**** BEGIN LOGGING AT Thu Apr 01 02:59:57 2010 Apr 01 06:22:03 heh bug in ffalarms. 8) Apr 01 07:12:33 morning Apr 01 07:38:45 morning Apr 01 07:39:51 TAsn: wrt profile.. someone already reported that something changed, then mrmoku said that with fresh image he doesn't see the change, I tried to find profile-changing commit and failed and I haven't seen any change on my neo.. Apr 01 07:50:35 TAsn: newest EFL building.. Apr 01 07:52:04 there is newer 2.6.32 uImage-2.6.32.10-oe1+gitr5+*-r5-om-gta02.bin with reenabled bluetooth, I haven't tested it (fearing to reboot remotely and loosing ssh conn for whole day), someone please test it :) Apr 01 07:52:35 morning Apr 01 07:54:03 JaMa: if you refresh my memory how to use nandwrite... Apr 01 07:54:26 it was /dev/mtd4 ? Apr 01 07:55:25 ahh... mtd3 Apr 01 07:55:49 JaMa: /etc/profile defently changed, i've just checked Apr 01 07:55:55 linux-openmoko.inc tells the whole story :) Apr 01 07:57:57 * mrmoku rebooting Apr 01 07:59:29 Old profile: http://shr.pastebin.com/mKPH0DbT new profile: http://shr.pastebin.com/nmHAvWwN Apr 01 07:59:40 It changed with base-file on 25.3. Apr 01 08:01:22 http://cgit.openembedded.org/cgit.cgi/openembedded/commit/?id=5f9d302b5de9c0a17bd3aab739d4a05984f9cb08 but there is nothing wrong in this commit: Apr 01 08:01:51 JaMa: heh... it says: Wacom driver registered :P Apr 01 08:03:23 but it boots and registeres Apr 01 08:03:48 mrmoku: yeah wacom was reenabled too (not sure why, maybe the Kconfig option was changed and it defaults to Y) Apr 01 08:04:18 JaMa: from nand it boots and works fine so far Apr 01 08:04:32 mrmoku: some other options were enabled already in defconfig also because of newer udev (and it's requirements which were forced in linux.inc before) Apr 01 08:04:39 mrmoku: thanks Apr 01 08:06:14 Heinervdm: and 3 commits before is just kaeilos ASCII art changes :) Apr 01 08:06:33 mrmoku, JaMa: we are setting things in profile in shr-image.bb Apr 01 08:06:38 shr-image.inc Apr 01 08:06:58 we should add our own profile to that package Apr 01 08:09:42 Heinervdm: I agree, shr-image tweaks should be moved to recipes where possible Apr 01 08:10:28 Heinervdm: as you lost those tweaks whenever you update recipe providing tweaked file from feeds (maybe that's the case of task-base PR bump and mrmoku's fine new image Apr 01 08:10:59 Ok, then i will move profile from shr-image to base-files Apr 01 08:17:31 Heinervdm: thanks for reply on thone :) I noticed to late and replied almost the same :) Apr 01 08:17:41 * JaMa needs morninig coffee Apr 01 08:20:05 YAY \o/ Apr 01 08:20:44 colleague resoldered the capacitor on my A7 next to the mic and call quality on remote side increased by a few orders of magnitude! Apr 01 08:21:38 :) Apr 01 08:26:35 ohh, coffee! Apr 01 08:39:59 blindcoder: Was it not soldered correctly or was it the buzz fix? Apr 01 08:45:31 TeLLuS: that's "factory buzzfix" but something went wrong and so the cap needed partial resoldering. Apr 01 08:46:42 blindcoder: cool :) Apr 01 08:47:26 * spaetz sighs. wants a high quality phone too. Apr 01 08:47:57 TeLLuS: not soldered correctly Apr 01 08:48:29 there's still some light buzz according to the called person, but that might as well have been the A/C Apr 01 09:01:32 blindcoder: interesting.. I have more complaints about audio quality on my A7+ than I did on the A5.. :S Apr 01 09:01:38 A6* Apr 01 09:01:58 I wonder if I just need to play with the settings some more, or whether something's wrong Apr 01 09:03:58 hi... Apr 01 09:04:15 mesa-dri is failing to build, undefined reference to 2 egl functions Apr 01 09:05:54 daniele_athome: rebuild glamo Apr 01 09:07:03 Weiss: I have tested my phone with help from Doc Apr 01 09:07:22 Weiss: and it turned out that the capacitor to the left of the microphone was not soldered correctly Apr 01 09:07:48 Weiss: we tested with a multimeter's resistance testing, set to beep on short circuit Apr 01 09:08:30 Heinervdm, what do you mean by glamo? Apr 01 09:08:30 Weiss: pressing black on ground and red to the right (closer to mic) side gave a short, but pressing lightly against the capacitor disconnected it from the board -> no sound, infinite resistance Apr 01 09:08:46 the xorg driver? Apr 01 09:08:58 Weiss: resoldering the capacitor now gives way better quality Apr 01 09:09:03 Weiss: all this on an A7 board Apr 01 09:09:07 daniele_athome: i don't remember the exact recipe, i think it was xf86-video-glamo Apr 01 09:09:18 so the xorg driver :) Apr 01 09:09:39 daniele_athome: can you paste the exact error? probably libdrm_glamo is the one that needs recompiling Apr 01 09:09:55 or something in the horrible dependencies of Xorg/Mesa Apr 01 09:10:58 Weiss: http://pastebin.com/PWPDp9B2 Apr 01 09:16:38 TAsn: contact app is case sensitive in sorting, is it a feature or a bug? :) Apr 01 09:16:44 Heinervdm, Weiss: xf86-video-glamo requires mesa-dri to build Apr 01 09:17:19 still same errors Apr 01 09:17:29 freesmartphone.org: 03mok 07cornucopia * r477510fedf41 10/libfsoframework/fsoframework/interfaces.vala: Apr 01 09:17:30 freesmartphone.org: fsoframework/intervaces.vala: add Preference and PIM namespaces Apr 01 09:17:30 freesmartphone.org: Signed-off-by: Klaus Kurzmann Apr 01 09:18:40 Heinervdm, daniele_athome: that another issue (glamo had mesa-dri in rdepends and rebuild changed that to make do_rootfs happy), daniele show me log Apr 01 09:19:00 ah I see Apr 01 09:19:05 JaMa, the error part is pasted Apr 01 09:19:08 http://pastebin.com/PWPDp9B2 Apr 01 09:19:40 and whole log? Apr 01 09:19:47 ok i'll paste it Apr 01 09:21:08 JaMa: http://pastebin.com/c70YVN28 Apr 01 09:23:46 Heinervdm: who requested that the timeout settings are asorted according to value in shr-settings? Apr 01 09:23:50 daniele_athome: strange log.. have you tried to -c clean -c build? Apr 01 09:24:04 I found it more intuitive when it was ordered logically along the line of states it followed Apr 01 09:24:05 JaMa, yes of both xf86-video-glamo and mesa-dri Apr 01 09:24:19 no biggie though. Apr 01 09:24:25 spaetz: isn Apr 01 09:24:35 't it sorted no in order they appear? Apr 01 09:25:28 I think it was sorted from the shortest timeout to the longest timeout but not in the order the steps are followed. I will recheck before I complain though :) Apr 01 09:25:35 daniele_athome: and this log is from clean rebuild? Apr 01 09:25:54 JaMa, no the log is from build process -after- clean has been executed Apr 01 09:26:00 they are 2 different commands Apr 01 09:26:28 spaetz: it was sorted in the order the states were returned by FSO Apr 01 09:26:43 daniele_athome: I'm not sure I understand you, but "Nothing to be done for `default'." doesn't look like build after clean Apr 01 09:26:45 mmh, ok. Apr 01 09:27:11 bb -c clean xf86-video-glamo; bb -c build xf86-video-glamo Apr 01 09:27:24 i'll make it again but I assure i did that Apr 01 09:28:01 s/xf86-video-glamo/mesa-dri/ Apr 01 09:28:12 i'm doing both now Apr 01 09:28:16 glamo is important only after mesa-dri builds ok Apr 01 09:28:20 s/doing/cleaning/ Apr 01 09:28:21 daniele_athome meant: i'm cleaning both now Apr 01 09:35:28 the "fixed in unstable" is really helpful. Apr 01 09:35:33 going through now Apr 01 09:35:58 JaMa, log is too large for firefox textarea Apr 01 09:35:58 :S Apr 01 09:36:44 can I send you the log by e-mail JaMa? Apr 01 09:38:03 hehe, sure Apr 01 09:38:16 daniele_athome: btw do you know wgetpaste ? Apr 01 09:38:40 JaMa, no i used pastebin utility in ubuntu but doesn't work anymore since last pastebin upgrade Apr 01 09:39:15 JaMa, paste your e-mail in query pls :) Apr 01 09:39:17 ah wgetpaste allows you to specify pastebin service.. and default changed to pocoo.org after pastebin upgrade :) Apr 01 09:42:32 JaMa, i sent the log Apr 01 09:44:08 daniele_athome: seen that, btw what version of libdrm do you have? Apr 01 09:44:42 JaMa, you mean in my system? or in OE? Apr 01 09:47:35 daniele_athome: in OE Apr 01 09:48:37 vanous123, I don't remember, because we did discuss it. Apr 01 09:48:55 TAsn: :) Apr 01 09:49:19 JaMa, is efl there? :P Apr 01 09:49:55 TAsn: i would opt for case insensitive if i had the choice Apr 01 09:50:33 IIRC we had a good reason why we used case sensitive, but I'w probably wrong. :P Apr 01 09:50:43 as it was case insensitive at one point in the past. Apr 01 09:51:25 hmm Apr 01 09:51:38 cannot see the user case... Apr 01 09:56:50 mrmoku: i forgot to add the view to ContactListData because of that win was NULL Apr 01 09:57:07 mrmoku: now it doesn't crash anymore, but the inwin isn't shown Apr 01 09:58:06 Heinervdm, try evas_object_show :P Apr 01 09:58:29 evas_object_show(layout); Apr 01 09:59:16 or do i have to do evas_object_show(inwin); too? Apr 01 10:00:26 is enabling xim in ecore a risk? as in can it break stuff? Otherwise, I am tempted to not change it for now. Apr 01 10:01:46 spaetz: we haven't seen any bugs from it till now Apr 01 10:01:47 Heinervdm, just see what we did in other places Apr 01 10:01:54 but I think you have to show the inwin too Apr 01 10:01:59 not sure though Apr 01 10:02:02 easy to check, sec. Apr 01 10:02:02 Heinervdm: fine :) Apr 01 10:02:15 http://heinervdm.dyndns.info/shr-git/libphone-ui-shr/src/view/sim-manager-view.c Apr 01 10:02:34 Heinervdm, ok, you don't ;| Apr 01 10:02:41 (need to show the inwin) Apr 01 10:02:50 Offtopic: how can i set the corret mimetype? Apr 01 10:03:34 Heinervdm, why would you want? It's great that the browser shows it and doesn't ask me to open it with an external app. Apr 01 10:04:39 TAsn: yes, but it shows it as HTML and not as plain text Apr 01 10:04:55 so formating is lost Apr 01 10:05:02 or is it just in my cache? Apr 01 10:05:19 oh it's my cache Apr 01 10:05:19 JaMa: 1:2.4.18+gitr4565+b5aec2bd3df736216e86eae28e278172d3ba3362-r0.4 Apr 01 10:05:21 lol (xkcd.com) Apr 01 10:09:39 I just saw the numeric-alt keyboard yesterday for the first time. It is really nice! Apr 01 10:10:09 Got me thinking on whether we don't want to remove the numeric row in the terminal keyboard which would let us make the keys a bit bigger. Apr 01 10:11:47 spaetz, stop thinking and start doing! :P Apr 01 10:12:07 spaetz, What I think: Apr 01 10:12:16 TAsn: want feedback first. Apr 01 10:12:24 spaetz: how do you get the chars with shift then? Apr 01 10:12:33 mrmoku, numeric-alt Apr 01 10:12:34 ... Apr 01 10:12:41 JaMa, that's strange... that function exists in libegl source Apr 01 10:12:45 but not in libegl binary Apr 01 10:12:50 make terminal keyboard only with only commonly used chars in the terminal Apr 01 10:12:55 mrmoku: will need to do some research as to what keys would be missing. Apr 01 10:13:00 the function is between #ifdef EGL_KHR_image_base/#endif Apr 01 10:13:04 could this be the problem? Apr 01 10:13:06 but we could save a few keys at least Apr 01 10:13:17 spaetz, I wonder if there's a way to put different stuff in Shift vs Capslock Apr 01 10:13:20 they are realy tiny now. Apr 01 10:13:22 TAsn: that means you can't type email addresses with terminal :/ Apr 01 10:13:26 if so, we'll put caps in Caps-lock Apr 01 10:13:51 TAsn: e.g. I like the N800: pressing shift a longer time makes it turn on CAPS_LOCK Apr 01 10:13:52 and cool stuff in Shift Apr 01 10:14:02 not sure if that is possible Apr 01 10:14:06 spaetz, that's a feature request for the keyboard :P Apr 01 10:14:18 but it would save us the shift key :) Apr 01 10:14:22 err, caps-lock key Apr 01 10:14:30 for mrmoku's @ sign :) Apr 01 10:15:39 need to have a look at the keyboard first and think as how I could save a few keys. Will send a mail when I'm done :) Apr 01 10:15:58 just wanted to get first opinions on whether it would be work in vain. Apr 01 10:17:10 freesmartphone.org: 03mickey 07cornucopia * rd9aed33f404c 10/tools/mdbus2/src/main.vala: mdbus2: fix dumping empty containers; fixes FSO #525 Apr 01 10:18:55 TAsn: not yet NOTE: Running task 11409 of 14923 Apr 01 10:19:54 mickey|office, thanks :P Apr 01 10:20:10 np :) Apr 01 10:21:13 daniele_athome: yes... possible, but strange that only you're seeing this :/ Apr 01 10:22:27 yes... Apr 01 10:22:35 JaMa, could be an autotools related issue? Apr 01 10:23:00 JaMa, forcing that define in CFLAGS generates another error Apr 01 10:23:09 so I guess this is not the right way :( Apr 01 10:25:34 i just want to try 3d on qvga :( Apr 01 10:27:29 daniele_athome: can you try to build mesa-dri_git if that works for you? Apr 01 10:27:56 JaMa, just -c build mesa-dri_git ? Apr 01 10:28:25 daniele_athome: -b recipes/mesa/mesa-dri_git.bb will be faster Apr 01 10:28:36 ok Apr 01 10:29:31 building Apr 01 10:30:55 sorry, lots of SHR trac bugspam Apr 01 10:31:36 spaetz: not as much as i spammed the last days ;) Apr 01 10:32:28 hehe Apr 01 10:32:57 true. The only annoying bit is that I cannot change something from "fixed in unstable" to "fixed". I have to reopen first and then close it again. Apr 01 10:33:04 so more mails than necessary... Apr 01 10:33:12 yes Apr 01 10:33:49 otherwise it is very useful. Especially when you mention the commit links that fix this bug :-). Much appreciated. Apr 01 10:34:11 i searched for sth to add a State Apr 01 10:34:34 State like: accepted, assigned Apr 01 10:34:47 but i couldn't add one Apr 01 10:35:05 Heinervdm, sim-manager: I see you are working on exporting to sim Apr 01 10:35:18 TAsn: yes Apr 01 10:35:23 please do it from the sim-manager, i.e, don't add a button in the contacts app :) Apr 01 10:36:08 TAsn: yes, i'm already doing it like that ;) Apr 01 10:38:47 was just making sure :) Apr 01 10:38:57 Heinervdm, saw my comments about checking sim status? Apr 01 10:39:00 *simready Apr 01 10:39:17 JaMa, works... Apr 01 10:39:23 mesa-dri git built Apr 01 10:39:31 TAsn: from yesterday? Apr 01 10:39:35 and now? lol Apr 01 10:39:46 Heinervdm, or the day before, I lost track of time :) Apr 01 10:40:35 TAsn: i don't know anymore who said what :) Apr 01 10:40:42 there were a lot of suggestions Apr 01 10:40:52 hehe Apr 01 10:41:06 i deactivated caching for now Apr 01 10:41:52 who saves the default brightness levels nowadays? phonefsod, right? Apr 01 10:41:56 someone said adding a warning, that sim isn't ready should be enough Apr 01 10:42:09 TAsn: in feeds Apr 01 10:42:19 so is there a DBUS API for changing that? or has shr-settings to modify that config file manually. Apr 01 10:42:27 and someone said that it should be enough to check the GError Apr 01 10:42:36 http://trac.shr-project.org/trac/ticket/891 backlight brightness setting survives suspend, but not charging or reboot Apr 01 10:42:54 spaetz: dbus Apr 01 10:43:07 ok, then shr-settings need to call the right things... Apr 01 10:43:16 apparently it doesn't Apr 01 10:43:17 org.shr.phonefso.Usage.SetDefaultBrightness Apr 01 10:43:20 JaMa, thanks. Apr 01 10:43:36 lunch... finally :;P Apr 01 10:44:25 hmm no raster here... Apr 01 10:44:44 TAsn: with that EFL_SRCREV in illume configs.. is it OK to bump ie 1.1 -> 1.2? Apr 01 10:45:01 yep, shr-settings is still modifying self.display = getDbusObject (self.dbus, "org.freesmartphone.odeviced", "/org/freesmartphone/Device/Display/0", "org.freesmartphone.Device.Display" Apr 01 10:45:01 yes as I updated the configs Apr 01 10:45:10 but having efl_srcrev there is useless Apr 01 10:46:21 TAsn: ah configure.ac doesn't have illume-configs version just AC_INIT(enlightenment, 0.16.999.050, enlightenment-devel@lists.sourceforge.net) Apr 01 10:46:34 spaetz, you can marked as fixed and put a comment in the same reply in trac:P Apr 01 10:46:42 TAsn: maybe that should be updated too Apr 01 10:46:50 JaMa, that's only because someone was lazy :P Apr 01 10:46:53 build of linux-openmoko-2.6.32 fails Apr 01 10:46:53 ERROR: TaskFailed event exception, aborting Apr 01 10:46:54 ERROR: Build of /media/disk/martix/shr/shr-unstable/openembedded/recipes/linux/linux-openmoko-2.6.32_git.bb do_unpack failed Apr 01 10:46:54 ERROR: Task 0 (/media/disk/martix/shr/shr-unstable/openembedded/recipes/linux/linux-openmoko-2.6.32_git.bb, do_unpack) failed Apr 01 10:46:59 s/only/only there/ Apr 01 10:46:59 TAsn meant: JaMa, that's only there because someone was lazy :P Apr 01 10:47:19 cp: cannot stat `/media/disk/martix/shr/shr-unstable/openembedded/recipes/linux/linux-openmoko-2.6.32/defconfig': No such file or directory Apr 01 10:47:42 Martix_: do you have MACHINE="om-gta02" ? Apr 01 10:47:51 mickey|office: mdbus2 feature request: bash shell completion with TAB :-) Apr 01 10:48:52 JaMa: in which cfg? :-) Apr 01 10:49:34 mrmoku: I just tried to set the default brightness, but that will not change /etc/phonefsod.conf, so it won't survive reboots. Right? Apr 01 10:49:42 env |grep MACHINE is plain Apr 01 10:49:47 Martix_: usually local.conf Apr 01 10:50:35 Martix_: there is defconfig only for om-gta02 (as it's only device where I tried that kernel) and its in recipes/linux/linux-openmoko-2.6.32/om-gta02/defconfig which is used only for MACHINE="om-gta02" Apr 01 10:50:40 no, just added Apr 01 10:53:05 JaMa, spaetz, mrmoku: PM. Apr 01 10:58:16 mrmoku: ok, thanks for the API. WOuld making phonefsod saving that value to the config file be acceptable? Apr 01 10:59:47 hi there Apr 01 11:00:17 i have an issue using intone and hope to find help here Apr 01 11:00:22 flashed SHR this morning Apr 01 11:00:41 i can only add single folders containing music, but the search seems not to be recursive :( Apr 01 11:00:46 what am i doing wrong? Apr 01 11:03:32 spaetz: sure, will do that Apr 01 11:03:48 actually I thought I have done that :/ Apr 01 11:10:55 I just changed it and the config file stayed the same. let me retry. Apr 01 11:13:44 SHR: 03Sebastian 07shr-settings * rb6aed65934cf 10/shr_settings_modules/shr_display.py: shr_display: Use phonefsod API GetDefaultBrightness and SetDefaultBrightness Apr 01 11:13:49 I just pushed shr-settings to use phonefsod API for Get/SetDefaultBrightness Apr 01 11:15:53 mrmoku: just rechecked, changes are not saved in /etc/phonefsod at least Apr 01 11:16:10 I'm stuck building SHR-U with an error when trying to run ncurses do_install Apr 01 11:17:10 The log says "... arm-oe-linux-gnueabi/bin/ld: cannot find -lncurses" Apr 01 11:17:31 Then it tells me "arm-oe-linux-gnueabi-libtool: install: error: relink `libpanel.la' with the above command before installing it" Apr 01 11:18:28 but when I try to execute the given command it fails with "arm-oe-linux-gnueabi-libtool: line 7855: arm-oe-linux-gnueabi-gcc: command not found" Apr 01 11:19:19 Does anybody have an idea what I am missing? I'm pretty sure the gcc task allready did run successfully Apr 01 11:20:29 doon: if you wan't to run that on your own, you have to set your PATH correctly Apr 01 11:21:09 JaMa: thanks, now I successfully builded 2.6.32 Apr 01 11:21:19 ah okay, so I'll add the location of arm-oe-linux-gnueabi-gcc to my PATH Apr 01 11:21:25 and then run the command again Apr 01 11:21:54 doon: you can edit temp/run.do_install*, remove exit in fatal() function and "source temp/run.do_install" Apr 01 11:22:28 doon: this way you get exactly the same environment and cmdlines as bitbake is using Apr 01 11:22:47 JaMA: okay, I'll try that, thank you Apr 01 11:23:45 JaMa: and then I try to build via "make image" again ? Apr 01 11:24:45 no that source command would execute that task and if it finish successfully Apr 01 11:25:01 then it's better to close shell (discard modified environment) Apr 01 11:25:08 and start new one with make image Apr 01 11:25:18 SHR: 03tom 07libphone-ui-shr * r280195334f9b 10/data/idle_screen/ (background.png logo.png): Fixed a couple of usability issues Apr 01 11:25:28 SHR: 03tom 07libphone-ui-shr * r07bee925192f 10/ (9 files in 5 dirs): Merge branch 'master' of git+ssh://git.shr-project.org/libphone-ui-shr Apr 01 11:31:22 JaMa: Hmm I can't find that temp/run.do_install file. Where is it supposed to be? Apr 01 11:31:51 doon: in tmp/work/ARCHITECTURE/ncurses Apr 01 11:32:01 2.6.32 builded from shr-unstable OE tree: Cannot open root device "mmcblk0p2" or unknown-block(0,0) Apr 01 11:32:17 using u-boot. but no problem with 2.6.29 Apr 01 11:32:19 Martix_: increase rootwait/rootdelay Apr 01 11:32:25 ok Apr 01 11:32:31 JaMa: found it, thanks Apr 01 11:32:49 Martix_: read README http://build.shr-project.org/tests/mrmoku/kms/images/om-gta02/ Apr 01 11:43:38 noone here to help out with intone? Apr 01 11:50:11 JaMa: The source command fails, just like the make image command. The error is "/home/azrael/shr/shr-unstable/tmp/cross/armv4t/lib/gcc/arm-oe-linux-gnueabi/4.4.3/../../../../arm-oe-linux-gnueabi/bin/ld: cannot find -lncurses collect2: ld returned 1 exit status arm-oe-linux-gnueabi-libtool: install: error: relink `libpanel.la' with the above command before installing it" Apr 01 11:50:40 doon: check where is ncurses lib and if it's included in some -L param Apr 01 11:50:57 doon: look inside libpanel.la Apr 01 11:51:29 mickey|office: I tried it now Apr 01 11:51:32 *uggh* Apr 01 11:51:49 it really kills suspend if you change onto the vt where the xoutput is Apr 01 11:51:53 >_> Apr 01 11:52:15 I didn't see that coming Apr 01 11:52:17 :( Apr 01 11:55:57 SHR: 03seba.dos1 07shr-themes * r8f2a82d36337 10/shr-splash/shr-splash-theme-tux/ (boot.fbz shutdown.fbz): shr-splash: enhanced compression, files take little less space now and are loaded faster Apr 01 11:57:11 JaMA: "[..]/ncurses-5.4/lib/libncurses.la" is part of the dependency_libs variable in libpanel.so Apr 01 11:57:54 I mean libpanel.la, sorry Apr 01 11:58:44 JaMa: increasing root delay doesnt work, tried 7, 10, 15 seconds also with http://build.shr-project.org/tests/mrmoku/kms/images/om-gta02/uImage-2.6.32.10-oe1+gitr5+14be1091928fcce66812a85129768fb253b36420-r5-om-gta02.bin Apr 01 11:59:02 and libncurses.so is located int [..]/lib/.libs/libncurses.so Apr 01 11:59:10 I have Sandisk 16 GB uSD card Apr 01 11:59:19 Martix_: try also debug in kernel cmdline Apr 01 12:02:51 ARGH I'm blind! THe goggles, they do nothing! Apr 01 12:03:12 nice april fool's joke to show a Windows logo on my SHR phone :) Apr 01 12:06:11 :o Apr 01 12:06:20 JaMa: debug didnt help Apr 01 12:07:44 menu_1=Boot from microSD part2 (ext3+ext3): setenv bootargs ${bootargs_base} debug rootfstype=ext3 root=/dev/mmcblk0p2 rootdelay=15 ${mtdparts} ro; mmcinit; sleep 1; ext2load mmc 1 0x32000000 uImage2.bin; bootm 0x32000000 Apr 01 12:07:49 is right? Apr 01 12:10:08 well i experimented Apr 01 12:10:11 with intone Apr 01 12:10:28 could it be, that it has problems with paths containing spaces and with locations on card? Apr 01 12:10:46 JaMa: the ncurses library is in tmp/work/armv4t-oe-linux-gnueabi/ncurses-5.4-r19/ncurses-5.4/lib/.libs Apr 01 12:11:21 JaMa: In libpanel.la the only -l param is -L/home/azrael/shr/shr-unstable/tmp/staging/armv4t-oe-linux-gnueabi/usr/lib Apr 01 12:11:57 JaMa: Should I edit libpanel.a or add a symlink to staging/[..]/usr/lib Apr 01 12:15:35 Good day all , How to use start Apr 01 12:15:57 Good day all , How to use stardict dictionaries with babiloo Apr 01 12:18:49 ~nf Apr 01 12:18:50 The #openmoko-cdevel Newsflash Bulletin Board. (continued at ~NF2. For help see ~NF-help) - - - Recommended URLs and channels(chanlogs etc):see ~RL - - - NEWSFLASH [2010-03-28 21:09 UTC]:buildhost got rebooted... new RAM will arrive beginning of next week Apr 01 12:21:12 ~nf2 Apr 01 12:21:13 (here you'll find additional info, if newsflash on ~NF is too short to hold all the info) Apr 01 12:23:57 moin Apr 01 12:25:22 moin Apr 01 12:44:56 alabd: I don't know if anyone here has ever used stardict. You might need to send a mail to the shr user list and ask. Apr 01 12:48:50 alabd: why not use estardict instead of babiloo? same author IIRC, zeusone Apr 01 13:01:20 hmm, observing the mails in shr now i wonder, as after upgrade reboot and uname -a still shows: Apr 01 13:01:33 Linux om-gta02 2.6.29-rc3 #1 PREEMPT Sun Feb 28 04:21:56 CET 2010 armv4tl GNU/Linux Apr 01 13:01:55 so what is the new_kernel talk about? Apr 01 13:02:12 or are repos still not fully sinced yet? Apr 01 13:04:11 hmm, no-one? Apr 01 13:04:27 SHR: 03mok 07phonefsod * rde3eab1df168 10/src/phonefsod-dbus-usage.c: make default brightness setting persistent Apr 01 13:10:49 everything is fixed Apr 01 13:11:13 vanous123: you should see it after first reboot or if you lock the screen.. Apr 01 13:11:34 i locked the screen and see a very lovely window Apr 01 13:12:09 managed to live without it for last ten years, now it's here :) Apr 01 13:12:30 stability cost something :) Apr 01 13:12:37 JaMa: what about kernel? Apr 01 13:12:54 is it a fools day joke only? Apr 01 13:22:40 JaMa: thanks can we stardict dictionaries with estardict ? Apr 01 13:24:40 alabd: ahh, don't know (I was expecting that you mean estardict..) Apr 01 13:32:00 how to use stardict dictionaries with estardict and babiloo on shr ? Apr 01 14:00:31 has anyone used bluetooth headset with shr ? Apr 01 14:25:08 hehe, subtle, real subtle prank ;) Apr 01 14:43:12 blindcoder: :) Apr 01 14:43:27 mrmoku|away: thanks for making the brightness persistent. Apr 01 14:46:57 hi Apr 01 14:47:38 where does the default profile get set at first boot? Apr 01 14:47:51 s/default/default enlightenment Apr 01 14:48:44 ~/.e I assume? Apr 01 14:49:31 ah but we have nothing in /home/root at first boot Apr 01 14:49:45 ? Apr 01 14:49:51 ahh that. no clue there, sorry Apr 01 14:49:59 misunderstood you Apr 01 14:50:07 else I'll use something else than enlightenment Apr 01 14:51:00 (if I can't debug that) Apr 01 14:51:07 because having no GUI is no fun Apr 01 14:51:30 ahh phoneuid-wrapper is killing my neo.. Apr 01 14:51:56 I've sshed inside the freerunner and found that : Apr 01 14:52:03 when dbus isn't running.. then launching phoneuid as quickly as possible won't help... bad wrapper! spank! Apr 01 14:52:19 inside /usr/bin/enlightenment_start.oe there is: Apr 01 14:52:33 /etc/enlightenment/default_profile Apr 01 14:52:40 that is sourced Apr 01 14:52:46 but that file is not there Apr 01 14:52:58 so I wonder how it set the default profile at fist boot Apr 01 14:54:21 maybe there is something in: Apr 01 14:54:30 /etc/X11/Xsession.d Apr 01 14:55:25 mmm no...still a mystery Apr 01 14:55:26 /usr/bin/x-window-manager -> /usr/bin/enlightenment_start.oe Apr 01 14:55:58 JaMa, heeh yeah, I know :P Apr 01 14:56:07 (bad wrapper) Apr 01 15:06:00 Folks, you should be seriously afraid of being subject to copyright infrigement investigation now. Apr 01 15:07:59 PaulFertser: i read that SCO lost ? :) Apr 01 15:08:25 wpwrak: http://scap.linuxtogo.org/files/80532009bb5ebb13db6f83c4e7383358.png Apr 01 15:12:14 PaulFertser, thanks. Apr 01 15:12:37 PaulFertser, though isn't ok in april's fool? :) Apr 01 15:13:17 SHR is now really Stable & Hybrid ;-) Apr 01 15:13:37 hehe Apr 01 15:13:50 SHR: 03tom 07libphone-ui-shr * r24e97267eeb5 10/data/idle_screen/ (background.png logo.png): Revert "Fixed a couple of usability issues" Apr 01 15:14:25 ^ lol Apr 01 15:14:53 PaulFertser, jokes aside, I don't think it's considered copyright infringement if it's not "used" to "better" our product Apr 01 15:15:14 it's like including the icon on tv/newspaper Apr 01 15:15:17 without approval Apr 01 15:15:28 PaulFertser: ;-)) Apr 01 15:15:32 And anyhow, dos is underage, we could blame it on him :P Apr 01 15:15:49 TAsn: revert also -tux logo then.. Apr 01 15:15:50 TAsn: looks like fair use to me. a parody. Apr 01 15:15:57 wpwrak, that's what I think. Apr 01 15:16:27 wpwrak, but if you play with the devil, you are bound to get burnt :P Apr 01 15:16:32 TAsn: all i know about that matters is that those are utter unbelievable crap. I do not think it's in fact can have any consequencies in the shr case :) Apr 01 15:16:33 burned* Apr 01 15:16:33 TAsn: but only few months left :P Apr 01 15:16:57 dos1, let's hope they won't drap it in court. Apr 01 15:17:00 drag* Apr 01 15:17:16 PaulFertser, so we'll wait a day Apr 01 15:17:28 JaMa, build only after 00:00 :P Apr 01 15:18:47 hopefully no SHR user is working in MS law department sending us copyright infringement now as his april's joke :) Apr 01 15:19:49 building now.. rsyncing later.. Apr 01 15:19:51 haha :P Apr 01 15:20:13 SHR: 03tom 07shr-themes * ra878c4b6dd82 10/shr-splash/shr-splash-theme-tux/ (boot.fbz shutdown.fbz): Revert "shr-splash: enhanced compression, files take little less space now and are loaded faster" Apr 01 15:21:05 Hey~! Apr 01 15:21:10 I had a long message :( Apr 01 15:21:13 ffs. Apr 01 15:21:23 evert "shr-splash: enhanced compression, files take little less space now and are loaded faster" That's not true, the images are now bloated and take a lot of RAM and CPU, furthermore, changing the pictures made my phone crash with BSoDs :| Apr 01 15:22:16 TAsn: if parody wasn't fair use, just imagine how the maker of this one would have gotten torn to pieces: http://members.liwest.at/mainzi/files/winnt40/winntneanderthal.gif Apr 01 15:23:07 wpwrak, Nah, I guess they know that suing him would mean even more publication Apr 01 15:23:14 and then everyone would know that windows sucks :) Apr 01 15:23:28 TAsn: yeah. can't let that secret out :) Apr 01 15:24:53 only 1% (according to internet stats) of the world population knows that, and only 98.9% know that subconsciously :| Apr 01 15:25:11 It's like the matrix, little know about it, but everyone know something is fishy :P Apr 01 15:28:47 ~seen leviathan Apr 01 15:28:49 leviathan <~quassel@ifw-public-dock-145-dhcp.ethz.ch> was last seen on IRC in channel #openmoko-cdevel, 3h 36m 32s ago, saying: ':('. Apr 01 15:33:08 TAsn: they probably just blame computers in general :) Apr 01 15:33:28 wpwrak, they officially do "Buy new hw" Apr 01 15:39:56 TAsn: the traditional solution for a software company - blame hardware :) Apr 01 15:40:12 yeah :P Apr 01 15:40:23 FINALLY Apr 01 15:40:30 After essentially HOURS Apr 01 15:40:46 I can finally install arch on my laptop Apr 01 15:41:11 (Had windows because it wasn't mine first, it was my father's with windows and I promised I'll copy stuff around bla bla) Apr 01 15:41:25 I didn't know right clicking a file can take >10secs Apr 01 15:41:26 :| Apr 01 15:43:07 congrats Apr 01 15:44:35 That was very painful ;P Apr 01 15:53:03 freesmartphone.org: 03mickey 07cornucopia * ra750fdd0d5f1 10/fsodeviced/src/plugins/kernel_input/plugin.vala: fsodeviced: kernel_input: initialize the hold timeout Apr 01 15:53:04 freesmartphone.org: 03mickey 07cornucopia * r1ddaff792ec6 10/fsodeviced/src/plugins/kernel_input/ (Makefile.am plugin.vala): fsodeviced: kernel_input: refactor input handling Apr 01 15:53:05 freesmartphone.org: 03mickey 07cornucopia * rb7a5ccd8aa2b 10/fsodeviced/configure.ac: fsodeviced: enable debug build Apr 01 16:03:00 mickeyl, hi Apr 01 16:03:06 hi GNUtoo Apr 01 16:03:25 I'll try harder for the backlight Apr 01 16:03:31 but I'm not shure it's bl's fault Apr 01 16:03:37 the resume issue can only be fixed if we implement two additional ioctls Apr 01 16:03:55 ah? Apr 01 16:03:57 which ones? Apr 01 16:04:33 sorry, one ioctl Apr 01 16:04:34 FBIOBLANK Apr 01 16:04:52 I'll look Apr 01 16:05:01 this we need to forward to the panel init/deinit Apr 01 16:05:16 otherwise the panel will only be reinitialized when you're on a console VT Apr 01 16:05:21 (which is not what we want :) Apr 01 16:07:41 apart from the panel reinit, suspend/resume works fine Apr 01 16:07:51 so I implement FBIOBLANK in msmfb.c? Apr 01 16:07:55 yeah Apr 01 16:08:04 but I wonder Apr 01 16:08:13 I've no access to bl inside that fiel Apr 01 16:08:15 *file Apr 01 16:08:35 the backlight level is another story Apr 01 16:08:43 less important since we can quickly set it after resume Apr 01 16:08:47 ok Apr 01 16:08:48 so noone will notice a flash Apr 01 16:08:59 so I didn't understood well what I should do Apr 01 16:09:35 ok, let me recap the problem Apr 01 16:09:53 problem is: after resume, the panel will not automatically come back Apr 01 16:10:10 right now it only comes back if you're on the console VT or on a VT that is set in TEXTMODE Apr 01 16:10:27 since we use another VT for both framebuffer or when in X, resume will fail to reinit panel Apr 01 16:10:42 the right way to fix this is to implement the FBIOBLANK ioctl Apr 01 16:11:06 then we can issue FB_BLANK_UNBLANK when we want the panel to come back Apr 01 16:11:29 (and we can power down the panel fully, even without suspend) Apr 01 16:11:52 so the from msmfb.c you need to get the handle to the panel and then call init/deinit depending on the value of FBIOBLANK Apr 01 16:12:22 mickeyl, under xf86-video-msm and xf86-video-fbdev suspend-resume suspends the pannel and resume it correctly Apr 01 16:12:44 interesting, how is that working? Apr 01 16:12:53 mickeyl, btw I saw your commit message about multistate: you should never edit defconfig by hand Apr 01 16:13:02 mickeyl, don't know Apr 01 16:13:04 if it's switching the VTs, then this is sheer luck Apr 01 16:13:13 switching the VTs will bring the panel back Apr 01 16:13:16 but will flicker Apr 01 16:13:22 mickeyl, I bet it does some switching Apr 01 16:13:27 but I didn't really look Apr 01 16:14:04 for defconfig by hand: look at a Kconfig and you'll see some dependencies,some things that pull other config in etc... Apr 01 16:15:26 yeah Apr 01 16:16:02 It doesn't flicker for me Apr 01 16:16:09 or maybe I don't see it Apr 01 16:21:32 mrmoku|away: phoneui again dislikes Cell phone. But now in multiple phone selection dialog. If contact has several Cell phones none of them shown. Apr 01 16:23:05 GNUtoo: is X running on a dedicated VT? Apr 01 16:23:12 or taking over vt1 ? Apr 01 16:23:27 mickeyl, don't know...how can I check? Apr 01 16:23:37 login via ssh Apr 01 16:23:44 run chvt Apr 01 16:24:07 that way you can switch vts Apr 01 16:24:15 and see on which is running what Apr 01 16:24:15 ok Apr 01 16:24:23 first one is 1 Apr 01 16:24:30 usually where the console is Apr 01 16:25:42 knowing a bit oe it should be the second one...but I'll check Apr 01 16:26:31 (second one = second vt) Apr 01 16:26:39 booting Apr 01 16:28:10 ok: Apr 01 16:28:14 vt1 : console Apr 01 16:28:17 vt2 : Xorg Apr 01 16:31:14 ok, very strange. i'll build xorg and check what happens here Apr 01 16:31:22 you're using the OE kernel? Apr 01 16:31:27 my own Apr 01 16:31:31 hmm Apr 01 16:31:33 but I can try with oe kenrel Apr 01 16:31:36 yes, please Apr 01 16:31:38 just to make sure Apr 01 16:31:49 if by kenrel you meant defconfig oe + leviathan's kenrel Apr 01 16:35:00 yep Apr 01 16:36:15 lol -> http://scap.linuxtogo.org/files/80532009bb5ebb13db6f83c4e7383358.png Apr 01 16:38:43 mickeyl, also work Apr 01 16:38:58 but I've some strange things in dmesg with your config Apr 01 16:39:14 of the other day + the early suspend thing Apr 01 16:39:51 [ 71.039886] Unable to handle kernel paging request at virtual address 00010074 Apr 01 16:39:53 and Apr 01 16:40:12 a NULL pointer dereference Apr 01 16:40:14 never seen that Apr 01 16:40:16 ok Apr 01 16:40:25 maybe it's my rootfs and filesystem corruption Apr 01 16:40:25 how do our configs differ? Apr 01 16:40:31 we should really try to use the same configs Apr 01 16:40:42 otherwise debugging will only be harder :) Apr 01 16:40:47 mickeyl, no,we should use different config Apr 01 16:40:57 else we won't find the magick thing that makes everything work Apr 01 16:41:00 for instance Apr 01 16:41:01 heh Apr 01 16:41:12 leviathan says it works for him Apr 01 16:41:19 if he used oe config Apr 01 16:41:20 no, yesterday night he confirmed it Apr 01 16:41:29 he may never have found how to make it work Apr 01 16:41:43 mickeyl, ? Apr 01 16:41:53 he confirmed the vt resume issue Apr 01 16:42:01 and i thought that's what's killing his X resume Apr 01 16:42:03 but if you have it working Apr 01 16:42:06 it must be something else Apr 01 16:42:07 *sigh* Apr 01 16:42:24 i started a build of xorg Apr 01 16:42:28 lets see.. Apr 01 16:42:29 ok Apr 01 16:42:44 what vt resume issue did leviathan have exactly? Apr 01 16:43:02 it stays black Apr 01 16:43:05 when you macke Apr 01 16:43:08 chvt 2 Apr 01 16:43:14 and then try to suspend Apr 01 16:43:16 then resume Apr 01 16:43:37 somehow it then uses the same vt as the xdriver wants Apr 01 16:43:48 and the tty is faster then the driver Apr 01 16:43:57 so it cant restart the refresh thread Apr 01 16:43:59 ... -.- Apr 01 16:44:25 only solution I see would be to reserve the ressource for the driver Apr 01 16:44:37 before killing the thread for suspend Apr 01 16:44:38 ... Apr 01 16:45:51 ah tthere he is Apr 01 16:46:12 hmm Apr 01 16:46:26 ok, worth a try, although i think we should implement FBIOBLANK anyways Apr 01 16:46:46 that way you can disable/enable the panel, save power, without going into suspend Apr 01 16:46:52 may have its usecase Apr 01 16:46:52 I'll have to eat Apr 01 16:46:57 ~bon appetit Apr 01 16:46:59 somebody said bon appetit was smacznego. Guten Appetit. Eet Smakelijk. God Appetitt. Buon Appetito. Buen apetito Bom Apetite. buen apetito Smaklig måltid!. Hyvää ruokahalua. Bo Proveito Apr 01 16:47:00 thanks Apr 01 17:05:48 Hi, does someone know the adress of the shr opkg repo Apr 01 17:06:09 I saw the adress yesterday but I can't remember it.. Apr 01 17:07:11 build.shr-projec.org Apr 01 17:14:17 opk-repo.org ? Apr 01 17:21:49 ~seen Slyon Apr 01 17:21:51 slyon <~lukas@p4FDB37B2.dip0.t-ipconnect.de> was last seen on IRC in channel #openmoko-cdevel, 17d 2h 55m 21s ago, saying: '(your local e settings will be lost)'. Apr 01 17:26:51 meep, phoneui bug :( Apr 01 17:27:08 having two "mobile phone" entries causes the dialogue appearing when pressing "Call" to fsck up Apr 01 17:39:02 hi GNUtoo Apr 01 17:39:04 wb Apr 01 17:40:40 leviathan, thanks Apr 01 17:40:46 np Apr 01 17:40:55 leviathan, so I do not understand and have the issue you and mickeyl have Apr 01 17:41:16 I'll build a new image on my ETH account Apr 01 17:41:20 and will upload it Apr 01 17:41:25 nearly finished ^^ Apr 01 17:41:28 (I may not understand also because I'm tired) Apr 01 17:41:34 anyway Apr 01 17:41:50 leviathan, how do you pass the issue of fetching the android utils? Apr 01 17:42:06 should I retry to build an image from scratch? Apr 01 17:42:15 stock oe.dev Apr 01 17:42:24 SRCREV_pn-android-image-utils-native = "${AUTOREV}" Apr 01 17:42:25 :) Apr 01 17:42:29 in local.conf Apr 01 17:42:35 ok Apr 01 17:43:01 was already set in another conf file but I'll try Apr 01 17:43:48 btw is there an illume specialist here: how is the profile chosen at first boot? Apr 01 17:45:03 dunno ;-) Apr 01 17:45:11 [2010-04-01 11:08:58] Weiss: resoldering the capacitor now gives way better quality Apr 01 17:45:54 blindcoder: could you open a ticket against hardware, describing the issue, pleae? Apr 01 17:46:02 please Apr 01 17:46:24 DocScrutinizer: sure Apr 01 17:46:37 It's clearly a manufacturing bug we need to pester openmoko.inc about Apr 01 17:47:12 * DocScrutinizer curses them for never sending MP production samples to him, for QA Apr 01 17:48:27 in fact that's roundabout 5th time it seems like some unknown actor tries to make buzzfix fail Apr 01 17:49:13 DocScrutinizer: docs.openmoko.org I presume Apr 01 17:49:49 blindcoder: err? Apr 01 17:49:57 DocScrutinizer: to report the bug? Apr 01 17:50:10 openmoko trac Apr 01 17:50:10 or SHR trac? Apr 01 17:51:04 TICKET_CREATE privileges are required to perform this operation Apr 01 17:51:05 :-( Apr 01 17:51:17 err WHAT? Apr 01 17:51:52 DocScrutinizer, since your're here, whats the progress of (dummy) scripts? Apr 01 17:51:58 my just created account is not allowed to open tickets Apr 01 17:55:58 https://docs.openmoko.org/trac/newticket ? logged in? Apr 01 17:56:47 JesusMcCloud|afk: slow :-( I'll ping you later Apr 01 17:56:58 yes and yes Apr 01 17:57:15 * logged in as blindcoder Apr 01 17:57:15 * Logout Apr 01 17:57:15 * Preferences Apr 01 17:57:15 * Help/Guide Apr 01 17:57:21 oops, sorry :( Apr 01 17:57:26 roh: ping Apr 01 18:00:27 blindcoder: you answered any confirmation msg? dunno if there's such msg for newly created trac accounts Apr 01 18:02:56 thanks spaetz! for th eurl Apr 01 18:13:19 mickey|bbiab, do you want me to commit oe config for htcdream? Apr 01 18:13:28 I just sftped the config Apr 01 18:13:30 and .... Apr 01 18:14:04 let me check before Apr 01 18:14:53 GarthPS, np. I coded it :-) Apr 01 18:15:07 my shr image is borked too, there's a Windows like logo in my idle screen!!!! Apr 01 18:15:19 :P Apr 01 18:15:24 hehe Apr 01 18:15:28 ah no so sorry Apr 01 18:15:32 old defconfig Apr 01 18:17:08 mickey|bbiab, it works with stock oe + xf86-video-fbdev Apr 01 18:18:24 DocScrutinizer: no message here Apr 01 18:19:40 spaetz: I saw that! very usefull thx! Apr 01 18:26:36 DocScrutinizer: ah, now an email arrived (just about 30 minutes late) Apr 01 18:27:16 mickey|bbiab, it looks like FSO is not calling bluetoothd when requesting BT resource or setting it to ON. Apr 01 18:33:52 leviathan, you've also get the issue mickeyl described? Apr 01 18:34:50 mickey|bbiab, leviathan the null pointer and kenrel problems were due to the loading of old modules Apr 01 18:35:39 DocScrutinizer: #2335 Apr 01 18:48:06 I just installed todays SHR image. The keyboard does not show up when I try to log into web-pages using Midori. Apr 01 19:09:18 spaetz: hi Apr 01 19:09:33 you can look on nariwork-03.ee.ethz.ch Apr 01 19:09:56 I'm try'n to build there a recent image Apr 01 19:10:29 GNUtoo: yes Apr 01 19:10:34 I also have this issue Apr 01 19:10:46 and I also have an idea Apr 01 19:10:50 where it comes from Apr 01 19:10:51 ... Apr 01 19:10:53 ok Apr 01 19:11:04 leviathan, Unable to connect Apr 01 19:11:18 GNUtoo: for what? Apr 01 19:11:28 nariwork? Apr 01 19:11:40 its for ETH students/employees only ;) Apr 01 19:11:41 yes Apr 01 19:11:48 ok Apr 01 19:12:04 so what's the issue? Apr 01 19:12:11 I thought spaetz is also on ETHZ Apr 01 19:12:22 so basically he can take a look there Apr 01 19:12:23 :) Apr 01 19:12:45 ok Apr 01 19:13:27 leviathan, but it worked before? (for mickeyl's issue) Apr 01 19:13:43 uhhm Apr 01 19:13:54 mickeys issue is surely tracable Apr 01 19:14:04 the reason is simple Apr 01 19:14:10 but the solution... Apr 01 19:14:19 I do not have ANY idea how to solve it ATM Apr 01 19:14:23 ok Apr 01 19:15:39 leviathan, for me chvt 1 works Apr 01 19:15:55 chvt 2 works Apr 01 19:16:02 which release? Apr 01 19:16:09 chvt 3 works Apr 01 19:16:18 26268253c2f3f670cea133668ccfa503ffcc800e Apr 01 19:16:21 seem that any chvt works Apr 01 19:16:25 I'll check Apr 01 19:16:28 its not the issue that they do not work Apr 01 19:16:29 :) Apr 01 19:16:38 its that chvt 2 kills resume Apr 01 19:16:44 26268253c2f3f670cea133668ccfa503ffcc800e Apr 01 19:16:48 yes Apr 01 19:16:52 ok Apr 01 19:16:58 so how do you reproduce it exactly Apr 01 19:17:03 chvt during suspend? Apr 01 19:17:05 or before/ Apr 01 19:17:14 s/\//?/ Apr 01 19:17:19 /etc/init.d/xserver-nodm Apr 01 19:17:21 stop Apr 01 19:17:39 direct on dream: chvt 2 Apr 01 19:17:42 over ssh Apr 01 19:17:45 apm -s Apr 01 19:17:51 /usr/bin/susres Apr 01 19:18:04 http://ftp.o2s.ch/4mickey/susres Apr 01 19:18:04 leviathan, I'm away from a computer now. What should I look at tomorrow? Apr 01 19:18:15 you're in ETH Apr 01 19:18:17 right? Apr 01 19:18:20 works for me Apr 01 19:18:32 I don't use susres tough Apr 01 19:18:42 /etc/init.d/xserver-nodm stop Apr 01 19:18:47 chvt 2 Apr 01 19:18:52 echo mem > /sys/power/state Apr 01 19:18:56 echo on > /sys/power/state Apr 01 19:18:58 spaetz: if you can login, you can watch into nariwork-03.ee.ethz.ch Apr 01 19:19:03 with oe Apr 01 19:19:06 's defconfig Apr 01 19:19:24 and mine too Apr 01 19:19:27 spaetz: I've set up a build suite there, you can use it, if you like Apr 01 19:19:40 ahh, nice :-) Apr 01 19:19:48 (didn't try the command I just typed with mine but only console and x suspend-resume) Apr 01 19:19:53 will check it out. Apr 01 19:20:07 spaetz: 16 cores... and because of hollydays: ALL mine ^_^ *wuahahah* Apr 01 19:20:27 hehe, I bet it is fast :-) Apr 01 19:20:33 ohh yeah Apr 01 19:20:36 4 threads Apr 01 19:20:40 make -j 10 Apr 01 19:20:45 2h Apr 01 19:20:57 for something my workstation takes a day ^_^ Apr 01 19:21:11 2h make -j 10 wow Apr 01 19:21:19 16 cores Apr 01 19:21:23 24Gig ram Apr 01 19:21:25 ^^ Apr 01 19:21:33 It took a week for building an image from scratch for bug device Apr 01 19:21:44 between failures and compilations Apr 01 19:22:38 leviathan, can't reproduce the bug Apr 01 19:22:48 grml Apr 01 19:22:49 ok Apr 01 19:22:57 maybe I should try apm? Apr 01 19:22:58 I'm building a most recent image atm Apr 01 19:23:00 yes Apr 01 19:23:16 apm works too Apr 01 19:23:25 altough it's fso-apm I think Apr 01 19:23:39 apm [OPTION...] - FSO APM Compatibility Utility V2 Apr 01 19:24:32 leviathan, fatal: failed to unpack tree object 48afbaf19428893b7aec0a385970f300c38300c3 Apr 01 19:24:45 I'll rm the dir in sources Apr 01 19:29:13 ok Apr 01 19:31:19 you need to set the VT into graphics mode to see the problem Apr 01 19:42:59 Hi, everybody Apr 01 19:43:10 I have trouble building eglibc in SHR-U Apr 01 19:43:17 It fails during do_install Apr 01 19:44:05 The errords in the log start with "In function `dest':memusage.c:(.text+0x60): undefined reference to `write" Apr 01 19:44:30 then a whole lot of other undefined references Apr 01 19:45:17 any idea what could cause this? Apr 01 19:52:55 blindcoder: thanks Apr 01 19:54:27 doon, no clue. but I think eglibc was bumped today.... Apr 01 19:55:57 hmm, ok Apr 01 20:00:20 spaetz: yes I could build it before, the failure only occured after I did an update Apr 01 20:01:02 with zhone2 using fb i can provoke the problem very easily Apr 01 20:01:11 mickeyl, ok do you have an image? Apr 01 20:02:12 let me check for uncommited changes... Apr 01 20:02:18 ok Apr 01 20:03:29 bitbake fso2-demo-image Apr 01 20:03:42 then ssh into it Apr 01 20:03:44 run zhone2 Apr 01 20:03:51 suspend Apr 01 20:03:52 resume Apr 01 20:03:55 no display Apr 01 20:04:02 okee Apr 01 20:04:05 strange Apr 01 20:05:11 if we could have FBIO_UNBLANK we could issue it after resume Apr 01 20:05:51 mickeyl, ok but I'm still stuck at fatal: failed to unpack tree object 48afbaf19428893b7aec0a385970f300c38300c3 whatever I do Apr 01 20:06:07 where/when? Apr 01 20:06:27 mickeyl, ah no sorry Apr 01 20:06:43 /home/embedded/oe/org.openembedded.dev/recipes/linux/linux-leviathan_git.bb so not the same than before Apr 01 20:06:56 gitorious tree is strange Apr 01 20:07:00 i always have to manually check that out Apr 01 20:07:05 go into the downloads dir Apr 01 20:07:05 ok Apr 01 20:07:08 I'll do that Apr 01 20:07:09 and clone it with your host git Apr 01 20:07:11 ok Apr 01 20:07:16 then relaunch the build and OE should pick it up there Apr 01 20:07:33 mickeyl, couldn't I do something like asume provided git? Apr 01 20:07:40 so it wouldn't break all the time Apr 01 20:08:01 i think it has to do with the environment and/or the git fetcher Apr 01 20:08:08 removing oe-built git doesn't even fix it Apr 01 20:08:17 ok Apr 01 20:08:22 i had no time to look into it and debug Apr 01 20:08:25 ok Apr 01 20:08:32 wanted to spend time rather with other things ;) Apr 01 20:08:32 I understand Apr 01 20:08:43 but at some point we have to fix it Apr 01 20:10:37 btw. Apr 01 20:10:48 looking at the code path in msmfb, i'm not surprised the panel does not wakeup Apr 01 20:10:54 ok Apr 01 20:10:54 or only if the console layer does something Apr 01 20:11:03 tatic void msmfb_resume(struct early_suspend *h) Apr 01 20:11:04 calls Apr 01 20:11:20 err, no Apr 01 20:11:24 msmfb_suspend calls panel->blank Apr 01 20:11:31 but msmfb_resume does _not_ call panel->unblank Apr 01 20:11:38 lol Apr 01 20:11:43 WILLFIX Apr 01 20:12:00 (the contrary of WONTFIX in the bugtrackers) Apr 01 20:12:50 also I don't understood why leviathan did what he did at the place where there are ioctls callbacks Apr 01 20:13:11 for fixing screen when resuming Apr 01 20:13:35 i think i understood at least that part Apr 01 20:13:40 ok Apr 01 20:13:54 android is not using any consoles Apr 01 20:13:55 nor VT Apr 01 20:14:02 and no refresh thread Apr 01 20:14:12 so when we build our kernel with consolefb Apr 01 20:14:17 and with the refresh thread Apr 01 20:14:29 then suddenly the kernel needs to stop drawing Apr 01 20:14:32 as opposed to userland (android) Apr 01 20:14:36 before you can suspend Apr 01 20:14:47 so we need to shut down the refresh thread, signalizing that we don't draw Apr 01 20:14:54 and then suspend can suceed Apr 01 20:15:00 (at least that's my interpretation of it ;) Apr 01 20:15:00 yes I understood that,but we don't call MSMFB_RESUME_SW_REFRESHER don't we Apr 01 20:15:02 ? Apr 01 20:15:14 no, he just relaunches a new kernel thread Apr 01 20:15:21 which should have the same effect Apr 01 20:15:21 yes but that's under MSMFB_RESUME_SW_REFRESHER Apr 01 20:15:26 ah, let me check Apr 01 20:15:56 git show 48afbaf19428893b7aec0a385970f300c38300c3 Apr 01 20:16:25 and strangely it works with xf86-video-fbdev Apr 01 20:16:50 i think someone else calls this for us Apr 01 20:16:56 uhm Apr 01 20:16:56 who? Apr 01 20:16:56 and we just have to kill/relaunch the thread Apr 01 20:17:12 should I do BUG() at theses ioctls? Apr 01 20:17:16 :) Apr 01 20:17:26 these ioctl are default one Apr 01 20:17:29 and specified Apr 01 20:17:51 in the struct Apr 01 20:17:54 of this board Apr 01 20:18:09 yes but they are MSM_ ioctls,userpace don't except MSM_ ioctls Apr 01 20:18:18 yes Apr 01 20:18:26 userspace doesnt expect THEM Apr 01 20:18:37 but it expects them in the struct Apr 01 20:18:45 as declarated actions Apr 01 20:18:54 for suspend Apr 01 20:18:56 and resume Apr 01 20:18:57 :) Apr 01 20:18:59 ah ok Apr 01 20:19:06 so the struct get called somehoe Apr 01 20:19:12 yes Apr 01 20:19:15 *struct functions Apr 01 20:19:22 ok Apr 01 20:19:29 board.suspend Apr 01 20:19:34 in pseudo code ;-) Apr 01 20:20:10 ok Apr 01 20:20:50 can we get the panel structure from within static int msmfb_ioctl(struct fb_info *p, unsigned int cmd, unsigned long arg) ? Apr 01 20:21:09 i see how they get it in early_suspend, but that needs their early_suspend structure which we don't have in msmfb_ioctl Apr 01 20:21:27 if we can get it from there, we can implement the blanking ioctl Apr 01 20:21:51 or we implement blanking and add Apr 01 20:22:03 .blank = MSM_BLANKING Apr 01 20:22:09 into the struct :) Apr 01 20:22:16 we have pannel access in msmfb.c Apr 01 20:22:21 btw what's that struct? Apr 01 20:22:26 uhmm Apr 01 20:22:30 panel I guess Apr 01 20:22:31 mom Apr 01 20:22:38 no Apr 01 20:22:43 I meant the one for suspend Apr 01 20:23:00 that's the damn android earlysuspend invention Apr 01 20:23:14 ok Apr 01 20:23:38 they use Apr 01 20:23:40 static void msmfb_earlier_suspend(struct early_suspend *h) Apr 01 20:23:40 { Apr 01 20:23:40 struct msmfb_info *msmfb = container_of(h, struct msmfb_info, Apr 01 20:23:40 earlier_suspend); Apr 01 20:23:40 struct msm_panel_data *panel = msmfb->panel; Apr 01 20:23:50 and then call things on the panel like panel->blank() Apr 01 20:24:16 no idea whether we can do the same without the earlier_suspend struct Apr 01 20:24:29 i can hardly follow taht code Apr 01 20:24:34 which is spread out in three and more files Apr 01 20:24:37 *sigh* Apr 01 20:25:08 leviathan@lilith:~/leviathan-incoming/drivers/video/msm$ grep BLANK . -r Apr 01 20:25:12 :) Apr 01 20:25:47 mickeyl: yes Apr 01 20:25:50 you also need the bourd-trout-panel Apr 01 20:25:59 thats the reason why I hate android Apr 01 20:26:11 the android main project is much worser Apr 01 20:26:13 >_< Apr 01 20:26:19 yes Apr 01 20:26:20 i know i suck at reading kernel code, but this i can't understand at all :/ Apr 01 20:27:27 JaMa|Off: fyi: .32.9 r4 wsod, .32.10 unable to mount (even with rootdelay and qi debug...) Apr 01 20:28:30 vanous123, i case .32.10 fubars your uSD ping me Apr 01 20:28:36 http://lxr.free-electrons.com/source/tools/perf/util/include/linux/kernel.h#L26 Apr 01 20:29:57 JesusMcCloud: went back to .29 Apr 01 20:30:07 ok Apr 01 20:31:13 cast a member of a structure out to the containing structure Apr 01 20:31:34 Hi everyone. I don't understand. I thoug that "opkg upgrade" upgraded the kernel too. but I did ity tree times today on shr-u and i am still with kernel 2.6.29 Apr 01 20:31:46 following thing Apr 01 20:31:50 I'm totaly tired Apr 01 20:32:00 me too lol Apr 01 20:32:00 felt shortly asleep right now Apr 01 20:32:09 moskitoes are comming back here Apr 01 20:32:14 fuck Apr 01 20:32:17 indeed Apr 01 20:32:23 I had one the other day Apr 01 20:32:27 ok, don't worry, i can work some days on palm pre in the meantime Apr 01 20:32:40 but I hope it's just a small ioctl that brings me back :) Apr 01 20:32:42 I had invented a good solution for removing them Apr 01 20:32:54 burner + axe-spray Apr 01 20:32:56 yD Apr 01 20:32:58 xD Apr 01 20:33:05 burner? Apr 01 20:33:12 fire lighter Apr 01 20:33:14 ok Apr 01 20:33:17 undersood Apr 01 20:33:23 :) Apr 01 20:33:23 very dangerous Apr 01 20:33:27 but worked Apr 01 20:33:30 lol Apr 01 20:33:38 they didnt fly around afterwards xD Apr 01 20:33:46 but the whole flat is in fire after that Apr 01 20:33:58 because what if they are hidding in curtains or near bed Apr 01 20:34:04 hmm Apr 01 20:34:24 another possibility is sanity smoke Apr 01 20:34:28 or how its called Apr 01 20:34:31 mickeyl, if it's a mather of doing that we can do it at once but I'll have no way to test: Apr 01 20:34:48 but msmfb_resume does _not_ call panel->unblank Apr 01 20:34:56 I'm at the beginning of the image Apr 01 20:35:07 ahh: olibanum was the name Apr 01 20:35:15 ok, you can do it w/ python Apr 01 20:35:15 makes them die'n Apr 01 20:35:17 python Apr 01 20:35:22 import os, fnctl Apr 01 20:35:33 hmm Apr 01 20:35:35 fd = os.open("/dev/tty2", 0 ) Apr 01 20:36:14 ok Apr 01 20:36:16 fnctl.ioctl( fd, 0x4B3A, 0x1 ) Apr 01 20:36:17 I've python Apr 01 20:36:25 this will set vt2 into graphics mode Apr 01 20:36:27 then Apr 01 20:36:29 chvt2 Apr 01 20:36:31 suspend Apr 01 20:36:32 resume Apr 01 20:36:33 I'm building ATM the most actual image Apr 01 20:36:33 no display Apr 01 20:36:34 boom Apr 01 20:36:40 now it should build trouth Apr 01 20:36:43 until then Apr 01 20:36:43 ok thanks a lot Apr 01 20:36:46 I'll sleep Apr 01 20:36:52 needs some time Apr 01 20:36:59 g'night lev Apr 01 20:37:08 takes some time to generate the eglibc-binary-localedata Apr 01 20:37:10 mickeyl: thx Apr 01 20:37:20 gn8 GNUtoo, mickeyl Apr 01 20:37:31 leviathan, good night Apr 01 20:37:34 thx Apr 01 20:37:37 uhm Apr 01 20:37:40 btw Apr 01 20:37:49 look onto Apr 01 20:38:02 http://people.ee.ethz.ch/~davidlan/downloads/ Apr 01 20:38:05 tomorrow Apr 01 20:38:12 then there will be the builds Apr 01 20:38:17 Could smbd give me a link to his almost-working 2.6.32 config? Apr 01 20:38:36 branches with .32 haven't def_config Apr 01 20:38:42 max_posedon: http://ftp.o2s.ch/4mickey/defconfig Apr 01 20:38:44 :) Apr 01 20:38:48 max_posedon: the one in oe doesn't work? Apr 01 20:38:56 so Apr 01 20:38:57 gn8 Apr 01 20:39:00 n8 Apr 01 20:39:02 oe have one?), I didn't know Apr 01 20:39:47 shr guys are building .32 out of oe Apr 01 20:39:54 and they have a defconfig in there Apr 01 20:40:16 opkg list | grep fnctl => nothing Apr 01 20:40:47 that should be in python-lang Apr 01 20:41:11 no Apr 01 20:41:17 python-fcntl Apr 01 20:41:19 ok Apr 01 20:41:24 lev did not bitbake it Apr 01 20:41:25 my typo Apr 01 20:41:31 ah ok Apr 01 20:44:33 mickeyl, I know why, instead of suspend-resume try that Apr 01 20:44:38 chvt 1 Apr 01 20:45:33 CONFIG_ARCH_MSM=y , leviathan thanks, but I asked for config for other device) Apr 01 20:45:40 will take from OE I think Apr 01 20:45:57 GNUtoo: yes, see my notes from yesterday night: Apr 01 20:45:59 yep. for some reason the framebuffer panel gets not initialized on resume Apr 01 20:45:59 it's almost as it's thinking it's still in blank mode Apr 01 20:45:59 changing to vt1 and back to vt2 forces it to reinit Apr 01 20:46:00 but that goes with an awful flicker that i don't want Apr 01 20:46:17 we do want to reinit the panel without relying on the fbcon and vt layer Apr 01 20:46:24 since we may chose not to compile it in Apr 01 20:46:33 and also we don't want the flicker and delay that goes with changing VTs Apr 01 20:46:37 ok Apr 01 20:47:13 mickeyl, you didn't try so I'll explain...chvt 1 doesn't work,it takes ethernally to do it Apr 01 20:47:25 ooh, that's bad Apr 01 20:47:27 then it's stuck Apr 01 20:47:32 that sometimes happens with the VT layer Apr 01 20:47:36 full of race conditions Apr 01 20:47:38 ok Apr 01 20:47:46 that's why kms is better I bet Apr 01 20:47:50 yep Apr 01 20:48:01 I had tons of issues with x700 lots of time ago with that Apr 01 20:48:13 ya, i can imagine Apr 01 20:48:19 the vga cards were sick Apr 01 20:48:24 flickr flickr flickr Apr 01 20:48:24 x<->vt could freeze the computer Apr 01 20:48:28 yep Apr 01 20:48:39 3d too Apr 01 20:49:01 what should we do then? Apr 01 20:49:17 please get me that ioctl and we are happy :) Apr 01 20:49:29 this allows us more control over the panel Apr 01 20:49:33 the one which blocks? Apr 01 20:49:33 ok Apr 01 20:49:37 we can chose to shut down Apr 01 20:49:42 without suspending Apr 01 20:49:59 I'll scp strace-static Apr 01 20:50:01 and we can chose when to reinit Apr 01 20:50:27 ah no the one you asked for before Apr 01 20:50:28 ok Apr 01 20:50:45 you want FBIO_BLANK and FBIOUNBLANK? Apr 01 20:50:48 yes Apr 01 20:50:52 ok Apr 01 20:50:59 have a look at drivers/video/* Apr 01 20:51:02 I'll go to sleep soon tough Apr 01 20:51:03 ok Apr 01 20:51:05 almost all fb drivers are implementing that Apr 01 20:51:09 ok, sleep well! Apr 01 20:51:11 thanks for listening :) Apr 01 21:14:05 bed time. see you Apr 01 21:14:33 night :) Apr 01 21:27:39 time for me too now... gnight all Apr 01 23:02:14 TAsn: rebuilding.. sync included :) Apr 01 23:22:32 JaMa|Off, cool :P **** ENDING LOGGING AT Fri Apr 02 02:59:56 2010