.openembedded.shr.images/0001-glamo-update-driCreateConfigs-call-wrt-http-cgit.patch Feb 11 11:33:56 PaulFertser ; in which file i need to update spi_board_info Feb 11 11:34:13 satish_: mach-gta02.c i guess Feb 11 11:34:24 [Rui]: what logfile has that line, dmesg? I've also got no GSM but a different error message Feb 11 11:35:01 PaulFertser : how to build that file and load on to the phone ? Feb 11 11:35:06 <[Rui]> jaacoppi: fsousaged.log Feb 11 11:35:10 satish_: build the whole kernel and boot it as usual. Feb 11 11:36:17 [Rui]: sth interesting in phonefso.log and phoneui.log? Feb 11 11:36:23 PaulFertser : both kernel and rootfs ? Feb 11 11:37:00 satish_: i'm sorry but i'm starting to doubt you've enough experience to write any driver, be it kernel or userspace... Feb 11 11:37:13 <[Rui]> Heinervdm: phoneui claims to initiate the call Feb 11 11:37:18 <[Rui]> 2010.02.11 11:37:05.573339 [libphone-ui] MESSAGE: Inititating a call to 5885695 Feb 11 11:37:22 <[Rui]> invented number Feb 11 11:37:28 <[Rui]> mickeyterm hangs Feb 11 11:37:35 PaulFertser : i'm newbe to linux driver programming Feb 11 11:37:35 <[Rui]> I would suspect something's wrong Feb 11 11:37:45 <[Rui]> very wrong, I mean Feb 11 11:38:16 [Rui]: have you dialed that number? Feb 11 11:38:25 PaulFertser ; i want to write user space driver Feb 11 11:38:36 <[Rui]> Heinervdm: a real number works just the same Feb 11 11:38:45 <[Rui]> I only invented it to not paste a real number :) Feb 11 11:39:25 [Rui]: i meant your problem was that the pin dialog doesn't show up, so it's should be impossible to dieal Feb 11 11:39:39 <[Rui]> Heinervdm: Dialer comes up Feb 11 11:40:10 I have a gsm resource error in phonefsod. I'll be back in 20min or so, have to walk home first Feb 11 11:40:29 [Rui]: yes, but without sim auth it's without use :) Feb 11 11:40:44 satish_: just recompile the kernel with the right changes and boot it, you can use any rootfs you're comfortable with. Feb 11 11:40:47 <[Rui]> should check for it, then... Feb 11 11:41:19 [Rui]: currently it expexts that every thing went fine Feb 11 11:41:27 but it hasn't went fine Feb 11 11:41:47 usally it's phonefsod failing to connect to frameworkd Feb 11 11:42:29 PaulFertser : previously i used bitbake command to build both kernel and rootfs Feb 11 11:42:55 bitbake world Feb 11 11:42:59 <[Rui]> 2010.02.11 11:42:21.702 frameworkd.resource INFO setting resource status for GSM from unknown to disabling Feb 11 11:42:59 <[Rui]> 2010.02.11 11:42:21.709 frameworkd.resource INFO setting resource status for GSM from disabling to disabled Feb 11 11:42:59 <[Rui]> 2010.02.11 11:42:21.766 frameworkd.resource INFO setting resource status for GSM from disabled to enabling Feb 11 11:42:59 <[Rui]> 2010.02.11 11:42:22.508 ogsmd.channel INFO : Creating channel with timeout = 3600 seconds Feb 11 11:43:01 <[Rui]> 2010.02.11 11:42:22.552 ogsmd.channel INFO : Creating channel with timeout = 300 seconds Feb 11 11:43:04 <[Rui]> 2010.02.11 11:42:22.629 ogsmd.channel INFO : Creating channel with timeout = 300 seconds Feb 11 11:43:07 <[Rui]> doesn't look good Feb 11 11:43:32 satish_: i'm not sure if it's easier to compile the kernel with OE for your particular purposes or not. Feb 11 11:43:54 [Rui]: enabeled is missing Feb 11 11:44:15 satish_: because i'm not familiar with OE Feb 11 11:44:32 PaulFertser : ok Feb 11 11:44:50 if i want to do it as separate .....without OE Feb 11 11:45:07 satish_: andy-tracking has a handy script called build Feb 11 11:45:20 where can i download kernel source for SHR Testing image ? Feb 11 11:45:32 can u post me the site addr / Feb 11 11:45:52 satish_: i thought OE already downloaded that for you Feb 11 11:46:09 satish_: it's b90406de472c1aa5371ab593a2bb79136d5de658 of andy tracking Feb 11 11:46:27 i didn't get u Feb 11 11:46:30 <[Rui]> Heinervdm: I know Feb 11 11:46:50 <[Rui]> Heinervdm: but that was all it had from a /etc/init.d/frameworkd start Feb 11 11:47:22 [Rui]: what does phonefsod say? Feb 11 11:47:32 http://people.openmoko.org/andy/ Feb 11 11:47:43 <[Rui]> Heinervdm: nothing on start Feb 11 11:47:50 <[Rui]> Heinervdm: only on stop, it tells it's stopping Feb 11 11:47:57 nothing in log? Feb 11 11:48:07 <[Rui]> Heinervdm: yup Feb 11 11:48:16 strange Feb 11 11:48:51 if there is no error everything should be fine Feb 11 11:49:32 <[Rui]> Heinervdm: however there is absence of evidence of everything should be fine Feb 11 11:49:42 <[Rui]> I suspect it has something to do with the gsm device Feb 11 11:49:48 <[Rui]> one of the upgrades was the kernel.... Feb 11 11:50:29 PaulFertser ; if i update the mach-gta02.c with spi board info and build the kernel and load on to the phone then...i'm able to see the /dev/spidev and able to use the user space API Feb 11 11:50:34 i'm upgrading now, let's see what happens Feb 11 11:50:48 <[Rui]> mickeyterm hangs Feb 11 11:50:49 PaulFertser ; to open, set SPI speeds and read , writes Feb 11 11:51:18 satish_: i think yes Feb 11 11:52:19 satish_: i think no Feb 11 11:53:01 larsc: why not? It looks like there's a userspace SPI API Feb 11 11:53:52 larsc: i mean it looks like he can write a userspace driver to use his coprocessor as an spi slave, why not? Feb 11 11:54:02 PaulFertser: yes, but you also need an spi master Feb 11 11:54:24 larsc: what's wrong with the one in andy-tracking? Feb 11 11:54:34 PaulFertser: afaik there is none Feb 11 11:54:39 larsc , PaulFertser : i don't see andy kernel source code Feb 11 11:54:41 That drives g-meters? Feb 11 11:55:05 http://people.openmoko.org/andy Feb 11 11:55:10 is this the site addr Feb 11 11:55:49 larsc: isn't spi_s3c24xx_gpio.c one? Feb 11 11:56:42 PaulFertser: yes it is Feb 11 11:56:55 but it's lines are not going to the debug port Feb 11 11:58:40 is it / Feb 11 11:59:06 i have connected the debug board with phone using the wire provided Feb 11 11:59:20 sure Feb 11 11:59:42 but schematics shows the debug board has SPI miso mosi cs lines Feb 11 11:59:42 but you need a different spi master then the one connected to the accelerometers Feb 11 11:59:43 satish_: no, larsc is maening that the spi bus to the accels is a different one than the debug board Feb 11 11:59:48 larsc: why can't he configure it to use the GPIO's that do go there? Feb 11 11:59:59 PaulFertser: he can Feb 11 12:00:16 [Rui]: are you using fsogsmd? Feb 11 12:00:26 either that or just add a second spi master Feb 11 12:00:55 <[Rui]> jl_gene: I don't even see that anywhere Feb 11 12:01:00 larsc : means the spi on debug board is dont have master on phone ? Feb 11 12:01:30 larsc ; and we need to create a new SPI master for it on phone ? Feb 11 12:01:30 satish_: it has but is a different bus. Bus 0 instead of one from the accels Feb 11 12:02:04 satish_: Sorry if I have to say it, but with your current knowledge about the problem I fear it won't work out until the weekend Feb 11 12:02:20 stefan_schmidt : ok... Feb 11 12:02:23 <[Rui]> Package fsogsmd (1:0.2.0+gitr803+42b94a43ef1dce1225c1f26301c0f4b8aab669e8-r2.2.4) installed in root is up to date. Feb 11 12:02:54 larsc: hm, i thought he connected to the same lines used for the g-meters stuff, because he was talking about adding his devices to that table with g-meters. Feb 11 12:03:03 stefan_schmidt : u mean i can't use the user space APIs avaiable on SPI driver in kernel ? Feb 11 12:03:55 [Rui]: Then try the patch mentioned on http://www.mail-archive.com/smartphones-userland@linuxtogo.org/msg02099.html Feb 11 12:04:05 satish_: you can, but you still need to setup everything correctly in the kernel first Feb 11 12:04:45 PaulFertser , larsc : stefan_schmidt ; i have solder debug board I2C/ SPI / IRD pins and connected external wires to my device platform Feb 11 12:04:49 stefan_schmidt: he can bitbang gpios from the userspace even :) Feb 11 12:05:43 satish_: thatt is the physical layer, still you need to configure it in the kernel... Feb 11 12:05:47 it is J11 on debug board Feb 11 12:05:59 PaulFertser: :) Feb 11 12:06:17 stefan_schmidt ; ok Feb 11 12:06:23 stefan_schmidt: in fact that might be an easier solution for someone new to kernel-level programming. Feb 11 12:06:48 stefan_schmidt ; how to configure the SPI lines on debug board in kernel Feb 11 12:07:41 PaulFertser: It might Feb 11 12:08:22 satish_: I don't walk you through all this step-by-step, sorry. PaulFertser and larsc are giving you pointer. Starting kernel dev is not something to be done in some hours. Feb 11 12:10:48 stefan_schmidt: s/giving/gave/ :-/ Feb 11 12:11:06 <[Rui]> jl_gene: trying, can't fix the patch from the web page, so I did all the changes manually Feb 11 12:12:07 <[Rui]> jl_gene: the pin now came up Feb 11 12:12:20 <[Rui]> I now have to go with my co-workers for lunch, brb Feb 11 12:15:58 [Rui]: in SHR ticket 890 is also mentioned than changing the GSM resurce policy to enabled during bootstrap also helps Feb 11 12:20:25 Weiss: and with last one http://jama.homelinux.org/org.openembedded.shr.images/mesa/ Feb 11 12:22:22 PaulFertser , stefan_schmidt , larsc : even if it takes much time also...i will try to learn and do it...i will initimate my guide abt ...it will take time.... Feb 11 12:22:45 PaulFertser , stefan_schmidt , larsc : please guide me how should i proceed Feb 11 12:22:52 i have confused little bit Feb 11 12:23:16 * PaulFertser wished he had a guide to intimidate Feb 11 12:23:30 satish_: 1) Find the git tree where the openmoko kernel ist hosted, 2) get it build 3) get it loaded on the phone and executed Feb 11 12:23:40 i want to do it in a proper way Feb 11 12:24:17 satish_: Once you have this setup with a kernel you can compile yourself and test you can get new guiding Feb 11 12:24:41 stefan_schmidt : ok...i will do that first Feb 11 12:25:13 can i go with andy tracking kernel ? Feb 11 12:25:45 satish_: that's what is used in that linux-openmoko-shr-devel in shr-t/shr-u, so yes Feb 11 12:27:17 satish_: yes Feb 11 12:27:32 satish_: that is compiled from the andy-tracking branch Feb 11 12:29:16 ok Feb 11 12:29:36 i'm picking it from Feb 11 12:29:36 http://wiki.openmoko.org/wiki/Kernel Feb 11 12:29:58 Building the andy-tracking branch Feb 11 12:30:03 column Feb 11 12:30:16 yes Feb 11 12:30:33 it has only 2.6 Feb 11 12:30:37 git clone git://git.openmoko.org/git/kernel.git linux-2.6 Feb 11 12:30:42 how abt sub version Feb 11 12:30:54 2.6.X ... what will be the X ? Feb 11 12:33:11 satish_: just use the andy-tracking branch Feb 11 12:33:16 ok Feb 11 12:34:49 as for i remember after building the kernel ...i just need to copy to the phone with some fixed name "uImage"....to boot from that particular kernel Feb 11 12:34:59 is that correct ? Feb 11 12:36:17 satish_: read up the wiki, its all described in there how you flash a uImage Feb 11 12:36:27 ok Feb 11 13:15:10 stefan_schmidt ; git-checkout is not working Feb 11 13:15:20 stefan_schmidt: any clue ? Feb 11 13:16:48 satish_: for clue I would need to know what _exactly_ you tried and what _exactly_ the error was Feb 11 13:17:06 stefan_schmidt: satish@satish-desktop:~/openmoko/kernel/linux-2.6$ git-checkout --track -b andy-tracking origin/andy-tracking git-checkout: command not found Feb 11 13:17:22 satish_: If you expect help from others for free try hard to not waste their time: http://catb.org/~esr/faqs/smart-questions.html Feb 11 13:17:45 satish_: and the answer is already given to you by git Feb 11 13:17:54 read the error message Feb 11 13:18:02 <[Rui]> trac.shr-project.org not resetting passwords adequately Feb 11 13:18:08 <[Rui]> Firefox has detected that the server is redirecting the request for this address in a way that will never complete. Feb 11 13:18:24 satish_: http://catb.org/~esr/faqs/smart-questions.html Feb 11 13:18:37 stefan_schmidt: :) but i have installed git on my ubuntu Feb 11 13:19:19 satish_: but you don't have git-checkout and this is because it is no longer called this way. It is git co nowadays. Feb 11 13:19:32 satish_: please upate the wiki accordingly Feb 11 13:20:11 stefan_schmidt: but git-checkout is not in the list of packages Feb 11 13:20:33 stefan_schmidt: do u know which package has this command ? Feb 11 13:20:39 satish_: it was renamed, google would have told you, Feb 11 13:20:57 satish_: it is still the git package but you need to type git co instead of git-checkout Feb 11 13:21:30 satish_: anyway, I stop helping out here until you come back with a setup where you can flash your own kernel. Feb 11 13:22:21 stefan_schmidt: ok... Feb 11 13:24:32 hmm those are pretty trivial questions.. I don't see how you can present working piece of hardware/driver in few days.. but you can surprise me :) Feb 11 13:25:55 JaMa: :) Feb 11 13:49:44 when i build shr, does it save the output of the compiling so i can see what went wrong? Feb 11 13:50:20 feitingen: yes and it also says where the log is Feb 11 13:50:48 feitingen: look for something like tmp/work/arch../package/temp/log.do_compile etc Feb 11 14:23:09 oooh lots-of-buttons-in-the-bottom illume bug is solved :D Feb 11 14:24:44 pespin: beer/cookies => Heinervdm Feb 11 14:25:51 jeje which was the problem? was it finally a theme or illume bug? Feb 11 14:26:24 pespin: http://git.openembedded.org/cgit.cgi/openembedded/commit/?id=aefc8c29f6f6014c20c86593cf15ffd69c28b1a8 Feb 11 14:26:36 stefan_schmidt: Hi Feb 11 14:27:03 stefan_schmidt: i have done with andy kernel build and load on to the phone and able to get phone up Feb 11 14:27:22 stefan_schmidt: root@om-gta02 / $ uname -r 2.6.29-GTA02_andy-tracking-mokodev Feb 11 14:27:38 pespin: hopefully I haven't broken it again with today's EFL bump :) Feb 11 14:27:57 could u please guide me the next step ? Feb 11 14:28:16 stefan_schmidt: Feb 11 14:28:27 JaMa, that looks like an ugly hack xD but if it works, hurray for us! Feb 11 14:31:01 pespin: http://trac.enlightenment.org/e/changeset/44462 Feb 11 14:33:08 stefan_schmidt: r u there ? Feb 11 14:33:36 satish_: now you need to modify mach-gta02.c appropriately, adding another spi master with proper GPIOs and then adding the slave to the corresponding slaves table. Feb 11 14:33:40 JaMa, hmm that change in e trac is the rev in which that got broken? Feb 11 14:34:02 PaulFertser: ok..thanks... Feb 11 14:34:13 PaulFertser: i will start that ... Feb 11 14:34:31 pespin: yes.. you can see there what was original intention for changing category and number of buttons Feb 11 14:36:33 JaMa: so.. does it work with all of those patches? Feb 11 14:36:35 JaMa, yeah, but isn't max the maximum number of buttons? I don't understand why is that actually drawing 10 buttons :S Feb 11 14:37:05 PaulFertser: can u expand little more details ... Feb 11 14:37:26 I mean, is should be set something (theme?) to draw as much buttons as max Feb 11 14:37:31 PaulFertser: i have to created SPI master first from drivers/spi folder Feb 11 14:38:35 pespin: I didn't checked it much (glad that Heindervdm fixed it), because I'm playing with illume2 instead.. Feb 11 14:39:21 pespin: but as there is FIXME: above all changed options I guess it's not configureable from theme yet Feb 11 14:39:34 satish_: there's already a driver, you need to create another structure to register it for the spi0 host with corresponding gpios. BTW, if it's too hard for you, why not bitbanging from the userspace? Feb 11 14:40:09 PaulFertser: ok... Feb 11 14:40:37 PaulFertser: but is it possible from userspace to probe my device Feb 11 14:40:44 satish_: (driver) look how it's done for spi1 and do something similar but probably bitbanging is easier for you. Feb 11 14:41:39 pespin: probably because bar_desktops = efreet_util_desktop_category_list("System"); is empty, then it creates max buttons as you can see in that while in the bottom of that file Feb 11 14:41:41 JaMa, yeah, just confused because of the name 'max' used, which should not affect the actual number of buttons displayed Feb 11 14:42:24 oh, ok Feb 11 14:46:28 PaulFertser: using spi_bitbanging.c ...i have to implement a wrapper to communicate with my device ...is it ? Feb 11 14:47:26 satish_: i'm afraid we're talking different languages :( Feb 11 14:47:35 yup... Feb 11 14:47:45 PaulFertser: but u mentioned that the SPI bus on the debug board is different from the one taking to accelarator ....so the available SPI master may not work Feb 11 14:48:35 PaulFertser: sorry ya....i might not understand u. Feb 11 14:49:13 Weiss: it compiles fine.. but error while starting Xorg is the same as without glamo driver in mesa :/ Feb 11 14:49:48 satish_: you just need to instantiate another instance of the spi_gpio driver. No need to write anything. It's just copy-pasting the one on mach-gta02.c and changing the gpio pins Feb 11 14:49:53 can be done in 10 minutes Feb 11 14:50:00 satish_: do you know what GPIO is? Do you know what bitbanging is? Do you understand how bytes travel over the SPI bus? etc etc Feb 11 14:50:01 PaulFertser: like spi_s3c24xx.c spi_s3c24xx_gpio.c....i have to create my spi_mydevice.c spi_mydevice_gpio.c ....... is it ? Feb 11 14:51:00 larsc: ok... Feb 11 14:52:21 JaMa: does it work with a non-DRM kernel, with no other changes? Feb 11 14:53:32 PaulFertser: i know GPIO and SPI protocol...but bitbanging term don't know Feb 11 14:54:30 satish_: bitbanging means you speak SPI by explicitly setting GPIO pins' state. Feb 11 14:54:46 PaulFertser: ok... Feb 11 14:55:06 Weiss: haven't tried with mesa glamo, as I cannot reboot from work Feb 11 14:55:52 PaulFertser: each bit transfer represent a MISO / MOSI gpio state is it ? Feb 11 15:00:01 PaulFertser: larsc : i'm about leave from my pc.....catch u tomorrow after finish this task... Feb 11 15:00:15 PaulFertser: larsc : thanks a lot ya..bye Feb 11 15:04:51 JaMa: I'm trying to determine whether it's a problem with DRI specifically, or if X is just broken Feb 11 15:23:03 <[Rui]> Weiss: I will re-enable omnewrotate on my phone, see if the black screen of death is happening again. Feb 11 15:23:40 <[Rui]> Weiss: I think suspend/resume doesn't like being associated with xrandr Feb 11 15:24:31 <[Rui]> I see the illume keyboard is still broken when horizontal... Feb 11 15:25:42 satish_: larsc: PaulFertser: SPI1 and SPI0 are *both* fully supported by SoC hw function. There's no need to do bitbang for SPI0 Feb 11 15:26:13 [Rui]: KMS or non-KMS? Feb 11 15:26:29 <[Rui]> Weiss: non kms I think. Feb 11 15:26:30 DocScrutinizer: and yet the kernel driver does bitbanging due to some issues with SPI api not fully available in IRQ contexts. Feb 11 15:26:44 <[Rui]> Weiss: for kms you need to install a different kernel than what is normally installed on shr-u right? Feb 11 15:26:45 damn Feb 11 15:27:07 DocScrutinizer: and i was suggesting that for a guy really new to kernel development it might be easier to write a simple userspace bitbanging driver than to understand how to hook up the kernel driver properly. Feb 11 15:27:10 <[Rui]> Weiss: so no kms :) Feb 11 15:27:31 stupid kernel issue, layering (sw problem) hindering to do the right thing Feb 11 15:27:37 :-((( Feb 11 15:27:53 <[Rui]> PaulFertser: a guy really new to kernel development should visit kernelnewbies.org Feb 11 15:28:05 <[Rui]> satish_: hint hint hint Feb 11 15:28:31 [Rui]: sometimes it's not worth learning kernel development when all you need is a simple spi driver, to me bitbanging from userspace seems like a reasonable alternative in this situation. Feb 11 15:29:04 <[Rui]> PaulFertser: I don't know spi from ips so I couldn't say :) Feb 11 15:30:04 PaulFertser: userspace means you reinvent all the protocol layer that is already there in kernel's SPI driver Feb 11 15:30:20 DocScrutinizer: SPI protocol layer is trivial anyway. Feb 11 15:30:28 maybe Feb 11 15:30:40 it's simpler than I2C for sure Feb 11 15:32:06 still I wonder how simple bitbanging from userland is anyway (except if your app is suid root) Feb 11 15:33:36 ah well, ignore me. I even fail to manage ssh pubkey auth Feb 11 15:35:30 and obviously kernel devels know much better, as they even decide to ignore a hw function in favor of doing it in sw so it won't violate their notion of a correct system design Feb 11 15:37:36 but hey, we got NODEBUG to optimize kernel. Why bother about 400 reads / second from accels done in bitbang Feb 11 15:38:07 Feb 11 15:39:09 slightly ot/ignorant: does the om have available gpio pins i can use? Feb 11 15:39:18 it has Feb 11 15:40:01 check schematics for TP-xxxx Feb 11 15:40:45 DocScrutinizer: btw, can the IRQ pin on the debug board be configured into normal GPIO mode so I could use it as second SS for SPI? Feb 11 15:41:24 stefan_schmidt: no idea off top of my head Feb 11 15:41:38 DocScrutinizer: ok, no problem. can look it up myself :) Feb 11 15:42:25 you're talking bout EINT3? Feb 11 15:42:49 satish_, were you looking for me? Feb 11 15:43:14 stefan_schmidt: EINT3 == GPF3 Feb 11 15:43:25 DocScrutinizer: moment, let me pull out the schematics. I mean the Interrupt pin we offer on the debug port Feb 11 15:43:36 you're talking bout EINT3? Feb 11 15:43:39 err Feb 11 15:43:43 stefan_schmidt: EINT3 == GPF3 Feb 11 15:44:00 that's the EINT we offer on debubo-conn Feb 11 15:44:14 DocScrutinizer: jup, that's it then Feb 11 15:45:00 DocScrutinizer: So it can be used as GPIO and one could have two SPI devices selected using a second slave select Feb 11 15:45:09 Good to know :) Feb 11 15:45:31 sure Feb 11 15:52:45 feitingen: see above dialog with stefan_schmidt. You can use all the SPI lines on debug board connector for GPIO primary function rather than SPI secondary function. same for EINT3. And then you'll find some more basically unused GPIO you can access via TP-xxx or some NC R on PCB Feb 11 15:53:15 thanks :) Feb 11 15:54:25 I seem to remember we counted up to some 16 GPIO you could use and access Feb 11 15:55:15 with minor reworks like removing a 0R or sth Feb 11 15:56:20 if you e.g. remove BT then you free furter GPIO for other usage Feb 11 16:01:02 [Rui]: hmm.. is this with .29 or .32? Feb 11 16:01:24 anybody an idea why the exactly same DSA-pubkey in .ssh/ works for loging in as root while it fails for non privileged user? Feb 11 16:02:01 DocScrutinizer: in the same .ssh folder? user at which end? Feb 11 16:02:06 SSH is fussy about permissions on .ssh Feb 11 16:02:54 i.e /root/.ssh/authorized_keys == /home/joerg/.ssh/authorized_keys Feb 11 16:03:17 login to root works, login to joerg fails Feb 11 16:04:05 far end obviously Feb 11 16:04:10 are the permissions right on both authorized_keys? SSH won't let you use pubkey authentication unless they have the right permissions Feb 11 16:04:23 locally I use same accound and private key Feb 11 16:05:13 I think they are identical, but maybe for root ssh doesn't care about them being incorrect? Feb 11 16:05:58 if anything, it'd care more about root's permissions being wrong.. Feb 11 16:06:08 owned by the right person? Feb 11 16:07:33 from man ssh I gather "authorized_keys: Feb 11 16:07:35 This file is not highly sensitive, but the recommended permissions Feb 11 16:07:36 are read/write for the user, and not accessible by others. Feb 11 16:07:37 " Feb 11 16:07:48 ah, I'm the wrong way round Feb 11 16:07:55 (because that file has the public parts, of course) Feb 11 16:08:28 ssh -v? Feb 11 16:08:44 doesn't yield much of a help Feb 11 16:09:45 debug1: Next authentication method: publickey Feb 11 16:09:46 debug1: Trying private key: /home/jr/.ssh/id_rsa Feb 11 16:09:48 debug1: Offering public key: /home/jr/.ssh/id_dsa Feb 11 16:09:49 debug1: Authentications that can continue: publickey,gssapi-with-mic,password Feb 11 16:09:51 debug1: Next authentication method: password Feb 11 16:09:52 DocScrutinizer: IIRC it was strict also about 700 for .ssh dir Feb 11 16:10:09 k, lemme check Feb 11 16:12:39 <[Rui]> Weiss: 29 Feb 11 16:13:34 mrmoku: See! Another guy with black screen. I suppose randr is the problem and omnewrotate. I too have it. Feb 11 16:13:38 Weiss: no diff Feb 11 16:14:12 mrmoku: waking up while in not same position as suspended forces black screen sometimes. Feb 11 16:14:35 aerr JaMa ^ Feb 11 16:15:53 <[Rui]> DocScrutinizer: make sure /home/joerg is 700, .ssh as well, authorized_keys is 400 Feb 11 16:15:54 Q-Master: you bet it will interfere. Just a guts feeling but still... Feb 11 16:16:16 <[Rui]> Q-Master: I think the problem is xrandr+suspend/resume Feb 11 16:16:45 [Rui]: ~ 700 ? o.O Feb 11 16:17:04 <[Rui]> Q-Master: and I (author of omnewrotate) can tell you for sure that rotation is not happening when it suspends nor when it resumes, but shortly after. Feb 11 16:17:24 <[Rui]> Q-Master: but perhaps the chip is so busy with something that it "dies" Feb 11 16:17:30 <[Rui]> DocScrutinizer: yeah, you have it like that? Feb 11 16:18:43 [Rui]: the problem is that it shows black screen while rotated against pre-suspended position Feb 11 16:18:49 I usually like to have my home 750 or 755 Feb 11 16:19:26 DocScrutinizer: if it gets as far as "offering key" for both cases, it sounds like the problem is at the other end.. Feb 11 16:19:40 you could play with ssh-add if it doesn't get that far Feb 11 16:19:58 it gets that far Feb 11 16:20:11 I'm trying to fix the other end Feb 11 16:22:01 <[Rui]> Q-Master: then you have another issue, because to me the phone is like powered off. Feb 11 16:22:04 <[Rui]> Q-Master: not even ping Feb 11 16:22:31 <[Rui]> DocScrutinizer: maybe you have strict permission checking, I don't know. Feb 11 16:22:32 wooho Feb 11 16:22:43 [Rui]: the problem is not only while suspend. this problem also happens if screen is dimmed off and then rotation happens Feb 11 16:22:46 <[Rui]> DocScrutinizer: was it permissions? Feb 11 16:23:06 chmod go-rx .ssh/authorized_keys ficed it Feb 11 16:23:08 <[Rui]> Q-Master: not here it just didn't. Feb 11 16:23:16 <[Rui]> DocScrutinizer: cool Feb 11 16:23:32 ~lart ssh manpage Feb 11 16:23:33 * apt changes ssh manpage's permissions to 0777 and tells the world Feb 11 16:23:42 cooool Feb 11 16:23:51 LOL Feb 11 16:24:02 ~botsnack Feb 11 16:24:02 :), DocScrutinizer Feb 11 16:24:04 [Rui]: why? the phone has active call, dimmed the screen, rotated -> voila. black screen. Feb 11 16:24:31 <[Rui]> Q-Master: active call? that's another thing, let me check Feb 11 16:25:16 [Rui]: exactly. not suspending, but dimming screen to black. Feb 11 16:25:16 thanks to all :-) Feb 11 16:25:33 <[Rui]> phone has active call now, screen is dimming Feb 11 16:26:02 <[Rui]> screen is now black and I have an ssh account active Feb 11 16:26:19 <[Rui]> just rotated. ssh is still alive and well Feb 11 16:26:23 [Rui]: make it rotated several times. Feb 11 16:26:30 <[Rui]> screen as well. Feb 11 16:26:35 <[Rui]> let's see it dim again. Feb 11 16:26:54 <[Rui]> Q-Master: I just did an upgrade today, so in all fairness maybe it's gone now.... Feb 11 16:27:07 <[Rui]> just rotated once Feb 11 16:27:12 <[Rui]> twice Feb 11 16:27:16 <[Rui]> thrice Feb 11 16:27:24 <[Rui]> ssh still on Feb 11 16:27:34 good if it is gone. Feb 11 16:27:38 and dim? Feb 11 16:27:45 [Rui]: so.. the problem is rotating while dimmed? Feb 11 16:27:50 <[Rui]> I give up Feb 11 16:28:01 <[Rui]> Weiss: I don't know. I always caught it after the fact Feb 11 16:28:10 [Rui]: is it lightens up again? Feb 11 16:28:18 <[Rui]> Weiss: I was now trying q-master's set and there was no problem at all Feb 11 16:29:01 <[Rui]> Q-Master: when it rotates it 1) saves the current value, 2) sets 0, 3) sets to the previous current value Feb 11 16:29:11 <[Rui]> if current value of brightness is 0, it will, of course, maintain 0. Feb 11 16:29:44 [Rui]: the problem seems that fsodeviced and omnewrotate fights for the value. Feb 11 16:29:52 <[Rui]> since I carry it horizontally, sideways, in my leather pouch, what I think that happens is: Feb 11 16:30:09 [Rui]: both trying to set the value. Feb 11 16:30:32 <[Rui]> a) receive call, b) I don't hear it so I don't pick it up, but c) the phone is still on and d) rotating as I walk around, and e) at some time it dies. Feb 11 16:31:02 [Rui]: I've had a black screen today while rotated upside down and ffalarms waked up the phone. Feb 11 16:32:11 <[Rui]> Q-Master: I believe there is some issue in the driver or it's the chip that is fault and can't take so many rotations in a certain amount of time Feb 11 16:32:41 [Rui]: 2010-02-11T06:25:02.900892Z [DEBUG] Kernel26Display : Brightness set to 0 Feb 11 16:32:41 2010-02-11T06:25:03.093476Z [DEBUG] KernelIdleNotifier : onState transitioning from s1 to s0 Feb 11 16:32:41 2010-02-11T06:25:03.112186Z [DEBUG] Kernel26Display : Brightness set to 0 Feb 11 16:34:28 <[Rui]> Q-Master: as I said, I couldn't distinguish the phone from a turned off phone. Feb 11 16:34:53 [Rui]: you'r problem is much worse then. Feb 11 16:34:59 <[Rui]> Q-Master: the only way I knew something was happening was because I pressed power for > 8 s and I heard a slight "pouf" as it turned effectively off. Feb 11 16:35:09 but anyway without SSH black screen is a black screen Feb 11 16:35:28 <[Rui]> Q-Master: yes, but plugging the cable there is not even IP address! Feb 11 16:35:32 <[Rui]> it doesn't ping Feb 11 16:36:20 kernel died then Feb 11 16:45:00 freesmartphone.org: 03tom 07framework * rb388de7f9f72 10/framework/subsystems/opimd/ (pimd_generic.py query_manager.py): opimd: Changed _limit to _pre_limit which limits before sorting, and changed _limit to limit after everything is done. Feb 11 16:45:05 <[Rui]> Q-Master: but it didn't panic either (no flashing red aux led) Feb 11 17:21:13 freesmartphone.org: 03mickey 07msmcomm * rd87d5631472c 10/libmsmcomm/vapi/msmcomm-1.0.vapi: libmsmcomm: vapi: fix callback and enums Feb 11 17:21:14 freesmartphone.org: 03mickey 07msmcomm * r15fca015edc5 10/msmvterm/src/ (commands.vala terminal.vala): msmvterm: register callbacks with context Feb 11 17:25:28 mickeyl, thanks for access. Feb 11 17:25:38 np, welcom on board Feb 11 17:25:46 thanks. Feb 11 17:25:54 lets hope I'll be productive enough :P Feb 11 17:26:00 hehe Feb 11 17:26:01 yeah Feb 11 17:26:23 * mickeyl boots the pre Feb 11 17:29:10 * TAsn envies mickeyl for having a pre. Feb 11 17:29:11 : Feb 11 17:29:12 :P Feb 11 17:29:52 Meh Feb 11 17:30:14 Now he has one, you can bet he'll play with it so much they won't be major FSO update for GTA :-) Feb 11 17:30:20 +s Feb 11 17:30:31 Argh my english is horrible tonight Feb 11 17:30:37 I should get myself a cup of coffee Feb 11 17:30:43 Ainulindale, but at least I got what you meant :P Feb 11 17:30:50 Anyhow, family guy is on, ciao. Feb 11 17:31:17 Family Guy \o/ Feb 11 17:48:10 Weiss: it works ok with non-kms kernel Feb 11 17:51:34 daniele_athome: ping Feb 11 17:51:51 hi zeusone Feb 11 17:52:07 daniele_athome: ciao I'm the telefoninux guy :) Feb 11 17:52:24 zeusone, ehm sorry I don't remember :( Feb 11 17:52:25 daniele_athome: I'm curious to know if you use edje in your program Feb 11 17:52:35 daniele_athome: no problem :) Feb 11 17:52:57 what program? :) Feb 11 17:53:17 edje but I bet that you don't use it :) Feb 11 17:53:30 actually I use it Feb 11 17:53:38 very often Feb 11 17:53:46 with elm? Feb 11 17:53:51 elm_layout? Feb 11 17:54:26 zeusone, yes Feb 11 17:54:37 daniele_athome: oh cool Feb 11 17:54:41 the new mokosuite written in C makes heavily use of edje Feb 11 17:54:47 for practically every window Feb 11 17:54:58 daniele_athome: so I will ask you the svn account for seeing your code :) Feb 11 17:55:04 eheh Feb 11 17:55:23 send an email with your preferred credentials Feb 11 17:55:32 daniele_athome: I saw the screenshot I was thinking that you use the normal elm Feb 11 17:55:36 you can take the address from my telefoninux profile Feb 11 17:55:44 zeusone, actually a mix of them Feb 11 17:55:44 daniele_athome: ok thanks Feb 11 17:56:08 I use edje for positionating objects and drawing special things (like the keypad buttons) Feb 11 17:56:19 daniele_athome: because I made my program in C with elm Feb 11 17:56:20 s/positionating/positioning/ Feb 11 17:56:20 daniele_athome meant: I use edje for positioning objects and drawing special things (like the keypad buttons) Feb 11 17:56:28 daniele_athome: but I never use elm_layout Feb 11 17:56:47 zeusone, the problem is that you need to know edje language Feb 11 17:56:53 but it is powerful indeed Feb 11 17:56:55 daniele_athome: ya I know Feb 11 17:57:08 daniele_athome: I'm studying it on Enna Feb 11 17:57:08 but I think that someone should write a widget design tool Feb 11 17:57:17 I mean edje is for graphical objects only Feb 11 17:57:19 daniele_athome: it is already done Feb 11 17:57:28 daniele_athome: by Gustavo Feb 11 17:57:44 zeusone, you mean editje? Feb 11 17:57:49 daniele_athome: yes Feb 11 17:57:54 it's not what I mean Feb 11 17:58:01 daniele_athome: ah Feb 11 17:58:03 I mean a widget design scripting language Feb 11 17:58:06 like edje, but for widgets also Feb 11 17:58:18 daniele_athome: ah I understood Feb 11 17:58:29 like glade for gtk, for instance Feb 11 17:58:34 :) Feb 11 17:58:58 there is one for qt too Feb 11 17:59:12 zeusone, I have to go now Feb 11 17:59:19 daniele_athome: miss only for Enlightenment Feb 11 17:59:23 daniele_athome: OK ciao Feb 11 17:59:23 have i nice night Feb 11 17:59:26 yes indeed Feb 11 17:59:26 you too Feb 11 17:59:32 thanks, bye! Feb 11 17:59:36 bye Feb 11 18:01:55 freesmartphone.org: 03mickey 07msmcomm * rc8065a4b5d74 10/libmsmcomm/vapi/msmcomm-1.0.vapi: libmsmcomm: vapi: can't create custom constructors in a .vapi; solve it slightly differently Feb 11 18:01:56 freesmartphone.org: 03mickey 07msmcomm * ra54f5e66fea6 10/libmsmcomm/src/ (message.c response_types.c): libmsmcomm: don't deliver the context as first parameter for the callback; use the user data instead Feb 11 18:02:17 WATCH OUT! Feb 11 18:02:19 * mickeyl hacking C Feb 11 18:02:27 8) Feb 11 18:02:34 oh no ;) Feb 11 18:02:37 i feel so dirty Feb 11 18:02:39 :D Feb 11 18:04:26 did you use C on your vibrator plugin again? :P Feb 11 18:10:16 heh Feb 11 18:11:26 freesmartphone.org: 03mickey 07msmcomm * r3950cea9db95 10/libmsmcomm/vapi/msmcomm-1.0.vapi: libmsmcomm: vapi: fix change_operation_mode Feb 11 18:11:27 freesmartphone.org: 03mickey 07msmcomm * r2ce0790a006e 10/msmvterm/src/ (commands.vala terminal.vala): msmvterm: basic communication with msmcommd seems to work; first two commands done Feb 11 18:15:55 * Q-Master is going to Moscow Feb 11 18:45:07 mrmoku, ping (PM) Feb 11 19:01:59 JaMa: ok... did you push the Xorg updates anywhere so I can try to reproduce it? Feb 11 19:09:13 bluez surprises me everyday... SIGNAL: DeviceFound --> 00:00:00:00:00:00 Feb 11 19:09:50 pespin, what about using bt handsets/keyboards? any planned support? Feb 11 19:09:50 :) Feb 11 19:10:00 TAsn, they are supported ;) Feb 11 19:10:15 really supported? Feb 11 19:10:18 or just supported? Feb 11 19:10:24 TAsn, you ware welcome to report if not :P Feb 11 19:10:34 pespin, I don't own neither Feb 11 19:10:37 so I can't report :) Feb 11 19:10:51 but if it works well, I can might consider getting one :) Feb 11 19:11:08 TAsn, there are toggles to bluez Input.Connect/Disconnect and Audio.Connect/Disconnect in emtooth ;) Feb 11 19:11:19 :) Feb 11 19:11:32 TAsn, I've got report from someone at spanish mailinglist and he has been able using emtooth Feb 11 19:12:14 :) Feb 11 19:12:59 TAsn, I don't have one neither, so I've to wait for other people to report xD Feb 11 19:17:46 Weiss: http://gitorious.org/~jama/openembedded/jama-shr-experimental/commits/shr/unstable Feb 11 19:19:31 Wow, we've been bumping efl like crazy people :) Feb 11 19:19:54 spaetz, what about migrating unstable to testing? everything works great now... Feb 11 19:20:20 TAsn: because libeflvala was moved from freesmartphone git to efl svn Feb 11 19:20:31 i c. Feb 11 19:20:50 do we use epsilon somewhere? Feb 11 19:21:01 no idea Feb 11 19:21:28 I fixed build of it today, but proposed at least python-epsilon removal too Feb 11 19:25:13 JaMa: ok, trying to compile.. (although I'm clearly jinxed today, so it'll probably explode my computer or something) Feb 11 19:31:14 Guys, please help, I bet that we all together can do wonders with this list: http://trac.shr-project.org/trac/wiki/Kernel%20Modules Feb 11 19:31:31 and even the non-tech guys, please add your list to the "lsmod" section Feb 11 19:34:38 TAsn: whole problem is imho this recipes/images/sdr-image.bb: kernel-modules \ Feb 11 19:34:50 what abou them? Feb 11 19:34:50 ah it's not shr-image but sdr-image :) Feb 11 19:35:00 :P Feb 11 19:35:28 but kernel-modules shouldn't be pulled to image, oe has other directives for pulling necessary modules Feb 11 19:35:45 and the rest *should* be built in feeds (just not included in image) Feb 11 19:38:54 TAsn: and those modules groups as someone suggested also already exists ie in dev/recipes/linux/linux-gumstix_2.6.15.bb:RDEPENDS_kernel-modules-wireless Feb 11 19:39:36 TAsn, yes, i really want to do that Feb 11 19:39:37 JaMa, the list at that page Feb 11 19:39:40 was generated from Feb 11 19:39:43 opkg list_installed Feb 11 19:39:43 :| Feb 11 19:39:56 probably during the weekend Feb 11 19:43:02 TAsn: yes but that's because whole kernel-modules package is pulled to image and that's because of MACHINE_EXTRA_RRECOMMENDS in om-gta02.conf Feb 11 19:43:31 JaMa, btw, of course I want to leave everything in the feeds, just not in the image. Feb 11 19:43:45 JaMa, that's why I have this list, to prevent this mayhem :P Feb 11 19:44:20 TAsn: ie om-gta01 has listed only few modules to install Feb 11 19:44:36 JaMa, maybe we should start with that list? Feb 11 19:45:16 JaMa: are you building something? Feb 11 19:45:21 yes with similar just g-ether+sound+other modules for network connection to download more Feb 11 19:45:26 mrmoku: on buildnost? no atm Feb 11 19:45:37 mrmoku: new efl was built already :) Feb 11 19:45:40 ok Feb 11 19:45:45 JaMa, wanna try it (in unstable) ? Feb 11 19:45:46 pyphonelog too? Feb 11 19:45:49 trim the list Feb 11 19:45:57 mrmoku, no Feb 11 19:46:09 neither phonelog nor framework were built Feb 11 19:46:11 TAsn: no problem with that Feb 11 19:46:21 JaMa, so please do your magic ;) Feb 11 19:52:58 dos1, /me fixed the limit thingie Feb 11 19:54:26 mrmoku: why did you asked about shr buildhost? Feb 11 19:54:38 mrmoku: I would restart 2.6.32.8 build as it failed.. Feb 11 19:54:40 dos1: hey, how do I add a multi value field with opimd-cli? Feb 11 19:54:45 JaMa, because I bugged him to build frameworkd and ppyhonelog Feb 11 19:54:51 yup Feb 11 19:54:52 :P Feb 11 19:54:59 mrmoku, probably using the python convention (I assume) Feb 11 19:55:02 i.e Feb 11 19:55:05 ok.. it doesn't colide :) Feb 11 19:55:23 opim-cli c edit 1 Phone ['bla', 'bla2'] Feb 11 19:55:27 but that's only an assumption :) Feb 11 19:55:33 no Feb 11 19:55:40 yeah, because I tried that :P Feb 11 19:55:48 Phone bla Phone bla2 Feb 11 19:55:48 ? Feb 11 19:55:49 mrmoku: opimd-cli c add Name dos IRC-nickname dos IRC-nickname dos1 IRC-nickname dos1|neo Feb 11 19:55:55 :P Feb 11 19:56:00 ohh Feb 11 19:56:01 :) Feb 11 19:56:24 mrmoku, supports looks promising? Feb 11 19:56:48 this ubifs on 2.6.29 is so nice, I have disabled bgt, but its still running since some time ago, and I can't mount nand anymore Feb 11 19:56:50 btw dos1 I hope I'll have time hacking frameworkd tonight :P Anyhow, I'm going to drop backend support completely (at least in my branch) Feb 11 19:56:51 :P Feb 11 19:57:01 TAsn: for now I can promise you only segfaults ;) Feb 11 19:57:14 mrmoku, sounds nice :) Feb 11 19:57:28 mrmoku, saw my comment about adding from phonelog? Feb 11 19:57:38 yep Feb 11 19:57:46 I think (assume) path is not initialized correctly Feb 11 19:57:59 (without even looking at the code/logs/anything) Feb 11 19:58:11 by path I mean contact path. Feb 11 20:00:16 mrmoku, http://shr-project.org/trac/ticket/881 <-- finished working on it? Feb 11 20:00:30 and ffs stupid thing won't register to network :( Feb 11 20:01:36 2010.02.11 21:57:18.923999 [phonefsod] MESSAGE: Error: The name org.shr.phoneui was not provided by any .service files Feb 11 20:01:37 :| Feb 11 20:02:15 heh... what version of phoneuid? Feb 11 20:02:17 but it is there (mdbus) Feb 11 20:02:25 mrmoku, the version is correct (I assume) Feb 11 20:02:33 because mdbus shows it correctly Feb 11 20:04:44 /usr/share/dbus-1/system-services/phoneuid-contacts.service Feb 11 20:04:46 mrmoku, Feb 11 20:04:52 you forgot to change these ^ Feb 11 20:10:31 TAsn, about "everythign works fine now": i have no service on most boots (->race condition?!) with current -unstable Feb 11 20:10:51 JesusMcCloud, wasn't it like that in testing? Feb 11 20:11:06 TAsn, with testing i only had the supend problem Feb 11 20:11:11 i c Feb 11 20:11:11 :| Feb 11 20:12:51 TAsn: they are for dbus-activation... Feb 11 20:13:00 don't need them Feb 11 20:13:05 2010.02.11 22:12:03.558496 [libphone-ui] MESSAGE: _get_signal_strength_callback: error -1: The name org.freesmartphone.ogsmd was not provided by any .service files Feb 11 20:13:06 setting gsm resource enabled via mdbus -s doenst wor either.. i'll pastebin logs in a sec Feb 11 20:13:09 :( Feb 11 20:14:35 it registered :P Feb 11 20:15:10 org.shr.phoneui - Rocks ;P Feb 11 20:15:23 mrmoku, fixed it? or do you want me to fix it? Feb 11 20:15:26 it's a 2sec fix Feb 11 20:15:46 fix what? Feb 11 20:16:14 TAsn: what fix? Feb 11 20:16:29 those services file Feb 11 20:16:33 s Feb 11 20:16:42 we don't use dbus-activation Feb 11 20:16:58 so why was it needed by phonefsod? Feb 11 20:17:05 and didn't work without it? Feb 11 20:17:34 org.freesmartphone.ogsmd was not provided by any .service files Feb 11 20:17:44 see? same with ogsmd... as long it's not up yet Feb 11 20:17:45 phoneuid/data/dbus-1/system-services Feb 11 20:17:55 we do ship them... Feb 11 20:18:26 yeah, fix them... but they should not cause any trouble Feb 11 20:18:40 TAsn, (maybe mrmoku, mickey|bbl aswell), http://pastebin.org/90266 Feb 11 20:18:43 TAsn: btw. multi value fields are BOXED Feb 11 20:18:52 mrmoku, ic :) Feb 11 20:19:47 SHR: 03tom 07phoneuid * rd6960ca0e8a1 10/ (8 files in 2 dirs): Fixed the service files to conform with the new bus-name renaming Feb 11 20:19:59 JesusMcCloud: do you have all /etc/freesmartphone/*.conf removed after opkg upgrade? Feb 11 20:20:15 TAsn, i have to correct myself: enabling gsm resource doesnt work it just hangs Feb 11 20:20:21 JaMa, i'll check Feb 11 20:20:47 JaMa, YES Feb 11 20:20:56 no conf left Feb 11 20:20:57 JesusMcCloud: and have you rebooted? Feb 11 20:21:30 this is why i think it might be a race condition, because it woks sometimes, but most of the time doesnt... Feb 11 20:21:48 the'res also a bug about this on shr trac Feb 11 20:21:51 JesusMcCloud, # Feb 11 20:21:51 2010.02.11 21:11:40.85 ogsmd.modems.ti_calypso INFO Requesting new channel from 'fso-abyss' Feb 11 20:21:54 I only get that Feb 11 20:22:02 and then nothing for a lot of time Feb 11 20:22:07 (I give up before I see anything else) Feb 11 20:22:51 JesusMcCloud: it worked for me in only about 20% before that libfsotransport downgrade, but after that I would say 90% (as I don't call with fr, so maybe sometimes I haven't noticed) Feb 11 20:23:12 mrmoku, A confession: I'm waiting to see exactly how you do it in order to copy it to other places :P Feb 11 20:23:21 I know theoretically where to look and what to do Feb 11 20:23:35 but the semantics (like knowing it's a box) I don't. Feb 11 20:23:44 JaMa, is the downgrade exeprimental or is it in feeds? Feb 11 20:23:58 JesusMcCloud: in feeds for about a week Feb 11 20:24:04 i have all upgrades Feb 11 20:24:09 i have the same Feb 11 20:24:15 JesusMcCloud: it has higher PV so it should be already installed for you.. Feb 11 20:24:18 dos1, same like who? Feb 11 20:24:30 kk... Feb 11 20:24:30 gsm resource hangs sometimes Feb 11 20:25:12 mrmoku, anyhow, http://shr-project.org/trac/ticket/881 - did you fix it or not? (comment says you are working on it, but it's 3 wks old) Feb 11 20:25:34 dos1, "sometimes" ... for me it's like most of the time (again not using latest unstable build, but all upgrades including today) Feb 11 20:26:00 dos1, any idea for a workaround apart from rebooting? Feb 11 20:26:00 JesusMcCloud: well, the same here, i'm just rebooting rarely so when it works, it's booted for few days ;) Feb 11 20:26:15 JesusMcCloud: you can try to disable fsodatad, fsotdld, fsonetworkd (move config somewhere will do) Feb 11 20:26:58 dos1, have you ried the proposed "fix" a link from the ticket comments point to? it did not work for me... Feb 11 20:27:08 no Feb 11 20:27:21 JaMa, woudl oyu mind to explain why this *could* help? Feb 11 20:28:42 TAsn: well that one in that form is fixed Feb 11 20:28:56 but as you know I cannot merge it into master Feb 11 20:29:02 so I did not close it Feb 11 20:29:03 mrmoku, good enough. Feb 11 20:29:16 it's ok to close it Feb 11 20:29:27 otherwise it'll just annoy you Feb 11 20:29:34 and you won't be able to work and look on other stuff Feb 11 20:29:35 btw. I think I just have to guess what type that boxed thing actually is... Feb 11 20:29:45 the main reason why not to close it, is users reporting about it again Feb 11 20:29:51 dos1, JaMa i found a workaround: (but i dont understand it, please try to reproduce) when gsm hangs, (and maybe when the fist otimed msgs are visible in frameworkd.log) DISable gsm resource Feb 11 20:29:54 yeah Feb 11 20:29:59 mrmoku, there are macros to check the types Feb 11 20:30:10 when I was testing it (with that libfsotransport which was failing for me often enouch for testing) then I've seen also channel requests from IIRC fsotdld if there is some race in fso-abyss then it *could* help IMHO Feb 11 20:30:12 well... which one Feb 11 20:30:26 TAsn: there are macros to check the type of a gvalue Feb 11 20:30:27 mrmoku, but in this case it's not as relevant (ticket) Feb 11 20:30:36 but the gvalue is of type boxed Feb 11 20:30:36 mrmoku, oh, it's not a gvalue? Feb 11 20:30:43 it is. Feb 11 20:30:45 ok Feb 11 20:30:47 so i use g_value_get_boxed to get a gpointer Feb 11 20:30:57 JesusMcCloud: just disable? or disable and then reenable? Feb 11 20:30:57 and I don't know what that gpointer points to :P Feb 11 20:31:06 just disable worked here... Feb 11 20:31:15 i'm rebooting and will try again Feb 11 20:31:45 mrmoku, to a boxed item :P Feb 11 20:31:58 JesusMcCloud: you can try my fso debug screen config :) Feb 11 20:31:59 TAsn: it might as well be a char ** Feb 11 20:32:12 JaMa, ? Feb 11 20:32:21 http://library.gnome.org/devel/gobject/unstable/gobject-Type-Information.html#G-TYPE-BOXED:CAPS Feb 11 20:32:39 thats where all my information is from :P Feb 11 20:33:05 http://library.gnome.org/devel/gobject/unstable/gobject-Boxed-Types.html Feb 11 20:33:44 JesusMcCloud: mmt Feb 11 20:34:47 * mrmoku tries GList first :P Feb 11 20:35:17 mrmoku, not verifying the type is bad :P Feb 11 20:35:19 but ok Feb 11 20:35:20 whatever. Feb 11 20:36:10 tell me how to... and I will do it ;) Feb 11 20:37:10 JaMa, confrimed it: when gsm hangs, just disable it (ifaik frameworkd will restarti it then) Feb 11 20:37:42 JesusMcCloud: http://jama.homelinux.org/org.openembedded.shr.images/fso-debug/ Feb 11 20:37:43 basically i diable it an voila, phoen starts to register and i can place calls Feb 11 20:38:04 JesusMcCloud: hmm interesting.. I'll try it if I can reproduce "gsm hang" Feb 11 20:38:18 just reboot :-P Feb 11 20:38:48 JesusMcCloud: well I rebooted about 30 times in last week Feb 11 20:38:55 dos1, would u mind to try aswell? Feb 11 20:39:16 btw how exactly do you disable it? Feb 11 20:39:46 org.freesmartphone.Usage.SetResourcePolicy ? Feb 11 20:39:47 JaMa, sec (having a phone conversation atm) -> i'll copypaste Feb 11 20:39:52 yup Feb 11 20:39:59 mrmoku, no idea :P Feb 11 20:40:02 but at least add a FIXME Feb 11 20:40:55 ok I have to finis watching LOTR3 :) Feb 11 20:41:01 s/finis/finish/ Feb 11 20:41:02 JaMa meant: ok I have to finish watching LOTR3 :) Feb 11 20:41:41 dammit gsm hangs again Feb 11 20:42:26 JaMa, you should Feb 11 20:42:54 (or i disabled gsm one time too often...) Feb 11 20:43:45 TAsn: don't worry I'm also building lite image without all modules for test :) Feb 11 20:43:52 JaMa, :) Feb 11 20:46:05 mrmoku, btw, I think we should allow passwordless logins again Feb 11 20:46:25 Now that we have the wizard that tells the users they should change their password Feb 11 20:46:32 JaMa, ok there might be something more to it... maybe my fault... i didn't set gsm to auto after setting it to disabled... Feb 11 20:46:42 the only reason why anyone whould use passwordless login is when something goes wrong Feb 11 20:46:49 and in that case, it's a good thing to have. Feb 11 20:47:51 mrmoku: I agree with TAsn (as I already asked passwordless twice :)) Feb 11 20:49:01 dos1, ping Feb 11 20:50:57 hey guys Feb 11 20:51:04 what's the proper config I need to have Feb 11 20:51:08 in order to make frameworkd Feb 11 20:51:11 to run only opimd? Feb 11 20:52:52 TAsn: are you going to use fsogsmd? Feb 11 20:53:01 TAsn: frameworkd -s opimd Feb 11 20:53:07 dos1, thanks :P Feb 11 20:53:09 JaMa, no Feb 11 20:53:15 JaMa, deving opimd Feb 11 20:53:21 ok :) Feb 11 20:53:59 ok, sweet, I have opimd running Feb 11 20:54:01 thanks dos Feb 11 20:54:07 TAsn: also, i think commenting out exiting when org.freesmartphone.frameworkd bus is already claimed could be also good idea Feb 11 20:54:28 TAsn: so this way you can have running every subsystem except opimd in one instance, and opimd in second Feb 11 20:55:42 dos1, testing it on pc anyway Feb 11 20:55:50 so I don't need that Feb 11 20:56:16 dos1, if you can give me a couple of quick pointers about design Feb 11 20:56:17 that'd be great Feb 11 20:56:23 as I don't know nothing about opimd design atm Feb 11 20:56:35 I'll of course investigate, just maybe you can make this process faster :P Feb 11 20:57:33 sorry, not today... bug me on weekend with everything i already delayed :P Feb 11 21:00:01 JaMa, seems one needs to set GSM back to auto (i did it as soon as possible even before my phone was fully registered) Feb 11 21:00:27 anhow, gtg eat some schnitzel Feb 11 21:05:34 JaMa, TAsn: well... as everybody wants passwordless login as default... what should I say :P Feb 11 21:05:49 :P Feb 11 21:05:55 bah, opimd is so complicated :P Feb 11 21:06:08 will take time to learn (especially with my rusty python abilities) Feb 11 21:06:30 TAsn: btw. it is char ** :) Feb 11 21:06:35 TAsn: you can write fsopimd instead :) Feb 11 21:06:46 mrmoku, sweet :P Feb 11 21:06:57 JaMa, nah, we need to know what we want before we do thta Feb 11 21:07:09 furthermore, the only reason why my vala is not rusty Feb 11 21:07:14 is because I don't know vala at all :P Feb 11 21:07:42 yeah seen that python~right_api reasoning :) Feb 11 21:07:51 * JaMa likes scala now Feb 11 21:11:44 Does window.keyboard_mode_set(elementary.ELM_WIN_KEYBOARD_NUMERIC) actually make a numeric keyboard popup in the said window when I focus on an entry? Feb 11 21:12:31 TAsn: hopefully I will finish the contacts tomorrow... Feb 11 21:12:35 now it's bed time Feb 11 21:12:39 happy opimd hacking :P Feb 11 21:12:41 ciao Feb 11 21:12:42 and gnight all Feb 11 21:12:43 thanks Feb 11 21:26:47 night Feb 11 21:27:36 gnite Feb 11 21:47:41 contact-messages-dialer bottom buttons are fixed? Feb 11 21:51:25 GNUtoo, afaik yes Feb 11 21:51:43 ok Feb 11 21:52:10 thanks a lot I'll upgrade then Feb 11 23:04:31 freesmartphone.org: 03mickey 07msmcomm * rf74b951eadf2 10/msmvterm/src/ (commands.vala terminal.vala): msmvterm: add command line completion Feb 11 23:41:15 freesmartphone.org: 03mickey 07msmcomm * rc6cf03e03b72 10/libmsmcomm/vapi/msmcomm-1.0.vapi: libmsmcomm: vapi: add eventToString() function; also fix event callback to deliver int; an "event" can actually also be a response Feb 11 23:41:16 freesmartphone.org: 03mickey 07msmcomm * rd5733bedc604 10/msmvterm/src/ (main.vala terminal.vala): msmvterm: print event type Feb 11 23:46:53 TAsn: grep -R kernel-module- installed-package-sizes.txt | wc -l => 21, not so little but not so bad Feb 11 23:47:26 a lot better though Feb 11 23:47:28 (I think) Feb 11 23:47:38 TAsn: we can remove few more from DISTRO_EXTRA_RRECOMMENDS in shr.conf Feb 11 23:47:45 JaMa: i added some comments the the kernel modules wiki page Feb 11 23:48:03 TAsn: it's from 2.6.32.8.. so quite minimal already.. but I had move modules installed before Feb 11 23:50:50 do we really care about "X" ? from my POV I would divide modules only to *installed* and *in_feed* Feb 11 23:51:39 and the second is decided by defconfig which can be as big as possible (buildable) Feb 11 23:52:12 X should be remove from image but keep in feed Feb 11 23:53:17 Heinervdm: ah I see Feb 11 23:54:54 * JaMa sleep, gnite Feb 12 00:16:21 uhm Feb 12 00:16:33 mickeyl: du hast also nun ein terminal für das msm modem? Feb 12 00:16:47 um AT befehle zu geben Feb 12 00:16:52 ? Feb 12 00:18:42 msm spricht kein AT Feb 12 00:19:16 wir haben ein terminal, was Befehle engegennimmt und die in das Binärprotokoll umsetzt Feb 12 00:19:42 diese Befehle haben wohl ähnliche Granularität wie AT Feb 12 00:19:47 aber ansonsten nichts damit zu tun Feb 12 00:20:01 s/msm/msm_palm/ Feb 12 00:20:56 hmm, wrong language :) Feb 12 00:24:05 ich hab vorhin mal überlegt was man ändern müsste um die proxies in fso-glib mit Object.new zu erzeugen Feb 12 00:24:21 hab aber kein schöndes basisobjekt gefunde Feb 12 00:26:52 errm Feb 12 00:26:59 * mickeyl notes at the channel name Feb 12 00:27:11 proxy sounds good Feb 12 00:27:21 we should probably not wait for ptitjes Feb 12 00:30:33 oops. i thought we are in #neo1973-germany Feb 12 00:31:56 hehe Feb 12 00:32:31 another would would be to define interfaces instead of object paths, but this would require some more magic Feb 12 00:51:52 freesmartphone.org: 03mickey 07msmcomm * r25fd84122a57 10/libmsmcomm/vapi/msmcomm-1.0.vapi: libmsmcomm: vapi: fix definition of response types Feb 12 00:51:53 freesmartphone.org: 03mickey 07msmcomm * rd04ed4b4a14d 10/libmsmcomm/src/response_system.c: libmsmcomm: response_system: properly compute IMEI Feb 12 00:52:04 g'night Feb 12 00:57:04 good night Feb 12 00:57:13 something went wrong: public struct GLib.HashTable Feb 12 00:58:27 freesmartphone.org: 03mickey 07msmcomm * ra9c3cdcd71f9 10/libmsmcomm/vapi/msmcomm-1.0.vapi: libmsmcomm: add questions for morphis Feb 12 00:59:35 playya_: I've got problems with merging my kernel into linux-2.6 of torvalds Feb 12 01:00:22 http://gitorious.org/htc-msm-2-6-32/leviathan-incoming/trees/msm-gnuconform Feb 12 02:44:44 freesmartphone.org: 03tom 07opimd-redesign * rbe95f18a8583 10framework/framework/subsystems/opimd/ (5 files): Feb 12 02:44:45 freesmartphone.org: opimd: Initial import of the redesign. Feb 12 02:44:45 freesmartphone.org: It's still a terrible mess, not everything works (iirc only adding and deleting contacts works) Feb 12 02:44:45 freesmartphone.org: This commit only includes cleanup and features for the contacts domain. Feb 12 02:44:45 freesmartphone.org: Fields pickle file is now stored in the db Feb 12 02:45:22 freesmartphone.org: 03tom 07opimd-redesign * r7c6e1db1846a 10framework/framework/subsystems/opimd/db_handler.py: opim: added a file I forgot to add in the previous commit Feb 12 02:45:25 phew Feb 12 02:45:28 * TAsn wants to die Feb 12 02:45:33 it's 4:50 :P Feb 12 02:45:56 And I was stuck for 2 hours on a stupid tiredness bug :P Feb 12 02:46:11 will probbaly continue working on it on monday night and not before :P Feb 12 02:46:13 anyhow, gn. **** ENDING LOGGING AT Fri Feb 12 02:59:56 2010