**** BEGIN LOGGING AT Sat Feb 18 10:59:56 2006 Feb 18 11:29:47 good morning Feb 18 11:35:03 03koen 07org.oe.oz354fam083 * r93cc6636... 10/conf/distro/familiar.conf: familiar.conf: set MAINTAINER ?= familiar-dev@handhelds.org Feb 18 12:06:17 moin Feb 18 12:10:59 moin zecke Feb 18 12:13:00 zecke: are you aware of a change that breaks some dependency thing? I keep getting this http://pastebin.ca/42043 regardless of which package I start to build. Feb 18 12:13:50 somehow the quilt dependency is not recognized. Feb 18 12:15:45 the log says: run.do_patchcleancmd.25780: line 241: quilt: command not found Feb 18 12:36:45 hi Feb 18 12:38:32 chris144: I'm not Feb 18 12:38:41 morning folks Feb 18 12:39:24 zecke: it seems that everything depends ultimately on gnu-config-native Feb 18 12:40:01 however gnu-config-native eeds quilt, which ist not being built yet. Feb 18 12:40:24 something IS wrong here. ok, I keep digging... Feb 18 12:40:44 chris144: well did someone touch the quilt files recently? Feb 18 12:41:01 chris144: try INHERIT += "patcher" as a workaround Feb 18 12:41:36 zecke|away: ok the patcher hint might be wortwhile. I'll give it a try.. Feb 18 13:09:08 anyone ever get psyco to work in ppc? Feb 18 13:11:31 zecke|away: the patcher hint doesn't really work, after -b'ing quilt it fails on autoreconf Feb 18 13:11:59 koen: why is it building quilt? Feb 18 13:13:54 zecke|away: I removed patcher afterwards as I like quilt better Feb 18 13:14:39 I should have said: gnu-config fails on autoreconf Feb 18 13:14:49 ehm Feb 18 13:14:55 libtool-native that is Feb 18 13:14:59 * koen gets some more coffee Feb 18 13:16:07 good night Feb 18 13:24:49 koen: it needs to depend on autoconf-native Feb 18 13:27:52 thought so Feb 18 13:30:06 RP: fix pushed Feb 18 13:30:30 koen: its should have that as a depend automatically so I'm not sure this is the right thing to do :-/ Feb 18 13:30:44 ok, 'fix' pushed Feb 18 13:31:02 it also isn't building quilt, so something is wrong in bitbake Feb 18 13:31:29 my 'fix' was to apt-get install quilt and/or -b quilt-native Feb 18 13:31:49 03koen 07org.oe.dev * r73ffcb18... 10/packages/libtool/libtool-native_1.5.10.bb: libtool-native: DEPEND on autoconf-native as well Feb 18 13:33:53 hi Feb 18 13:33:59 hey law_ Feb 18 13:34:00 any ideas? http://pastebin.com/561006 Feb 18 13:34:07 hey koen Feb 18 13:34:38 RP: it seems versioned RDEPENDS don't work anymore: http://pastebin.com/561006 Feb 18 13:35:56 folks, here are tow pastbins: the 1st with bb 1.3.2.1 which builds gnu-config-native cleanly: http://pastebin.ca/42044 .The 2nd is bb 1.3.3.1 which somehow screws up on the dependancies. http://pastebin.ca/42045 Feb 18 13:37:19 The 2nd should start building quilt-native, as soon as it finds that quilt-native has no more dependencies. But it keeps looking at dependency paths for some reason. Feb 18 13:42:34 chris144: agreed. There is something very wrong with bitbake. I can't debug it this moment in time but will look at it later on Feb 18 13:43:58 RP: cool, I have some time now, I will start looking ito it. We have to see what the outcome will be. I am quite firm with .bb files but not yet well worked into the bitbake source.... we'll see.. Feb 18 13:48:29 btw: hi koen Feb 18 13:48:34 Howdy Feb 18 13:48:57 heh Feb 18 13:49:04 'where to find oe cross toolchain' as a mail subject Feb 18 14:01:10 morning Feb 18 14:01:11 koen: ping Feb 18 14:01:19 pong Feb 18 14:01:40 koen: I would love to try setting up a tinderbox3 on ewi :} Feb 18 14:01:51 koen: do you have some time to spare? Feb 18 14:02:06 zecke|away: yes Feb 18 14:07:09 * koen just finished doing his taxes Feb 18 14:10:47 koen: what if statement is missing in my display switcher script ? Feb 18 14:11:45 law_: do you have a bug number? Feb 18 14:12:03 http://bugs.treke.net/show_bug.cgi?id=699 Feb 18 14:12:52 http://bugs.treke.net/attachment.cgi?id=535 below '#display landscape' Feb 18 14:13:14 there are 4 'fi's and 3 'if's Feb 18 14:13:49 I suspect it should be 'if [ $1 = "1" ]; then' Feb 18 14:14:24 koen: uups ill clean up this Feb 18 14:14:40 no statement 1 isn?t available Feb 18 14:14:55 it was just a guess Feb 18 14:22:46 RP: I just sent a mail to oe# Feb 18 14:22:56 on how to work around these problems Feb 18 14:23:26 koen: fixed thanks Feb 18 14:25:03 koen: what do you prefer? pg or mysql? Feb 18 14:25:25 pg Feb 18 14:26:06 koen: could you create me a db and give me a username+password? Feb 18 14:27:40 hmm eds-dbus doesnt build. NOTE: Applying patch 'disable_orbit.patch' fails Feb 18 14:29:27 koen: could you remove /data/tinderbox2 and clean /data/tinderbox1? Feb 18 14:36:24 * koen rm /data/tinderbox2 -rf Feb 18 14:39:43 zecke: postgres had been installed Feb 18 14:42:14 zecke: but I'm a bit lost on how to create a db Feb 18 14:42:57 hehe Feb 18 14:43:10 http://www.johnkeiser.com/mozilla/tbox3.html Feb 18 14:43:56 koen: wait, I will try it locally first Feb 18 14:45:03 postgres@bitbake:~$ createuser zecke Feb 18 14:45:03 Shall the new user be allowed to create databases? (y/n) y Feb 18 14:45:03 Shall the new user be allowed to create more new users? (y/n) y Feb 18 14:45:03 CREATE USER Feb 18 14:45:08 go ahead :) Feb 18 14:45:28 oh Feb 18 14:48:10 woglinde: hey Feb 18 14:49:17 hey woglinde Feb 18 14:49:32 zecke: the bugzilla integration looks nice Feb 18 14:49:52 zecke: and with monotone.cvssync we could use teh cvs features Feb 18 14:51:05 zecke: http://lists.gnu.org/archive/html/monotone-devel/2006-02/msg00161.html Feb 18 14:51:14 hi koen Feb 18 14:51:22 if that get implemented initial pull will be even faster Feb 18 14:53:31 koen: for the 'cvs' integration, one could use a fake bonsaio server Feb 18 14:54:25 zecke: or a real bonsai server with a fake (tailor+hg,git/monotone) cvs tree Feb 18 14:55:52 koen: wow Dr. Hip looks at monotone Feb 18 14:58:58 zecke: I don't know dr hip Feb 18 14:59:28 koen: author of sqlite (maybe i skipped some 'p''s..) Feb 18 14:59:33 ah, ok Feb 18 15:01:10 zecke: did the db creation work? Feb 18 15:02:10 koen: I have read into the tinderbox code... Feb 18 15:02:19 koen: and now I will to setup a box locally... Feb 18 15:49:54 koen: Where you will be using postgreSQL? Feb 18 15:53:56 lamikr: on ewi546 for the tinderbox Feb 18 16:03:16 Good evening Feb 18 16:03:59 mtd-utils-native does not build. patch "add-exclusion-to-mkfs-jffs2.patch" fails. Feb 18 16:10:25 Ok, postgresql homepage has very good docs how to configure and handle it. For example backup/restore in chapter http://www.postgresql.org/docs/8.1/interactive/backup.html Feb 18 16:11:09 koen: And user management, roles, access rights, etc.. in http://www.postgresql.org/docs/8.1/interactive/user-manag.html Feb 18 16:12:04 and fast installer guide, db creation and start guide: http://www.postgresql.org/docs/8.1/interactive/installation.html Feb 18 16:17:33 hello oll Feb 18 16:21:03 hey hrw Feb 18 16:21:25 uv1: can you post the log file to pastebin? Feb 18 16:21:26 koen: did you saw thread which I gave you at night? Feb 18 16:21:31 hrw: yes Feb 18 16:21:39 koen: funny people Feb 18 16:21:41 hrw: especially funny about the softfloat remark Feb 18 16:21:51 afaik softvfp isn't faster as softfpa Feb 18 16:24:16 I wonder how many Z distros will update their kernel after 3.5.4 to get WPA working Feb 18 16:25:50 http://wickedweasel.com/ Feb 18 16:40:42 chris144: Where is the log file gone ? Up to today I was only working with the prebreakage stuff. Feb 18 16:42:52 uv1: prebreakage? I think I missed something. For me the log file ist still in tmp/work//temp Feb 18 16:43:11 uv1: can u tell me what prebreakage is? Feb 18 16:43:41 chris144: forgive me, found it "http://pastebin.com/561252". prebreakage is the state frombefore reworking bitbake (jan2006) Feb 18 16:45:32 uv1: ok, your log file shows that you are not suffering from the same problem than me. Feb 18 16:46:09 chris144: Thought you were a crack and fix the problem, heh ;-) Feb 18 16:46:32 chris144: Whats your problem then ? Feb 18 16:46:48 uv1: :-) trying to be a crack, but not there yet... Feb 18 16:49:15 uv1: http://pastebin.ca/42043 . Something is wrong with the dependencies calculation in the latest bb Feb 18 16:50:06 uv1: bitbake leads to this problem. Feb 18 16:50:58 uv1: I am about to work through the bitbake code, what I undertand right now, there was a change recently which either breaks something or needs package dependencies updated in the metadata Feb 18 16:51:31 uv1: bitbake (from scratch) leads to this problem Feb 18 16:52:55 chris144: Will care for mtd-utils first and possibly struggle on the same proble, then you. Will come back tomorrow ... Feb 18 16:52:59 Tschau ... Feb 18 16:53:11 uv1: ok see ya Feb 18 16:56:34 hi all! Feb 18 16:58:35 anyone knows about intel flash support? I need to talk to p30 chips, but mtd creates kernel-dumps Feb 18 16:58:48 chris144, out of curiosity when was the last time you updated bitbake? Feb 18 16:59:11 emte: 2/16 Feb 18 16:59:38 emte: Feb 1 still worked Feb 18 16:59:50 yeah Feb 18 17:00:41 emte: I am already going through the bitbake code, and i have the point where it fails. But i do not understand enough I think (yet) Feb 18 17:01:36 in bitbake.py: if fn in self.building_list: return self.addRunDeps(fn, virtual , buildAllDeps) Feb 18 17:02:19 this finds quilt-native already in another build chain and returns without building it. Feb 18 17:02:28 the only that recently got changed was stripping i think Feb 18 17:02:34 thing* Feb 18 17:02:46 emte: stripping is part of OE, not bitbake Feb 18 17:03:19 true Feb 18 17:04:06 is anyone else getting chris144 's errors? Feb 18 17:04:21 evening Feb 18 17:04:58 emte: you need to do a bitbake bootstrap-image (or something else) from scratch to see it. Feb 18 17:05:13 i only get my normaly sysvipc problem (which i need to submit the fix for) Feb 18 17:05:34 yeah i build images regularly Feb 18 17:06:01 emte: could you try with bb ad oe from yesterday, clean tmp dir? from very scratch? Feb 18 17:07:15 sure, gonna take ~ 10 min to partch fakeroot fisrt tho Feb 18 17:07:20 patch* Feb 18 17:07:41 emte: no prob, I keep going through bb in the meantime... Feb 18 17:09:43 and 64 revs Feb 18 17:09:59 i'll brb need to drop my wife off at work Feb 18 17:11:20 * chris144 starts a build of bootstrap-image from scratch with oe and bb from 20060217 Feb 18 17:11:54 chris144: that's likely to fail. read the mailing list. Feb 18 17:12:05 reenoo: ok Feb 18 17:12:34 reenoo: still I can see if it produces my dep prob or something else Feb 18 17:14:16 chris144: I covered up the bug a bit: http://handhelds.org/hypermail/oe/65/6559.html Feb 18 17:14:51 * chris144 looks at koens link Feb 18 17:16:14 Hi, I am also seeing build error if I build dev branch. It will fail in the first package. (gnu-config). I am using new bitbake and oe.db, both checked out today and started building from scratch. fam083 builds fine from scratch. Feb 18 17:16:35 koen: yes, that might help. let me try it... Feb 18 17:18:25 lamikr: see the link I just posted :) Feb 18 17:19:38 cu Feb 18 17:20:43 lamikr: yes, that is what I am seeing too. I know the line in bitbake where it fails but i do not have enough bb knowledge to fix it quickly. I think we have to wait or use koens hint. Feb 18 17:21:33 chris144: where in bitbake would that be? Feb 18 17:22:37 reenoo: look for # See if this is a runtime dependency we've already built Feb 18 17:22:45 reenoo: # Or a build dependency being handled in a different build chain Feb 18 17:22:46 koen: Yeh, just opened after pressing enter :-) Feb 18 17:23:23 reenoo: the if after that finds quilt native already in another build chain and returns without building it. Feb 18 17:23:54 reenoo: this leads to gnu-config trying to do its patches without having quilt built. Feb 18 17:24:26 reenoo: that is as far as i got. I am not sure whether I am looking at a side efect hre or at the root cause. Feb 18 17:24:39 chris144: ok, thanks for the info. Feb 18 17:24:51 reenoo: ok Feb 18 18:01:49 is someone actually taking care of the domain names? this is getting silly. Multiple people have offered to pay for it now. Can one of them just send a check to kergoth? Feb 18 18:04:25 22:22 < kergoth> i really need to get those domains transferred after i pay them. easydns is charging $35 a year, a piece. craziness. want to move them onto godaddy Feb 18 18:04:28 22:22 < pb_> yeah, good plan Feb 18 18:04:31 22:23 < pb_> did you get your paycheck yet? Feb 18 18:04:33 22:23 < pb_> heh Feb 18 18:04:36 so it looks like money isn't the problem anymore Feb 18 18:04:38 22:23 < kergoth> deposited it yesterday Feb 18 18:05:07 taking care of it as we speak. been busy at work Feb 18 18:05:19 hey kergoth Feb 18 18:05:30 dns is fluff anyway Feb 18 18:05:48 and it is slow because it doesn't access the framebuffer directly Feb 18 18:06:24 hehe Feb 18 18:08:56 DNS is overrated Feb 18 18:13:19 hrw|gone: hh.org bugzilla was cleaned up a bit Feb 18 18:21:07 koen|food: heh. a bit much it seems Feb 18 18:21:07 today is oe hackery day for me. Feb 18 18:21:38 who has stuff they want looked at? my focus is going to be gcc4 compilation fixes, i can look into other bugs if they are in bugzilla, also Feb 18 18:31:00 hmm Feb 18 18:31:07 up to 3074 packages Feb 18 18:32:13 * emte is wondering if its time to split packages into branches? Feb 18 18:32:26 just to increase parsing time Feb 18 18:33:57 core, wm specific, and other seems like good choices Feb 18 18:34:38 then again i suppose you really only parse once at tehbegining of a new build Feb 18 18:34:41 no, that's going to end up in chaos. use BBMASK and BBFILES if you want to reduce parsing time Feb 18 18:35:04 i am using BBMASK and BBFILES Feb 18 18:35:11 still takes 5min to parse Feb 18 18:35:46 and thats on a p4 3.2GHz E with 2gb ram Feb 18 18:35:55 emte: each time or just first one? Feb 18 18:35:59 hi Feb 18 18:36:02 just the first time Feb 18 18:36:10 after that its fine Feb 18 18:36:16 emte: it has to build cache ;( Feb 18 18:36:26 this is more me complaining while i notice it so you can ignore me :) Feb 18 18:36:35 emte: in time before cache we had it each time... Feb 18 18:36:42 and it eat 600M Feb 18 18:36:44 yeah i know Feb 18 18:37:10 its just that i never build opie, qte, etc Feb 18 18:37:20 only bootstrap and gpe Feb 18 18:37:29 and e17 once in a while Feb 18 18:37:57 the point being? Feb 18 18:38:19 with BBFILES set to what you want things will get as fast they can Feb 18 18:38:21 not much, as i said you can ignore me for the moment while i complain :) Feb 18 18:38:50 http://pastebin.com/561429 Feb 18 18:39:00 gives me something to do besides cures at dbus Feb 18 18:39:03 and i tried do build opie image :-( Feb 18 18:39:04 curse* Feb 18 18:39:14 failed really early :-) Feb 18 18:39:22 which reminds me Feb 18 18:39:46 apperently we are not using dbus to read hw like the battery and cpu ? Feb 18 18:39:57 law_: lack of quilt? Feb 18 18:40:10 ~lart sd cards Feb 18 18:40:15 root@c7x0:~# while true;do md5sum /dev/mmcblk0;done Feb 18 18:40:15 3d537381ae18772794eb234f4efe61b8 /dev/mmcblk0 Feb 18 18:40:15 f63ab67190a69fb6f8e967f85b2cc0db /dev/mmcblk0 Feb 18 18:40:25 not quite nice... Feb 18 18:41:11 hrw: i?ll do bitbake opie-image again mom Feb 18 18:41:55 test5 opie works fine on my collie, but qpegps doesnt work, so i?ll try to build latest cvs version Feb 18 18:42:42 hrw: if i start bitbake opie-image gnu-config-native is the first package Feb 18 18:42:47 law_: qpegps does not like newer gpsd Feb 18 18:42:58 law_: yep - build quilt-native first by hand Feb 18 18:42:59 i found the most useful gps mapping program Feb 18 18:43:07 http://roadnav.sourceforge.net Feb 18 18:43:12 hrw: thanks Feb 18 18:43:14 it actually works, and looks halfway decent Feb 18 18:43:22 shadows, as opposed to navsys? Feb 18 18:43:39 hrw: latest cvs version works with newer gpsd i think Feb 18 18:43:41 emte, would have to look into what navsys is i don't know Feb 18 18:43:45 http://www.navsys.org/ Feb 18 18:44:19 shadows: is there an roadnav .bb ? Feb 18 18:44:26 law_: don't know Feb 18 18:44:27 law_: great - when you get it working then please add it into OE and give sign - I will add it into .oz354fam083 Feb 18 18:44:36 i've been wondering tho if anyone has looked at Network Manager Feb 18 18:44:58 its goals seem well suited for a pda Feb 18 18:45:00 shadows: does it work with gpe? Feb 18 18:45:10 http://www.gnome.org/projects/NetworkManager/ Feb 18 18:45:15 law_: i think it's Gtk based Feb 18 18:45:53 law_: i've used roadnav on debian desktop boxes, and it fetches tiger data here in the USA very easily, so you get maps and useful navigation software without much effort Feb 18 18:46:01 i'm not sure how it would work for non-US Feb 18 18:46:08 hmm :-( Feb 18 18:46:37 * emte curses at fakeroot Feb 18 18:46:51 i was not able to make "roadster" program do anything useful in my experiments, though the screenshots looked nice Feb 18 18:47:58 i need to figure out how to make OE recognise my flags Feb 18 18:48:04 what flags? Feb 18 18:48:23 ${NOSYSVIPC} Feb 18 18:48:44 its not quite reading it correctly Feb 18 18:49:15 hm Feb 18 18:49:23 seems i am one of the few who doesnt have SYSV support in thier systems and it makes problems for fakeroot Feb 18 18:49:34 oh Feb 18 18:49:39 what system is that Feb 18 18:49:41 so i am trying to test for a flag and alter OECONF Feb 18 18:49:42 slackware? Feb 18 18:49:46 debian Feb 18 18:49:50 uhh Feb 18 18:49:54 you should have SYSV support Feb 18 18:50:00 i'm on a debian box Feb 18 18:50:03 not if you dont install it Feb 18 18:50:16 so install it, what's the big deal about that Feb 18 18:50:26 stable/release images have SYSV support for compatability Feb 18 18:50:43 sd organization should die! Feb 18 18:51:23 shadows, for me it makes more sense to set a flag to change it to TCPipc support instead Feb 18 18:51:37 hrw: wait until Blu-Ray organizations acquire the right to fly in a jet plane and bomb non-licensed abusers' homes Feb 18 18:51:45 hrw quilt-native 0.39 or 0.42 ? Feb 18 18:51:52 law_: any Feb 18 18:51:57 ok Feb 18 18:52:16 i am sure more systems will pop up that do not support SYSV and this is a way to handle those cases Feb 18 18:52:23 shadows: I hope that EU will not get as much fscked Feb 18 18:52:45 hi Greentux_nokia Feb 18 18:52:51 emte: oh. feel free to submit patches, that's cool Feb 18 18:52:59 yeah thats waht i am doing :) Feb 18 18:53:05 it's just counter to what 99% of people are developing on Feb 18 18:53:09 helping increase fault tolorance Feb 18 18:53:13 Hi hrw Feb 18 18:53:20 shadows, not really Feb 18 18:53:30 only RH and BSD are natice SYSV Feb 18 18:53:34 native* Feb 18 18:53:41 uh huh Feb 18 18:53:49 what's wrong with SYSV? Feb 18 18:54:06 why use it if you do not need it in your system? Feb 18 18:54:18 apparently you do need it Feb 18 18:54:20 ;) Feb 18 18:54:40 no, oe doesnt requier it aside from it defaulting to enabled in fakeroot Feb 18 18:54:45 require* Feb 18 18:55:31 as far as i've discovered anyway Feb 18 18:59:52 you really haven't stated what is wrong with SYSV that it should not be used Feb 18 19:03:40 i never said taht anything is wrong with it, i said that on systems without the support a simple flag change is preferable to require the user to install unneeded libraries Feb 18 19:03:56 requiring* Feb 18 19:05:35 hrw: next problem http://pastebin.com/561468 :-) Feb 18 19:06:06 law_: solution: read OE ml Feb 18 19:06:13 law_: build autoconf-native Feb 18 19:10:48 looks like I found better way to capture kernel log without serial cable Feb 18 19:11:05 ssh device -t cat /proc/kmsg|tee logfile Feb 18 19:13:48 talking about ssh, does anyone have any objections of making dropbear use update-alternatives so one can install openssh into the the install w/o both beeing started? Feb 18 19:14:29 CoreDump|home: good idea for me. remember that it means ssh-daemon, ssh, scp Feb 18 19:14:53 right Feb 18 19:16:52 CoreDump|home: you probably want to remove the symlinks in /etc/rc?.d Feb 18 19:17:12 CoreDump|home: instead of abusing update-alternatives even more Feb 18 19:17:34 that's not an abuse, it's the type of thing it was _designed_ for Feb 18 19:17:43 RP: well, the kernel module compiles and can be loaded and shows some output in dmesg....it's just "HP Status: 0" or 1, depending on whether I hold a key or not. Feb 18 19:17:53 it wasn't designed to handle init scripts Feb 18 19:18:15 and removing dropbears init.d links with a postinst off openssh server would be an ugly hack Feb 18 19:18:29 JustinP: That HP Status message is from the sound code Feb 18 19:18:31 it was designed to handle all kind of files Feb 18 19:18:39 RP: yeah, that's what I thought Feb 18 19:18:52 JustinP: I suspect you need to disable that so you can register the interrupt for the rc driver Feb 18 19:19:26 but FWIW i was going to use the same rc script _name_ for both anyways Feb 18 19:19:36 CoreDump|home: you asked for objections. I do object messing with core packages. Feb 18 19:19:38 RP: I should probably also remove KERN_DEBUG from the printk, right? Feb 18 19:20:31 JustinP: KERN_ERR will make sure you see it. KERN_DEBUG will be visible in dmesg Feb 18 19:21:00 reenoo: objection noted, but beside it beeing a core package, you didn't tell me a reason why that change would be bad and / or wrong in the first place Feb 18 19:21:27 CoreDump|home: make them both conflict/replace Feb 18 19:21:49 hrw: now that's a productive idea Feb 18 19:22:19 RP: so KERN_DEBUG should be shown in dmesg? well, I'm not seeing anything...suppose I should disable that sound interrupt then... Feb 18 19:22:27 I'll check wheter replacement works when dropbear is a DEPENDS of task-bootstrap Feb 18 19:22:39 CoreDump|home: a) dropbear and openssh should conflict each other in the first place b) the less hacks involved before dropbear starts the better Feb 18 19:22:56 on that we can agree Feb 18 19:25:32 JustinP: yes, it does appear there Feb 18 19:26:07 hmm.. fail card, fail! Feb 18 19:26:27 * koen wonders where zecke is Feb 18 19:26:39 RP: should I just disable sound altogether in the defconfig? Feb 18 19:26:48 RP: good news: in monday/tuesday my serialcable will be created so week later I will have it to test Feb 18 19:27:18 RP: or unload the module...? Feb 18 19:29:12 JustinP: As you have corgi_rc compiled in and sound is modular, just don't install the module Feb 18 19:29:31 JustinP: As its compiled in, your driver should win the race of course :-/ Feb 18 19:29:48 JustinP: Can you pastebin the current code you're using - I'll give it a glance over Feb 18 19:30:20 JustinP: trying to get remote control working? Feb 18 19:31:05 re Feb 18 19:31:12 18 20:28 * koen wonders where zecke is Feb 18 19:31:14 ;) Feb 18 19:31:20 koen: Our metadata is faulted - quilt-native RDEPENDS on patch, diffstat etc Feb 18 19:31:27 koen: This is the real problem... Feb 18 19:32:00 hrw: yes Feb 18 19:32:07 hey zecke Feb 18 19:32:09 The question is whether REPENDS get resolved before or after a package gets built Feb 18 19:32:15 RP: actually I compiled and installed as a module Feb 18 19:32:38 RP: I can install my kernel where it's compiled in, though Feb 18 19:32:56 * JustinP wonder is he should pastbin this....it's not very....good looking Feb 18 19:33:05 koen: I was in a Gym :) Feb 18 19:33:05 I just did enough to make it compile basically.. Feb 18 19:33:06 RP: it should rdepend on patch-native, diffstat-native, etc Feb 18 19:33:47 bitbake-native, python-native, sh-native, cvs-native, wget-native Feb 18 19:34:25 koen: agreed Feb 18 19:34:44 JustinP: A module is probably best Feb 18 19:35:00 JustinP: Just disable the sound modules (rm/rename them) Feb 18 19:36:01 RP: I'll try that Feb 18 19:36:03 RP: http://oe.pastebin.com/561507 Feb 18 19:36:14 I'll add a fix for the quilt-native RDEPENDS and add patch-native, diffstat-native etc to our default ASSUME_PROVIDED Feb 18 19:36:35 Of course that won't work as they're RDEPENDS, not DEPENDS and we have nothing to resolve the RDEPENDS into DEPENDS :-/ Feb 18 19:36:38 RP: we never had diffstat in needed software iirc Feb 18 19:36:51 hrw: Its an RDEPENDS of quilt Feb 18 19:37:07 o.. never noticed Feb 18 19:37:22 RP: moving the sound directory (out of lib) enough? Feb 18 19:37:34 hrw: You're right, we haven't though. We don't really need it Feb 18 19:37:44 JustinP: Yes, it should be Feb 18 19:38:00 * JustinP reboots Feb 18 19:38:12 (hiz Z) Feb 18 19:38:23 1.2M mmc.log already... when Z will finally get faults.... Feb 18 19:39:07 zecke: Any ideas what to do about ASSUME_RPROVIDED? Add some dummy bitbake files? (or real ones?) Feb 18 19:40:03 that issue is weird Feb 18 19:40:29 zecke: Its not - quilt-native RDEPENDS on patch, diffstat and bzip2 Feb 18 19:40:37 oh hehe Feb 18 19:41:01 RP: fwiw, -native doesn't produce any packages, right? Feb 18 19:41:06 so how can it have rdepends? Feb 18 19:41:07 =-( Feb 18 19:41:19 zecke: The quesiton is how to ASSUME_RPROVIDE patch-native, diffstat-native, bzip2-native? Feb 18 19:41:43 RP: I find koen's comment right. No package is emitted for them Feb 18 19:41:54 RP: do we want to have RDEPENDS in native packages? Feb 18 19:41:59 It still has a runtime depenedency on those other tools being present Feb 18 19:42:43 RP: to get more independant from the host system we might add real -native versions for diff,patch,bzip2 Feb 18 19:43:03 zecke: That's one good approach Feb 18 19:43:17 RP: nope, nothing...:-( no output at all Feb 18 19:43:21 (in dmesg) Feb 18 19:43:27 RP: and then people get put ASSUME_PROVIDED in auto.conf or site.conf :) Feb 18 19:43:35 RP: and gzip, tar, zip, cvs, subversion, git? Feb 18 19:43:40 zecke: or teach OE to use multimachine stuff for -native packages Feb 18 19:43:52 hrw: yes, we already set some of them Feb 18 19:43:52 and gnu-tools Feb 18 19:44:29 what about small code which check does tools are available and then set A_P? Feb 18 19:44:36 it boils down to the question if we want to assume fundamental tools being present or not Feb 18 19:44:42 JustinP: You need to add something to arch/arm/mach-pxa/spitz.c Feb 18 19:44:50 RP: oh... Feb 18 19:45:15 zecke: yep. or we could end in A_P with gcc, libc-dev etc.. Feb 18 19:45:18 JustinP: add a "spitz-remote-control" platform device definition which the driver can match Feb 18 19:45:35 * JustinP looks Feb 18 19:45:36 See that file and you'll get the idea - copy the corgikbd one Feb 18 19:45:47 hrw: well gcc is the only thing we need to assume Feb 18 19:46:06 JustinP: Also, call it sharpsl-remote-control so we can share it with other devices like corgi Feb 18 19:46:17 RP: makes sense Feb 18 19:46:31 okay I would be happy with ASSUME_PROVIDED or real -native bbfiles Feb 18 19:46:51 33 minutes and card does not yet failed.. mmc_debug is big change.. Feb 18 19:46:51 and if we feel lucky we could use some code from the nslu2 people and automatically generate auto.conf Feb 18 19:47:13 zecke: We can't just set ASSUME_PROVIDED as there is nothing to map the RDEPENDS = patch-native to the DEPENDS patch-native Feb 18 19:47:58 RP: -native packages don't have RDEPENDS. it's all DEPENDS from bitbake's POV Feb 18 19:48:11 * france is away: Away Feb 18 19:48:12 RP: you need them at build time Feb 18 19:48:48 reenoo: The concept of RDEPENDS is perfectly valid for a -native package Feb 18 19:49:07 reenoo: native.bbclass should set BUILD_ALL_DEPS of course (it might already) Feb 18 19:49:28 I'll give this more thought - I have to go Feb 18 19:49:30 ~lart flash animations with sound Feb 18 19:50:33 so if dbus doesnt read the battery and cpu, does anyone know what service is? Feb 18 19:51:19 it looks like the only thing that dbus currently reads are terminal logins and hotplug events Feb 18 19:51:35 RP: later Feb 18 19:51:37 RP: thanks Feb 18 19:54:21 maybe i jsut need to make apci and dbus talk ... Feb 18 19:54:31 unless we are still using apm ... Feb 18 19:55:02 AFAIK we're using apm....but I'm not the oracle of OE... Feb 18 19:55:03 emte: dbus doesn't read anything, it's just a bus Feb 18 19:55:28 yeah, just attach an apm/acpi messager....or something... Feb 18 19:55:36 emte: you'd probably want the kernel-event-layer, which can generate events which can be put on the bus Feb 18 19:55:44 aha Feb 18 19:55:49 thanks koen Feb 18 19:55:55 k-e-l is 2.6, but bedboi should have a backport for 2.4 Feb 18 19:56:01 * emte goes to find the kernel-event-layer docs Feb 18 19:57:00 hmm what do you know ... Feb 18 19:57:06 http://www.linuxjournal.com/node/7744/print Feb 18 19:57:10 thats kinda handy :) Feb 18 19:58:10 looks liek its already been done somewhere Feb 18 19:58:47 * emte goes to find sysfs docs now Feb 18 20:02:41 hmm Feb 18 20:03:04 koen, do you have a link for bedboi ? Feb 18 20:11:44 koen, do you have a link for bedboi ? Feb 18 20:15:17 #gpe i suppose Feb 18 20:15:28 ~seen bedboi Feb 18 20:15:41 hmm Feb 18 20:15:45 ibot is still AWOL Feb 18 20:17:46 koen: I am testing the build of gomunicator in fam083 (I copied gomunicator package from dev to 083) branch and have found two problems. Feb 18 20:18:25 !!seen bedboi Feb 18 20:18:26 CoreDump|home: Don't know, don't care Feb 18 20:18:31 heh Feb 18 20:19:51 good night folks Feb 18 20:20:04 cu chris144|gone Feb 18 20:20:05 n8 chris144|gone Feb 18 20:20:05 'night chris144|gone Feb 18 20:20:18 cu Feb 18 20:22:36 hmm Feb 18 20:22:46 * emte hasnt ran into chris's error yet ... Feb 18 20:23:40 koen: I know how to fix those problems, but as you are the gomunicator maintainer it might be better if you make commits. Feb 18 20:24:02 lamikr: do you have a patch? Feb 18 20:25:11 koen: Not yet. But at least the first one is easy. Feb 18 20:26:51 First fix is that gomunicator uses alsa, and there needs to be added dependency to alsa in gomunicator_cvs.bb. (I added dependency to alsa-utils, but propably the dependency to alsa-libs would be enought) Feb 18 20:28:37 * emte goes and washes the car while bootsrap-image builds Feb 18 20:28:51 koen: From the second problem I am not 100% sure. I copied the gomunicator package from dev to fam083 and at least in there I needed to use "export SRCDATE=20060219" to get it building. My dev branch is still building so I have not yet tested in there. Feb 18 20:29:17 change SRCDATE to CVSDATE Feb 18 20:29:55 we are allowing floating cvsdates in the oz/fam stream? Feb 18 20:30:26 oz maybe Feb 18 20:30:34 familiar has it tied to a global CVSDATE Feb 18 20:30:48 YES! finally error ;) Feb 18 20:30:56 1h17m Feb 18 20:31:01 btw isnt 0219 tomorrow? Feb 18 20:31:32 emte: it is Feb 18 20:32:02 thats good, thought my watch mighta been a day off Feb 18 20:33:00 hi does anybody know build problems for fontconfig-native Feb 18 20:33:45 build bootstrap-image works on h2200 and hp ipaq 2100 Feb 18 20:33:46 ~kill sf.net cvs Feb 18 20:34:06 but both fail for gpe-image Feb 18 20:38:07 koen: So dev uses SRCDATE variable that floats, and fam083 uses fixed CVSDATE? "IF" gomunicator could be moved to 083, would that mean that there should be made a gomunicator snapshot tar.bz2 that would be always used. (no retrieves from cvs head after that anymore) Feb 18 20:38:39 lamikr: CVSDATE got renamed to SRCDATE after we branched Feb 18 20:39:32 ok, well that's was wise as there are now svn, git, etc... Feb 18 20:40:20 koen: But is fam083 branch still getting some packages from the cvs head or using always tagged versions? Feb 18 20:41:16 lamikr: for familiar it will use the global cvsdate (20051212) Feb 18 20:41:33 you can specify a different CVSDATE per package Feb 18 20:43:08 koen: Ok, so changing CVSDATE for example in the gomunicator_cvs.bb will just make a local change, (not global) and will not affect to other packages? Feb 18 20:43:21 I think so Feb 18 20:43:45 I'd just put a CVSDATE_gominucator = "20060218" in local.conf or something like that Feb 18 20:44:44 oops should have mentioned that it is the org.openembedded.oz354fam083 branch Feb 18 20:45:27 any ideas? Feb 18 20:45:56 koen: Ok, this _gomunicator was magic that I did not knew. Thanks. I will test tomorrow how the gomunicator works in fam083, and if things are ok, I would like to see it added to fam083. Feb 18 20:48:00 RP: still not seeing anything...I suspect it may not be attaching to the platform device correctly? Feb 18 20:48:03 RP: http://oe.reversefold.com/sharpsl-remote-control.patch Feb 18 20:48:39 if there are any other kernel hackers around who could give me pointers.... Feb 18 20:53:37 lamikr: that is the override of bitbake Feb 18 20:58:22 zecke: So when CVS(SRC)DATE is specified, does bitbake try to get labeled version from repository where label match with CVSDATE or does this CVSDATE be used from getting already bzipped source packages from handhelds ftp-site? Feb 18 20:58:49 lamikr: no I meant the VARIABLE_foo = "thing" Feb 18 20:59:21 lamikr: these are two features. If SRCDATE is specified it will try to fetch this date from svn,svk,cvs... Feb 18 20:59:42 lamikr: and then we have the TARBALL STASH which looks on a server for a prepackaged tarball :) Feb 18 21:00:06 zecke: Ok, so that will work with almost every variable... Feb 18 21:03:05 lamikr: we often have SRC_URI_arm = "this.patch;patch=1" Feb 18 21:03:27 lamikr: then this patch gets applied if you build for TARGET_ARCH arm Feb 18 21:06:51 zecke: Thanks for explaining this. In addition it seems that bitbake is bzipping the sources retrieved from cvs automatically to sources dir. (At least I seems to now have sources/gomunicator_cvs.sourceforge.net__20060218.tar.gz) in addition of having sources in sources/cvs/gomunicator Feb 18 21:08:03 hi zecke Feb 18 21:09:17 lamikr: http://familiar.handhelds.org/releases/v0.8.3/feed/x11/gomunicator_0.1.3+cvs20060218-r0_arm.ipk ;) Feb 18 21:10:08 zecke: any luck with tb3? Feb 18 21:10:11 lamikr: right. It zips it, (task fetch) and then you have a task unpack Feb 18 21:10:27 koen: I'm setting up a postgres db here and documenting it step by step Feb 18 21:10:35 zecke: cool Feb 18 21:12:16 koen: postgres is done (if the script matches the tables used...) now onto configuring apache Feb 18 21:12:51 hmmm Feb 18 21:12:55 postgres isn't in OE Feb 18 21:35:15 koen: Do you are there any plans with oe people who are coming to fosdem from the evening happenings or hotels/hostels where to live? I have now arranged my flights but have not reserved hotel room. Feb 18 21:35:45 afaik we go out for drinks on friday evening Feb 18 21:36:06 and a lot of us are staying at the astrid hotel on rue de samedi Feb 18 21:36:36 ouch the default f.d.o tbox3 is a security nightmare Feb 18 21:37:06 tbox3 uses bugzilla to log in, but f.d.o's tbox does a return 1 Feb 18 21:37:08 ok, I will also be there on friday. Is it ok, if I attend there on friday? Feb 18 21:37:43 nice quote: 'She sells C shells...' Feb 18 21:38:10 lamikr: sure Feb 18 21:38:15 lamikr: the more the merrier :) Feb 18 21:38:24 :-) Feb 18 21:38:45 * koen wants to learn some more finnish Feb 18 21:39:33 koen: I heard that it is easy language Feb 18 21:39:43 like hungarian on japan Feb 18 21:39:52 it's related to hungarian Feb 18 21:40:10 and I do speak some hungarian :) Feb 18 21:40:13 * lamikr is always ready to teach within beer Feb 18 21:40:44 I know - at 2000/2001 party one friend gave long talk about finish/hungarian (worked as german<>polish translator of StarOffice) Feb 18 21:41:00 It's belgium, of course there will be beer :) Feb 18 21:41:21 hmm... think i have been await for to long .... how did you disable ccache for a specific pkg now again :P Feb 18 21:41:30 ldc: CCACHE="" Feb 18 21:41:40 ah, thanks! Feb 18 21:41:52 Yes, some people in Finland keep very close contacts with Hungarian. My fiwes father for example offered me very nice hungarian self made spirit that he had got in last visit from hungarian... Feb 18 21:42:03 lamikr: palinka! Feb 18 21:43:01 good bimber is good Feb 18 21:43:06 Yes, something home made, really strong one but one could easily smell that it was made from the garden fruits, grapes, etc.. Feb 18 21:43:55 hi CoreDump|home Feb 18 21:44:31 hi hrw|tv Feb 18 21:46:40 koen: Is this astrid hotel in good location? (I have newer been in Bryssel earlier). I was first thinking this chap which had a link from the fosdem site. http://www.chab.be/ Feb 18 21:47:11 lamikr: it is Feb 18 21:47:29 lamikr: some OE people wll be there iirc Feb 18 21:47:50 koen: Ok, I will start googling to it. I think I could also try to book then from there. Feb 18 21:48:07 http://astridhotel.be/ Feb 18 21:50:13 Yes, looks good :-) I will try to get a room from there tomorrow. Feb 18 21:51:01 any perl guru around? Feb 18 21:51:12 how can I print a map? (the content) Feb 18 21:52:17 sorry, no help from me. I have always tried to run away when seeing perl. Feb 18 22:03:44 *perl god* Feb 18 22:04:17 yuck perl Feb 18 22:05:06 perl.. executable linenoise Feb 18 22:05:07 my $sane_login =~ s/\@/%40/; Feb 18 22:05:31 NA|Zzz: heh good one Feb 18 22:05:37 "How would you describe your perl knowledge?" "I know it inside out" Feb 18 22:06:52 Perl - The only language that looks the same before and after RSA encryption. Feb 18 22:07:03 HEHE Feb 18 22:07:44 lol Feb 18 22:08:18 I clean my taglines and some friends share stuff Feb 18 22:12:47 If GOD had intended computers to have 1,024 bytes per kilobyte, Feb 18 22:12:47 HE/SHE would have given us SIXTEEN FINGERS!! [Dan Barrett] Feb 18 22:14:01 hrw|tv: you mean you don't have 16 fingers! Feb 18 22:14:52 does anyone have the CVS sources for minimo-0.0+cvs20060218 ? Feb 18 22:16:49 I guess they are in mozilla cvs Feb 18 22:17:05 which apears to be b0rked Feb 18 22:30:55 03nail 07org.oe.dev * rfdd12ff2... 10/ (2 files in 2 dirs): slugos-packages: Add netcat to feeds Feb 18 22:30:59 03nail 07org.oe.dev * rde843404... 10/packages/linux/ (4 files in 2 dirs): ixp4xx-kernel: Add coda module to feed Feb 18 22:31:36 netcat.. that reminds me that I need udp netcat Feb 18 22:36:00 hmm Feb 18 22:36:21 someone broke h3600_backpack .. Feb 18 22:36:55 * emte needs to buy an orbital ... Feb 18 22:37:08 i forgot how long it takes to wax a car by hand ... Feb 18 22:37:18 especially if you do it twice ... Feb 18 22:37:20 koen you can't leave now :} Feb 18 22:38:38 KOED: check SETUP in my home dir of ewi Feb 18 22:42:31 ok. time for Robocop Feb 18 22:42:46 have fun :) Feb 18 22:43:40 * reenoo read that as RoboCup Feb 18 22:44:00 * reenoo looks at his todo list for that and sighs Feb 18 22:47:40 hmm Feb 18 22:47:55 where do you think sysfs would live in if it exists in the 2.4 kernel? Feb 18 22:48:00 JustinP: Your device name needs to match in the driver - .name = "spitz-remote-control", .name= "sharpsl-remote-control", don't matcg Feb 18 22:50:37 emte: sysfs was introduced in 2.5 Feb 18 22:52:14 emte: as for h3600_backpack, I haven't upgraded the sa kernel for ages... Feb 18 22:53:25 its a timer error it says, duplicate definition Feb 18 22:53:56 and yeah i know it was introed in 2.5, tryiong to find a 2.4 port of it Feb 18 22:54:43 otherwise i am gonna have to come up with another way ... Feb 18 22:58:37 mm Feb 18 22:58:45 time to finish waxing the car ... Feb 18 23:05:55 koen|sleep: hey Feb 18 23:06:04 KOED: check SETUP in my home dir of ewi Feb 18 23:09:00 koen|sleep: now I only need to update the tinderclient bbclass Feb 18 23:13:00 yeah linux boots on the intel imac Feb 18 23:16:57 Hi, I am trying to build OE for Nokia770 and getting the following error: | checking checking for d-bus... configure: error: Can not locate the D-BUS, please use the switch "--with-dbus-root-dir" to specify its location. | FATAL: oe_runconf failed Feb 18 23:17:38 Any one know were do I need to add the --with-dbus-root-dir=PATH switch? Feb 18 23:18:44 ghellman: which 'package' was build? Feb 18 23:19:10 libosso-0.8.9-r2 Feb 18 23:19:41 there should be something like a libosso*.bb Feb 18 23:19:58 and for some reason it does not specifiy this option Feb 18 23:22:32 Where can I find the libosso*.bb file? Feb 18 23:23:33 If I look at ~/org.openembedded.dev/packages I can' find any subdirectory called libosso* at all Feb 18 23:23:54 This is the usual place were the *.bb files for the packages are located in Feb 18 23:24:04 ghellman: man find Feb 18 23:24:08 ghellman: find ~/org.openembedded.dev/packages -name 'lib*osso*.bb' Feb 18 23:25:52 bingo it is under ~/org.openembedded.dev/packages/maemo/ :-) Feb 18 23:26:35 there could be something like a EXTRA_OECONF in this file... Feb 18 23:28:01 So what is the dest way to add this switch? Feb 18 23:28:18 I can't see EXTRA_OECONF in the file Feb 18 23:29:04 ghellman: understand the issue first :) Feb 18 23:36:56 ok who forgot to pay the openzaurus.org bill? Feb 18 23:44:56 why is this important? Feb 18 23:45:08 who forgot to give the guy the money :) Feb 19 00:00:14 RP: oops.... Feb 19 00:07:48 hehe, only important because its slightly more annoying digging through google's cache for doc's Feb 19 00:11:01 Is there any chess boards on OE? I have seen one engine, but no boards. Feb 19 00:14:24 poli: knights for opie Feb 19 00:14:46 hrw|tv: nothing for X, I suppose? Feb 19 00:15:04 poli: I do not know that part Feb 19 00:17:10 REdOG: oe.handhelds.org? Feb 19 00:22:24 ooh yummy Feb 19 00:25:16 cu Feb 19 00:35:56 good nite Feb 19 00:42:18 RP: ok, the device is registered and all but I don't think it's actually doing anything....no output from the driver that I can see Feb 19 00:42:51 RP: I did tail -f on /dev/input/event2 (which should be the remote) but nothing happened....not sure if that's the right way to check... Feb 19 00:44:30 hi Feb 19 00:45:51 has anybody tried building gpe-image Feb 19 00:46:29 oh many many times Feb 19 00:47:39 do you get a fail on fontconfig Feb 19 00:48:37 check the mailing list or backlog in this channel Feb 19 00:53:27 cheers I'll check the backlog but for now need to sleep (in uk) Feb 19 00:54:04 * Harvy is back Feb 19 00:54:47 * Harvy is back Feb 19 01:17:46 ttfn Feb 19 01:36:08 how do I get into the diag menu on a sl-6000? Feb 19 01:42:07 nm got it Feb 19 03:21:48 well building oe/oz/fam seems to be working much better in 32bit mode than 64 Feb 19 03:26:08 * emte forgets where i saved my sa-1110 datasheet Feb 19 03:26:53 heh Feb 19 03:37:38 simcop2387: please work with me to make that process better o_O Feb 19 03:37:59 simcop2387: as a test, i just built today's checkout using gcc4 on an amd64 pure box, with gpe-image and spitz hardware target Feb 19 03:38:07 build succeeded without any failures Feb 19 03:38:10 :) Feb 19 03:38:14 what should i try next? Feb 19 03:38:21 opie-image? Feb 19 03:38:32 good luck Feb 19 03:38:52 shadows: honestly i'm not sure what was going wrong, lots of linking problems with opie-image/collie, it could have just been something being stale in the build dir, but i decided to try again in 32bit and see how it would go, so far not a single problem Feb 19 03:38:55 i dont know if a non-release opie will build Feb 19 03:39:17 emte: is opie having troubles? Feb 19 03:39:35 i belive so with the recent changes Feb 19 03:39:45 hmm i'll tell you if it builds, just pulled earlier today around 3pm Feb 19 03:39:46 est Feb 19 03:40:05 simcop2387, from oz-fam or unstable? Feb 19 03:41:10 oz354 branch Feb 19 03:41:26 hey why the heck does wpa-supplicant-nossl_0.2.6-r3 build? Feb 19 03:41:33 i would think a newer one would build Feb 19 03:42:10 conf/machine/include/zaurus-clamshell-2.6.conf:5:PREFERRED_VERSION_wpa_supplicant = "0.4.7" Feb 19 03:42:10 shadows: why are you asking why it does build? Feb 19 03:42:14 but that's not building Feb 19 03:42:50 ah Feb 19 03:42:53 simcop2387: wpa-supplicant 0.2.6 is rather useless Feb 19 03:43:08 thats weird why wouldn't it build Feb 19 03:43:14 you need a newer version of wpa-supplicant for it to work with 2.6 kernel Feb 19 03:43:28 i found that wpa-supplicant version 0.4.7 works fine Feb 19 03:43:31 but it is not building Feb 19 03:43:42 if i bitbake it manually, it builds fine Feb 19 03:43:43 can wpa_supplicant run a script when it sees certain APs? Feb 19 03:44:02 i don't know about that, wpa_supplicant provides a socket interface to control it Feb 19 03:44:21 there is a Qt+ widget based gui called wpa_gui to scan and select/configure wpa info Feb 19 03:44:25 it works well on laptops Feb 19 03:44:57 hmm Feb 19 03:45:02 this "nossl" thing is the problem Feb 19 03:45:47 why can we not include openssl in the gpe-image? Feb 19 03:53:28 http://www.jinx.com/scripts/details.asp?affid=-1&productID=122 Feb 19 03:53:48 kergoth: i'm wearing a carpe noctem hoodie Feb 19 03:53:57 hehe Feb 19 03:54:08 quality stuff. i'm upset that the material tears easily in the washer/dryer machines Feb 19 03:54:30 though being a total geek, i know how to sew and mend the holes ;) Feb 19 03:56:36 hmm Feb 19 03:56:58 simcop2387: what do you think, is it so bad to drag openssl into the rootfs image of clamshell openzaurus targets? Feb 19 03:57:14 you could make a BB for Network Manager Feb 19 03:57:29 not sure why its not in OE yet Feb 19 03:57:35 Network Manager last i checked doesn't touch WPA Feb 19 03:57:46 shadows: i've wondered that too, my only guess was that it added too much of a size for collies and such but for clamshells they've got alot more space don't they? Feb 19 03:57:55 yeah Feb 19 03:58:17 screw it, i'm s/nossl/new package/ Feb 19 03:58:17 shadows: maybe its a remnant of a simpler build system? the bitbake system should be able to do it Feb 19 03:58:28 http://www.gnome.org/projects/NetworkManager/users/ Feb 19 03:59:19 emte: that directed towards me? Feb 19 03:59:54 or shadows Feb 19 04:00:15 * emte shrugges Feb 19 04:00:15 ah Feb 19 04:00:29 i'm checking it out, if its simple enough to make a .bb for i'll try it Feb 19 04:01:21 emte: any idea if that'd work in opie? Feb 19 04:01:48 says its independent or wm and works in most Feb 19 04:01:52 of* Feb 19 04:02:31 yea once i'm done building world i'll play around with making a .bb for it (of course i'm going to try again to get POSE running on my collie) Feb 19 04:02:38 first Feb 19 04:04:00 simcop2387: i poofed up a IIIc today Feb 19 04:04:10 poofed up? Feb 19 04:04:11 flashed a different Palm OS on it, now the screen comes up black :( Feb 19 04:04:15 ah Feb 19 04:04:20 poof meaning, bad, fucked it up Feb 19 04:04:28 magic smoke let out = poof Feb 19 04:04:38 though i didn't actually let the magic smoke out physically Feb 19 04:04:46 i've wanted to get POSE running on my collie simply for some old software i like Feb 19 04:04:49 yeah Feb 19 04:04:58 do you know anyone who could rip the OS off a IIIc ? Feb 19 04:05:06 useing osflash or romeo or something Feb 19 04:05:25 it's for my dad's dad, he's in need of a PDA, thought i'd be nice and upgrade the OS but i really didn't know what i was doing Feb 19 04:05:29 shadows: there is a way to completely restore them to factory settings i think, i've got a IIIxe i think it is Feb 19 04:05:38 oh Feb 19 04:05:43 the IIIxe is upgradable Feb 19 04:05:46 i've done it, love it Feb 19 04:05:48 i could be wrong though Feb 19 04:06:44 i think it was like datebook + up when hitting the reset button Feb 19 04:06:53 hmm Feb 19 04:07:07 i guess i need to look at the h36 suspend/sleep code Feb 19 04:07:55 why? Feb 19 04:08:30 http://www.handhelds.org/~bugzilla/show_bug.cgi?id=1551 Feb 19 04:20:53 hmm Feb 19 05:41:42 * shadows tries out an OZ build Feb 19 05:41:45 neat Feb 19 05:41:52 the libsqlite bug is fixed ;) Feb 19 05:43:33 moo Feb 19 05:44:43 bugger i don't know how to get wpa_supplicant 0.4.7 onto the gpe-image Feb 19 05:55:55 shadows, if you figure it out let me know as i'd like to add macchanger to the wireless interface as well Feb 19 07:03:04 opie built fine on first build, oz354 Feb 19 07:03:26 only problem was gcc having an internal error once but wasn't reproducable (heh, that happens alot) Feb 19 08:11:19 shadows, emte: what do you mean you don't know how to get wpa_supplicant? just build it and install it Feb 19 08:32:36 JustinP: no, how to make wpa-supplicant 0.4.7 show up in the rootfs tarball of gpe-image target Feb 19 08:32:56 JustinP: currently with my patches, it is including still the version 0.2.6 of wpa-supplicant-nossl Feb 19 08:33:02 which is non-functional on kernel 2.6 Feb 19 08:34:12 ah, I see Feb 19 08:35:00 if you need to do some conditionals on kernel version I recently did some magic on gcc-cross version you could use as a template Feb 19 08:35:26 wait....those were your bbs that I altered weren't they... Feb 19 08:35:33 * JustinP is always too tired.... Feb 19 08:37:01 I need a simple program to listen to a kernel 2.6 event device and let me know if anything comes through Feb 19 10:24:03 NOTE: package inkscape-0.43-r0: task do_build: completed Feb 19 10:24:07 that was pretty painless Feb 19 10:27:29 03koen 07org.oe.dev * rf101beff... 10/packages/inkscape/ (files/no-boehm-version-check.patch inkscape_0.43.bb): inkscape: add 0.43 Feb 19 10:27:30 koen: nice Feb 19 10:27:42 JustinP: Running tail on the /dev/input/event2 device would show you if anything was happening. It sounds like its time to start adding printk's - see if the headphone interrupt triggers for a start Feb 19 10:28:05 koen: scap ? Feb 19 10:28:27 law_: I don't have scap on this board Feb 19 10:28:54 koen: aha Feb 19 10:31:42 JustinP: http://www.rpsys.net/openzaurus/temp/evtest.c will let you view an input device properly Feb 19 10:51:58 morning **** ENDING LOGGING AT Sun Feb 19 10:59:57 2006