**** BEGIN LOGGING AT Mon Apr 02 02:59:58 2012 Apr 02 11:03:39 SHR: 03GNUtoo 07meta-smartphone * r19d3dc5ed5ea 10/meta-nokia/recipes-kernel/linux/linux-nokia900-meego_git.bb: meta-nokia: bump srcrev to unbreak nokia900 kernel build Apr 02 12:47:20 Where can I find ShortOm or similar? Both opkg.org and projects.openmoko.org, where it's supposed to be hosted, is down. Apr 02 13:00:10 stiell: seems like there might be a copy in this SVN repo: https://code.google.com/p/mbac/source/browse/trunk/mbac/usr/bin/shortom-bin?spec=svn3&r=3 Apr 02 13:09:21 DocScrutinizer: are the old projects.openmoko.org VCS repositories on a server somewhere? stiell is looking for some code Apr 02 13:50:40 pabs3: the svn vserver still down, Roh doesn't feel like brushing out the bugs until whole infra moved to the new box. Scheduled for 'in a short while' Apr 02 14:25:12 DocScrutinizer51: IIRC projects.openmoko.org VCSen were completely separate to git.openmoko.org/svn.openmoko.org Apr 02 14:25:24 thanks for the info re svn.om.o though Apr 02 17:24:27 GNUtoo: dbus not coming up does happen on n900 and crespo only? Apr 02 17:27:24 not on gta04 Apr 02 17:27:27 for me Apr 02 17:27:34 I don't know about other devices Apr 02 17:27:45 crespo, n900 -> no-dbus Apr 02 17:27:50 gta04 -> dbus Apr 02 17:31:38 hmm Apr 02 17:32:01 let me check on gta04 anyway... maybe that's the reason for the missing PIN dialog Apr 02 17:36:26 2000-01-01T00:03:35.763763Z [ERROR] DBusServiceDevice <>: Can't find modem for modem_type option_gtm601; corresponding modem plugin loaded? Apr 02 17:37:59 GNUtoo: heh, I got no module for option_gtm601 ? Apr 02 17:38:24 root@om-gta04:~# ls /usr/lib/cornucopia/modules/fsogsm/ Apr 02 17:38:25 dbus_service.so modem_dummy.so pdp_nokia_isi.so pdp_ppp.so pdp_ppp_mux.so ppp2fsogsmd.so Apr 02 17:38:27 modem_cinterion_mc75.so modem_singleline.so pdp_option_gtm601.so pdp_ppp_internal.so pdp_qmi.so Apr 02 17:41:19 you must opkg install it Apr 02 17:41:24 like it is said in the install Apr 02 17:41:29 *install page Apr 02 17:41:32 mrmoku, ^^^ Apr 02 17:42:12 * mrmoku thought not to be in need of that page ;) Apr 02 17:42:21 why is that? Apr 02 17:42:34 missing dep in fsogsm-modules? Apr 02 17:42:50 fsogsmd-modules that is Apr 02 17:45:56 RDEPENDS_${PN}_gta04 += "fsogsmd-module-modem-option-gtm601" Apr 02 17:46:39 GNUtoo: missing om in there Apr 02 17:46:45 JaMa: ^^ Apr 02 17:49:04 ah right Apr 02 17:49:28 the page was a quick way to make users aware of what they should do Apr 02 17:50:59 ok, preparing a patch Apr 02 17:51:11 that one is soo obvious that I would tend to push without testing ;-) Apr 02 17:52:01 said and done Apr 02 17:53:42 ahh, I was missing the feeling of having fixed something :-) Apr 02 17:53:57 guess I will have to fix something else now Apr 02 17:59:08 hi, little offtopic question. does someone know something about OpenMaxIL and Android? I'm trying to find if there's an implementation of it there (libOmxCore.so looks like so) but I'm not sure if it's there or I need to compile some other implementation of the OpenMaxIL API (such as bellagio one) Apr 02 18:44:33 GNUtoo, JaMa: btw. whoever did it... my desktop icons are stable now... no reload needed anymore after boot :-D Apr 02 18:45:11 ok nice Apr 02 18:45:26 GNUtoo: but I get no softkey Apr 02 18:53:13 that's because of HOME it seems Apr 02 18:53:27 HOME is not set Apr 02 18:53:33 which means that most apps won't work Apr 02 18:53:39 eve won't work Apr 02 18:53:42 intone won't work Apr 02 18:53:44 etc... Apr 02 19:11:04 hmm, ok Apr 02 19:11:29 root@om-gta04:~# echo $HOME Apr 02 19:11:29 /home/root Apr 02 19:11:36 GNUtoo: at least via ssh it is set Apr 02 19:13:07 but indeed... in valaterm it is not Apr 02 19:39:46 GNUtoo: it is *not* missing HOME Apr 02 19:40:13 I changed /bin/sh to /bin/bash in passwd and added a .bashrc setting it Apr 02 19:40:18 it is set in valaterm now Apr 02 19:40:25 still no softkey Apr 02 19:42:20 is shr_elm_softkey.desktop in right .e/e/applications/startup/.order ? Apr 02 19:42:20 e.g. not in /tmp/.e? Apr 02 19:42:41 root@om-gta04:~# cat .e/e/applications/startup/.order Apr 02 19:42:41 shr_elm_softkey.desktop Apr 02 19:43:30 SHR root@gjama ~ $ cp .e/e/applications/startup/.order /tmp/.e/e/applications/startup/ Apr 02 19:43:34 SHR root@gjama ~ $ systemctl stop xserver-nodm.service Apr 02 19:43:36 SHR root@gjama ~ $ systemctl start xserver-nodm.service Apr 02 19:43:40 and shr_elm_softkey seems to be running here Apr 02 19:43:44 hmm Apr 02 19:44:14 * mrmoku will try exactly that when the reboot finished Apr 02 19:45:00 JaMa: the one in /tmp is empty after boot Apr 02 19:45:55 JaMa: yeah, copying it and restarting x 'fixes' it Apr 02 19:46:53 * JaMa testing EnvironmentFile to set HOME Apr 02 19:48:14 mrmoku, ok Apr 02 19:48:32 ok nice Apr 02 19:50:01 1st time wizard.. good sign Apr 02 19:50:33 ahh, yeah. I did not get that one Apr 02 19:50:39 and running Apr 02 19:50:54 should I push it? Apr 02 19:51:06 yeah, push Apr 02 19:51:25 JaMa: btw. I forgot (or better thought it would not be needed) the PR bump for fsogsmd-modules Apr 02 19:51:39 I guess it would have been better to bump Apr 02 19:54:16 hmm looking at it again I guess I've fixed it a while ago with /etc/profile which is sourced in /etc/X11/Xsession Apr 02 19:54:24 so it's weird that this issue returned Apr 02 19:54:47 and EnvironmentFile is work around, should be fixed "again" in /etc/profile Apr 02 19:55:58 was your fix before or after switching to bash? Apr 02 19:56:07 maybe this was only for PATH fix http://git.shr-project.org/git/?p=meta-smartphone.git;a=commit;h=95760613ab0cff3fd754e6276e9e13643867ffd1 Apr 02 19:56:28 IIRC switching to bash was with shr-core switch Apr 02 19:56:38 ok Apr 02 19:57:00 but I somehow remember some PATH issues Apr 02 19:58:25 ah it was about $HOME ;) 'if [ "$HOME" = "/home/root" ]; then' wasn't working for us.. Apr 02 19:58:45 ahh, yeah, that I have heard I think :) Apr 02 20:00:16 so we need to find out why USER is set soon enough, but HOME isn't Apr 02 20:07:27 hmm and in shell login I also have HOME set Apr 02 20:09:26 indeed... logging in via ssh it is set just fine Apr 02 20:10:25 also when logged locally Apr 02 20:10:58 who is responsible for setting it? I guess some pam module now, but without pam, was it shadow itself (as /bin/login provider)? Apr 02 20:14:02 yup login should parse it from /etc/passwd or set '/' if not specified Apr 02 20:15:11 hmm ok Apr 02 20:15:13 hmm and it does Apr 02 20:15:20 SHR root@gjama /etc/pam.d $ unset HOME Apr 02 20:15:20 SHR root@gjama /etc/pam.d $ /bin/login Apr 02 20:15:21 GNUtoo: missing PIN is due to phonefsod not being started Apr 02 20:15:25 and still have HOME set Apr 02 20:20:32 ok Apr 02 20:20:37 sigh n900 sound quality Apr 02 20:20:46 I spent hours on it without a real result Apr 02 20:21:01 buffer underruns Apr 02 20:21:03 etc... Apr 02 20:21:35 ALSA lib pcm.c:7339:(snd_pcm_recover) underrun occurred Apr 02 20:21:58 and it works only on plug:dmix:0 **** BEGIN LOGGING AT Mon Apr 02 20:24:03 2012 Apr 02 20:25:21 mrmoku: what exactly do you want me to PR bump? Apr 02 20:25:36 mrmoku: ah you pushed it to master branch Apr 02 20:25:45 mrmoku: use shr for meta-smartphone please Apr 02 20:27:38 SHR: 03mok 07meta-smartphone * raf989834b7f4 10/meta-fso/recipes-freesmartphone/freesmartphone/fsogsmd-modules_1.0.bb: fsogsmd-modules: fix typo in modem module dependency for om-gta04 Apr 02 20:28:45 JaMa: ohh, did that change or does my memory suck? Apr 02 20:28:50 anyway... sorry Apr 02 20:29:38 not sure when I did that change, but it's a while, and NP :) Apr 02 20:30:09 ok :P Apr 02 20:30:14 2011-12-21 Apr 02 20:30:16 :P Apr 02 20:30:39 ähm... well then my memory sucks... Apr 02 20:31:17 GNUtoo: what was wrong with linux-nokia900-meego build? Apr 02 20:32:17 JaMa: ahh, and we're back to udev now? Apr 02 20:33:02 it didn't built anymore Apr 02 20:33:11 same for crespo Apr 02 20:33:16 GNUtoo: it did for me Apr 02 20:33:20 so I'm curious Apr 02 20:33:22 the error should be in the commit message Apr 02 20:33:35 basically it said that -mno-unaligned-access was not valid Apr 02 20:33:40 and interestingly last 2 commits are Apr 02 20:33:42 Makefile: add -mno-unaligned-access to fix boot on nokia n900 Apr 02 20:33:49 Revert "Makefile: add -mno-unaligned-access to fix boot on crespo" Apr 02 20:34:56 and by "current openembedded toolchain" you mean gcc-4.6 from toolchain-layer, right? Apr 02 20:35:07 I mean the one I've got here Apr 02 20:35:13 I didn't look more Apr 02 20:35:37 I don't know if you have enabled toolchain-layer after gcc-4.6 bbappends were moved from meta-oe Apr 02 20:36:01 I didn't touch anything Apr 02 20:36:18 should I bitbake -e | grep something? Apr 02 20:37:27 tmp-eglibc/sysroots/x86_64-linux/usr/bin/armv7a-vfp-neon-oe-linux-gnueabi/arm-oe-linux-gnueabi-gcc -mno-unaligned-access main.c -o main Apr 02 20:37:30 is legal Apr 02 20:37:38 try ^ Apr 02 20:37:53 if it doesn't work for you then add toolchain-layer to your bblayers.conf Apr 02 20:40:17 ah I don't have toolchain-layer then Apr 02 20:40:27 I've not changed my layers recently Apr 02 20:40:51 "after gcc-4.6 bbappends were moved from meta-oe" Apr 02 20:41:06 ok Apr 02 20:41:13 about 10 days ago Apr 02 20:41:14 http://git.shr-project.org/git/?p=shr-makefile.git;a=commit;h=ae917e3eca82c196df271673eebdc05f7af33a80 Apr 02 20:41:17 yes I know Apr 02 20:41:31 but I tought the toolchain-layer was for experimental stuff Apr 02 20:41:35 can you please revert that revert and SRCREV bump? Apr 02 20:41:38 and that I was not concerned Apr 02 20:41:46 ok Apr 02 20:42:20 it also fixes couple of bugs in ARM code generation Apr 02 20:42:26 such as firefox? Apr 02 20:42:53 it's the same as what was in meta-oe, so no AFAIK Apr 02 20:43:04 ok Apr 02 20:43:07 if firefox was broken with meta-oe gcc-4.6 then it's still broken Apr 02 20:43:12 ok Apr 02 20:43:18 it's broken since a while Apr 02 20:43:23 ld.gold is the issue Apr 02 20:43:40 maybe we should use ld.bfd Apr 02 20:43:52 btw we will have chromium in a meta-browser soon Apr 02 20:44:03 I'm working on it for work Apr 02 20:44:53 nice Apr 02 20:47:10 hmm "meta-oe: recipes-graphics: xserver-nodm-init: fix DPI" Apr 02 20:47:17 is not really fix of DPI Apr 02 20:47:23 it does a lot more than that Apr 02 20:48:29 it's sort of correct as long as we have xserver-common in RDEPENDS but that commit message is bad Apr 02 20:50:05 JaMa, I can't do the revert right now tough Apr 02 20:50:34 ok Apr 02 20:50:51 could you fix the commit message? Apr 02 20:51:27 not now Apr 02 20:51:30 thanks a lot for pointing me the meta-toolchain change Apr 02 20:51:31 ok Apr 02 20:51:49 * JaMa off to bed Apr 02 21:02:34 mrmoku: ping Apr 02 21:16:22 SHR: 03shr-devel 07buildhistory * rcce958b7319f 10/packages/all-oe-linux/xserver-nodm-init/ (8 files in 8 dirs): Build 201204022305 of shr 20120402 for machine om-gta02 on opmbuild Apr 02 21:51:30 SHR: 03shr-devel 07buildhistory * r489bbed88587 10/images/crespo/eglibc/chroot-image/ (4 files): Build 201204022345 of shr 20120402 for machine crespo on opmbuild Apr 03 00:19:12 GNUtoo, http://git.freesmartphone.org/?p=cornucopia.git;a=blob;f=fsoaudiod/src/plugins/gsmvoice_alsa_cmtspeechdata/cmthandler.vala;h=ba2c7f963c554c2a7eaf5e5dd7eb091e1d93b250;hb=3530b692cafbd2bf0d7d400b281ea9b991ceed10 line 339 Apr 03 00:19:35 did you change dmix there? Apr 03 00:22:50 not yet Apr 03 00:23:08 it was put there long time ago Apr 03 00:25:21 ok. **** ENDING LOGGING AT Tue Apr 03 02:59:58 2012