**** BEGIN LOGGING AT Sat Jul 16 23:59:56 2005 Jul 17 02:47:05 mickeyl: did you have a chance to respond to my mail yet? Jul 17 02:49:26 heh Jul 17 02:49:29 nevermind Jul 17 02:50:25 ya i did Jul 17 02:50:27 :) Jul 17 02:50:36 40 minutes ago or so Jul 17 02:50:47 " there is a guy who was working Jul 17 02:50:49 on something like for his university" Jul 17 02:50:52 that would be me Jul 17 02:50:57 heh, no no Jul 17 02:50:59 it wasn't you Jul 17 02:51:07 zecke would now Jul 17 02:51:10 his nick was Chaos or so Jul 17 02:51:14 s/now/know/ Jul 17 02:53:33 mickeyl: non-monolithic means upgrading (and downgrading) between releases should be relatively painless Jul 17 02:53:59 that would also mean the first release is after we have a stable EABI build Jul 17 02:54:11 right Jul 17 02:54:50 * koen|train sees the right trainstation out of the window Jul 17 02:54:56 talk to you later Jul 17 03:56:34 hi ! Jul 17 04:31:12 world champion :) Jul 17 04:31:13 *yipee* Jul 17 04:35:34 zecke: very good Jul 17 04:35:42 world champion in what respect, exactly? Jul 17 04:36:52 pb_: Soccer Jul 17 04:37:07 pb_: Small-Size Robots, and 2nd place in the Mid-Size League Jul 17 04:37:38 I would have never expected to get world champion in football/soccer Jul 17 04:38:07 hehe Jul 17 04:38:19 congratulations Jul 17 04:53:29 Hello, could I please ask a prop. really stupid question. I try to get the OE environment going and wonder if I need to install a seperate tool chain for x-compiling ? or is it already in bitbake and the other files I have donwloaded ? Jul 17 04:55:10 oho: when you run bitbake the first time it will setup a toolchain for you.. Jul 17 04:56:40 ah ok - thanks Jul 17 04:57:43 I followed the manual on the webpage and get the following output - any ideas where the problem lies ?: (510:phm4@dhcp2936:~/progs/oe/build)$ bitbake nano Jul 17 04:57:43 NOTE: Using cache in '/home/phm4/progs/oe/build/tmp/cache' Jul 17 04:57:43 ERROR: no files to build. Jul 17 04:57:43 NOTE: Parsing finished. 0 cached, 0 parsed, 0 skipped, 0 masked. Jul 17 04:57:43 ERROR: Nothing provides nano Jul 17 04:58:42 did you set the BBPATH env variable? Jul 17 04:59:38 yes: export BBPATH=/home/phm4/progs/oe/build:/home/phm4/progs/oe/org.openembedded.dev Jul 17 05:02:29 and I have the latest version of bitbake installed: (544:phm4@dhcp2936:~/progs/oe/build)$ bitbake --version Jul 17 05:02:29 BitBake Build Tool Core version 1.3.0, bitbake version 1.3.2 Jul 17 05:04:43 hm, strange.. Jul 17 05:05:50 do you think I might have sth. wrong in my local.conf ? Jul 17 05:07:44 did you set the packages path in the local.conf? Jul 17 05:09:02 do you mean this one ? : BBFILES := "/home/phm4/progs/oe/openembedded/packages/*/*.bb" Jul 17 05:11:04 I think this might be where the prob lies - just checked the directory and it seems to be screwed up, I will check it out again Jul 17 05:16:27 I don't have := in the BBPATH declaration.. I just have a = there.. although I don't know it that matters.. Jul 17 05:19:28 no that doesn't make a difference, but what is strange is, that my penembedded directory is empty after checking out ? is that normal ? Jul 17 05:19:50 I tried to build mutt directly: bitbake -b org.openembedded.dev/packages/mutt/mutt_1.5.4i.bb Jul 17 05:20:14 and that seems to work, therefore I think I might not be pointing to the sources correctly Jul 17 05:20:47 did you use monotone 0.20 ? Jul 17 05:21:08 no 0.19 Jul 17 05:21:47 on vanille.de it says: "Note that you need monotone 0.20 to connect to this server" Jul 17 05:23:20 ah ok - didn't get any errors, but it might be good to update then Jul 17 05:24:08 the db creation will take a while - thank you for your help Jul 17 05:24:28 you're welcome.. Jul 17 06:01:28 howdy people Jul 17 06:03:53 heya Jul 17 06:03:56 03freyther 07org.oe.dev * r42235576... 10/packages/patcher/ (files/freebsd_gpatch.patch patcher-native_20040913.bb): Jul 17 06:03:56 patcher on FreeBSD: Jul 17 06:03:56 -Patch patcher to use gpatch on FreeBSD Jul 17 06:03:59 03freyther 07org.oe.dev * r842279e2... 10/ (3 files in 3 dirs): Jul 17 06:03:59 conf/bitbake.conf: Jul 17 06:03:59 -Add the BUILD_OS to the overrides. This allows us for example Jul 17 06:03:59 to use gpatch on FreeBSD instead of patch Jul 17 06:04:01 packages/patcher/patcher-native_...bb: Jul 17 06:04:03 -Change the PATCHCMD to use gpatch as well Jul 17 06:04:05 - s/build-fbsd/build-freebsd/ Jul 17 06:08:57 could anyone take a look at this error, please, and tell me what is wrong? http://pastebin.ca/17908 Jul 17 06:09:24 I didn't enable uclibc in the local.conf.. Jul 17 06:18:22 Delta_: monotone pull ; monotone update Jul 17 06:19:13 Delta_: http://ewi546.ewi.utwente.nl/tmp/viewmtn/revision.psp?id=d2ba131f96b5ba9ebe63fc924cc95179c0623844 Jul 17 06:19:56 ah, merci.. Jul 17 06:24:25 03freyther 07org.oe.dev * rc2d29cd1... 10/classes/base.bbclass: Jul 17 06:24:25 base.bbclass: Jul 17 06:24:25 Make it compatible with FreeBSD Jul 17 06:24:25 Replace --no-derefence with -P according to the GNU cp manpage Jul 17 06:24:25 this option is the same as --no-derenfe and FreeBSD knows -P as Jul 17 06:24:26 well Jul 17 06:24:48 koen: how do you connect to the internet from the train? do they have wlan in the trains now? Jul 17 06:27:24 good afternoon all Jul 17 06:29:42 hi Jul 17 06:30:31 shouldn't "squashfs-tools[-native] depend on lzma ? Jul 17 06:31:04 I get compilation errors here Jul 17 06:31:23 if adding lzma to DEPENDS works, file a bug against it Jul 17 06:31:24 or best would be to make this optional Jul 17 06:31:37 ok Jul 17 06:31:40 thanks Jul 17 06:31:43 s/works/fixes it/ Jul 17 06:33:08 pb_: I'm at the same point neil stevens was Jul 17 06:33:21 should I start a flamewar on ipkg-native? Jul 17 06:34:34 * koen gathers some asbestos clothes Jul 17 06:35:09 zecke: while you're at it, you could suggest that maemo should swith to qt/e too Jul 17 06:35:46 zecke: I think standard procedure at this point is to throw a tantrum, post some flames to the list, and then never be heard of again. Jul 17 06:36:33 with a 30 percent chance that gentoo get's mentioned Jul 17 06:37:32 koen: wifi on trains? Jul 17 06:37:46 where/who/what/when? Jul 17 06:37:58 mithro: gprs on trains Jul 17 06:38:28 pb_: do you've an idea how to not include features.h (directly) but have GNU_LIBRARY defined Jul 17 06:38:37 mithro: dopler effect happens in the train ;) Jul 17 06:38:56 zecke: dunno. why do you not want to include ? Jul 17 06:39:08 pb_: libbb.h does ;) Jul 17 06:39:28 pb_: to check for __GNU_LIBRARY__ obviously I do not have a glibc ;) Jul 17 06:39:38 zecke: right, but why can it not include that header? Jul 17 06:39:48 pb_: I do not have it ;) Jul 17 06:40:25 ah, I thought was part of Single Unix. Jul 17 06:40:27 I guess not. Jul 17 06:41:16 * zecke scratches head Jul 17 06:41:30 zecke: I suppose you could write an autoconf test. Jul 17 06:41:40 AC_CHECK_HEADER() or some such Jul 17 06:42:09 * zecke ignores pb_ as he said the evil word Jul 17 06:43:00 koen: you have afforadble GPRS? Jul 17 06:43:03 alternatively, you could check for defined(__linux__), and assume that non-linux machines might not have features.h Jul 17 06:43:49 mithro: one free meg/month, more is 2 euro/mb Jul 17 06:43:52 actually pigi should fix ipkg now, and right now... Jul 17 06:44:10 mithro: about 3 hours of irc + mailchecking Jul 17 06:44:27 zecke: to be honest, I think that whole chunk should just be deleted from libbb.h. Jul 17 06:44:31 nobody cares about libc5 anymore. Jul 17 06:45:06 pb_: I bet Neil Stevens return if you remove libc5 support Jul 17 06:45:16 pb_: and maybe lorn complains we break compability Jul 17 06:45:16 ah Jul 17 06:45:19 in that case, let's not remove it Jul 17 06:45:31 we wouldn't want to do anything that might bring Neil back Jul 17 06:46:58 ~lart patcher Jul 17 06:46:58 * ibot hooks into a hydrant and hoses patcher down Jul 17 06:47:27 it's mildly farsical anyway, since nothing in ipkg or libbb actually uses socklen_t, daemon() or strtok_r. Jul 17 06:47:31 koen: ahh, I wouldn't call that affordable :P Jul 17 06:47:52 i have it a bit problem with building bootstrap-image at final stage when it install packages into fake root. but it doesn't find packages task-bootstrap depends on. Jul 17 06:48:00 zecke: you should get with the programme and use quilt Jul 17 06:48:01 when i run by hand: /stuff/tmp/staging/i686-linux/bin/ipkg-cl -f /stuff/tmp/work/bootstrap-image-1.0-r0/temp/ipkg.conf -o /stuff/tmp/rootfs install task-bootstrap Jul 17 06:48:26 pb_: that is FreeBSD: I made patcher work with replacing patch with gpatch ;) Jul 17 06:48:33 it list all packages which depends Jul 17 06:48:34 pb_: for quilt I would need to mess with getopt Jul 17 06:48:40 ah Jul 17 06:48:48 that does sound unpleasant Jul 17 06:49:04 maybe we should add glibc-native to oe, so that you can build yourself a tolerable development environment Jul 17 06:49:16 i tried also only "base-files", it's the same Jul 17 06:49:16 hehe Jul 17 06:49:39 I wonder if Debian/FreeBSD uses glibc, or the native bsd libc. Jul 17 06:50:03 aaargh Jul 17 06:50:10 pb_: /me too Jul 17 06:50:10 the jlime kernel problem surfaces again Jul 17 06:50:29 pb_: I think it only uses the FreeBSD kernel Jul 17 06:51:02 /stuff/tmp/work/bootstrap-image-1.0-r0/temp/ipkg.conf it point it correct ipk pool, and "Packages" index is also ok Jul 17 06:51:35 zecke: yah, that's what I suspected also Jul 17 06:51:36 ~lart schurig for a broken patcher Jul 17 06:51:36 * ibot wallops schurig with a main rotation server that needs rehubbing. It won't take long for a broken patcher Jul 17 06:51:48 zecke: so, theoretically, you could install debian in a chroot under your freebsd system Jul 17 06:51:50 heh Jul 17 06:52:23 pb_: psst, I love to waste my weekend Jul 17 06:52:56 i tried with debuf output too, but it doesn't tell me anything usefull Jul 17 06:55:42 zecke: heh Jul 17 06:56:41 pb_: I would one want to include Jul 17 06:57:13 i think it miss some criteria by finding packages, but i don't know how to follow that process Jul 17 06:57:18 ich@opiezilla:~$ cat /usr/include/sys/vfs.h | wc -l Jul 17 06:57:18 4 Jul 17 06:57:53 opiezilla Jul 17 06:57:54 heh Jul 17 06:58:44 oh someone hacked me ;) Jul 17 06:59:07 I hate it when that happens Jul 17 06:59:15 pb_: could it be the new bitbake doesn't like PREFERRED_PROVIDER_virtual/kernel_kernel24="handhelds-pxa" ? Jul 17 06:59:15 * pb_ checks that none of his hostnames have gotten set to opiezilla Jul 17 06:59:29 anyone give me some hint ? :) Jul 17 06:59:37 koen: dunno, you'd have to take that up with the Champ. Jul 17 06:59:48 * koen looks at zecke Jul 17 07:00:23 koen: It should like it? Or an update_data() is missing Jul 17 07:00:53 I just svn upped bitbake and it tries to bitbake the jlime kernel for h3900 Jul 17 07:01:38 * koen thinks about installing OE on osx Jul 17 07:02:47 pb_: Lord of autostuff, there is no test for statfs(2) right? Jul 17 07:10:29 when i do "ipkg.cl ... list " i get list of all packages Jul 17 07:11:35 wow now it only has to make linkable Jul 17 07:13:55 zecke: not that I know of, but it should be easy enough to test for it with AC_CHECK_FUNC. Jul 17 07:18:33 I will make it compile and link and then try to make use of autoconf Jul 17 07:19:31 good plan Jul 17 07:20:20 heh, what is it with lzma? that package seems to be perpetually broken. Jul 17 07:20:57 strndup is considered more secure than strdup I guess? Jul 17 07:21:06 I just field a bug report Jul 17 07:21:18 zecke: not really Jul 17 07:21:25 marvin24: yeah, I saw it Jul 17 07:22:06 pb_: who is responsable ? Jul 17 07:22:15 dunno Jul 17 07:22:28 does the bb file have a MAINTAINER? Jul 17 07:22:39 otherwise, check monotone's history to see who added it Jul 17 07:22:43 no Jul 17 07:23:39 lol Jul 17 07:23:44 there is replace/strndup.c Jul 17 07:23:50 and xstrndup in libb Jul 17 07:24:12 it is tested for strndup but ipkg_config ignores HAVE_STRNDUP completely... Jul 17 07:24:16 heh Jul 17 07:24:16 pb_: I'm not very familiar with monotone ... Jul 17 07:25:30 me neither Jul 17 07:25:39 mickeyl and koen are the masters of monotony. Jul 17 07:26:11 monotone annotate Jul 17 07:26:20 that should give some clues Jul 17 07:27:05 heh Jul 17 07:27:12 what a metaphor Jul 17 07:30:44 hey mickeyl, heading to Innsbruck now Jul 17 07:30:52 keon: the file lzma-native belongs to you ;-) Jul 17 07:31:00 then Vienna then back into Germany Jul 17 07:31:04 anyway have to run now Jul 17 07:31:06 see ya! Jul 17 07:31:41 take care Jul 17 07:31:53 zecke: I should be in berlin in about a week Jul 17 07:32:02 will tell you more after I reach Vienna Jul 17 07:32:28 marvin24: see http://oe-devel.bkbits.net:8080/openembedded/src/packages/lzma?nav=index.html|src/.|src/packages Jul 17 07:33:57 mithro: heya! have fun and take care Jul 17 07:33:57 koen: ok, old file, who is br1 ? Jul 17 07:34:31 ~br1 Jul 17 07:34:49 he's the guy working on the Nylon distribution for the meshcube hardware Jul 17 07:35:36 ~seen zerochaos Jul 17 07:35:37 koen: i haven't seen 'zerochaos' Jul 17 07:36:46 oh, br1. Jul 17 07:36:48 that explains it Jul 17 07:37:20 hehe Jul 17 07:37:51 found him and will cc him Jul 17 07:38:56 might as well reassign the bug to him Jul 17 07:39:02 same with that other lzma-native bug that mickey filed Jul 17 07:42:43 pb_: done Jul 17 07:50:07 wow ipkg-native builds and links Jul 17 08:07:37 hmm flex-native fails to build :} Jul 17 08:13:46 zecke: very good! Jul 17 08:13:53 (ipkg, not flex) Jul 17 08:14:13 I revived replace/ Jul 17 08:20:17 03mickeyl 07org.oe.dev * r888665f8... 10/packages/python/ (python-imdbpy_1.9.bb python-imdbpy_2.0.bb): upgrade python-imdbpy to 2.0 Jul 17 08:40:04 afternoon Jul 17 08:44:08 morning kergoth Jul 17 08:44:12 afternoon all Jul 17 08:44:18 hey RP Jul 17 08:44:26 Have I missed much? :) Jul 17 08:52:51 hi RP Jul 17 08:52:56 yeah, quite some stuff Jul 17 08:53:08 to start with the most important one Jul 17 08:53:20 inotify is in 2.6.13-rc Jul 17 08:53:37 and we found the problem of the SD/MMC in 2.6.12 Jul 17 08:53:37 mickeyl: yay :) Jul 17 08:53:56 What was it? Jul 17 08:53:58 2.6.12 removed a sysfs ACK attribute, hence udev needs 20 seconds until the device node appears Jul 17 08:54:19 udev-062 fixes this, however... it doesn't work on the Z. I get lots of alignments issues and it crashes completely Jul 17 08:54:26 hmm :-/ Jul 17 08:54:39 someone would need to look into this Jul 17 08:54:44 ok, that's kind of progress :) Jul 17 08:55:01 * RP is still reading a backlog of thousands of emails... Jul 17 08:55:07 and perhaps also adapt the init script which seems to have progressed quite a lot Jul 17 08:55:21 but that's the only major issue kernel-wise Jul 17 08:55:31 my BT card now works out of the box Jul 17 08:55:42 thanks to some tweaks to module autoloading and hotplug Jul 17 08:55:56 I assume the CF issues are still there? Jul 17 08:55:59 oh, and udev since 058 completely changes the way it works at all Jul 17 08:56:04 yes, CF issues still there. Jul 17 08:56:27 so we have to catch up with all then new developments in hotplug,udev,and kernel Jul 17 08:56:33 s/then/the/ Jul 17 08:56:47 but i'll leave you reading through your mails now :) Jul 17 08:56:48 bbl Jul 17 08:57:26 mickey_away: :) Jul 17 09:21:27 good afternoon Jul 17 09:28:00 hi where i can find dettailed info about zaurus SL-C3100 and e port status on it ? thanks Jul 17 09:29:15 gremlin[it]: www.openzaurus.org maybe? Jul 17 09:29:24 gremlin[it]: it runs a 2.4.18er linux kernel Jul 17 09:29:38 heheh too easy :) .. thanks zecke :) Jul 17 09:30:03 gremlin[it]: and the kernel is likely to deserve major clean up... Jul 17 09:33:40 a thig i cannot find is if 1000/3000/3100 have a graphic chip or not ... Jul 17 09:54:56 gremlin[it]: I think it uses the pxafb Jul 17 10:00:07 yes seem is it ... so could be easy to port linux (graphic chip seem really closed source) .. :( Jul 17 10:02:02 i'm looking for a nice Linux pda ... ASUS MyPal 730w hand wifi on sdio so is unusable ;( Jul 17 10:02:26 dell axim seem good ... except bigger one that use intel graphic ... Jul 17 10:02:47 gremlin[it], you mean e16/e17? Jul 17 10:03:00 axim ? Jul 17 10:03:20 i am not done the wm for embedded devices ... maybe in a month or so Jul 17 10:03:22 i'm looking on dell axim x50 serie ... Jul 17 10:04:00 emte, what are u working on ? Jul 17 10:04:19 the embedded wm for e17 Jul 17 10:04:28 dell e17 ? ... Jul 17 10:04:49 enlightenment is not device specific Jul 17 10:05:11 er well ... i wouldnt try to compile it in windows anyway Jul 17 10:05:25 anyother platform and you should be fine Jul 17 10:05:25 ahhh ... i remember i user enlightenment many years ago on desktop :) Jul 17 10:31:20 * dent bows to you Jul 17 10:31:45 anyone up for a stupid question? - what's the default password in base-passwd? Jul 17 10:34:40 nothing i belive Jul 17 10:35:23 emte: nope - or i'm too stupid to type nothing on my dropbear prompt ;) Jul 17 10:35:45 mhhh try 'rootme' ... but should be empty Jul 17 10:36:04 via ssh ... well if you havent set it yet i dont think you can ssh in Jul 17 10:36:31 rigght u first have to do the initial setup Jul 17 10:36:32 emte hmm, sucks - since i cannot connect via the conole :( Jul 17 10:36:52 use the touch screen ? Jul 17 10:37:04 gremlin: usb-serial Jul 17 10:37:15 no, he is telling you Jul 17 10:37:17 gremlin: no display at all Jul 17 10:37:39 guys.. this is still not a distro support channel Jul 17 10:37:40 then you will need to edit your base-passwd adn set something before buil;ding the image Jul 17 10:37:59 hmm Jul 17 10:38:13 byyyyyyyyeeeeeeeeeeeeeee Jul 17 10:38:23 emte is there some neat way without forking my private base-passwd bb? Jul 17 10:38:28 reenoo_, how is trying to connect via ssh to a device with no display distro specific? ... Jul 17 10:38:53 emte: it's not at all OE related. which makes it off topic here Jul 17 10:39:15 pretty sure the default base-passwd is OE related ... Jul 17 10:39:34 no, that's a distro decision Jul 17 10:39:38 then again maybe not Jul 17 10:40:13 sorry guys if i asked an off-topic question Jul 17 10:40:16 well dent i cant help you then Jul 17 10:40:31 maybe try #oe-butnotoespecific Jul 17 10:40:32 emte: thanks for trying! Jul 17 10:40:54 emte: well i'm using a vanilla image from oe Jul 17 10:41:09 reenoo_ says too bad, thats distro specific Jul 17 10:41:11 emte: so... i had the wrong impression this could be oe related Jul 17 10:41:19 * dent shrugs Jul 17 10:41:25 * dent thanks emte Jul 17 10:45:36 cya Jul 17 11:03:40 * kergoth chuckles at the current userfriendly Jul 17 11:09:37 mickey_away: there's something incompletely implemented about this branch deletion stuff. I keep getting zecke from vanille: Jul 17 11:09:39 monotone: bytes in | bytes out | revs in | revs out | revs written Jul 17 11:09:39 monotone: 15.4k | 55.1k | 2 | 25 | 0 Jul 17 11:09:40 monotone: warning: Dropping branch certs for unwanted branch org.openembedded.zecke Jul 17 11:09:57 * chouimat is away: stuff Jul 17 11:10:08 jbowler: did you set somewhere to not pull.zecke? Jul 17 11:10:18 Yes Jul 17 11:10:42 jbowler: Oh, BTW, I'm now getting sqlite3 missing errors for monotone-5 Jul 17 11:10:46 Here is what I sync: 'org.{nslu2-linux.*,openembedded.{dev,dreambox,nslu2-linux}}' Jul 17 11:11:10 JustinP: then you need to build sqlite... Jul 17 11:11:28 It's probably missing from the DEPENDS. Jul 17 11:11:35 I put it in there and no change Jul 17 11:11:52 I also just tried bitbake sqlite3 and it didn't compile anything extra... Jul 17 11:12:47 Ah, no - it uses a built in version. Jul 17 11:13:24 --with-bundled-sqlite Jul 17 11:14:35 ok.... Jul 17 11:14:46 what did you do to exclude the zecke branch? Jul 17 11:22:02 <_law_> monotone: warning: caught bad_decode exception decoding input from peer monotone.vanille.de: 'protocol version mismatch: wanted '4' got '5'' Jul 17 11:22:05 I sync the branches explicitly - see above. Jul 17 11:22:20 <_law_> any ideas why i cant update? Jul 17 11:22:25 _law_: upgrade monotone to 0.20 Jul 17 11:22:34 <_law_> ok Jul 17 11:24:11 <_law_> monotone: misuse: no branch pattern given and no default pattern set Jul 17 11:24:14 <_law_> :-( Jul 17 11:25:31 _law_: read the website or mickey's posting to oe@ Jul 17 11:25:36 mickeyl: those 25 revs are because I got the nslu2-linux pattern, not the openembedded one, so that's ok. Jul 17 11:27:12 The corrected 'default pattern' is: 'org.openembedded.{dev,dreambox,nslu2-linux}' and the only problem seems to be that monotone.vanille.de has org.openembedded.zecke. Jul 17 11:31:02 <_law_> jbowler: where an how can i set them? i only do monotone pull; monotone update Jul 17 11:31:33 monotone --help Jul 17 11:37:31 If anyone gets it wrong, the default pattern can be unset using the 'unset' command and set explicitly with the 'set' command. Jul 17 11:37:59 For read-only access it is only necessary to pull the branch you actually want (e.g. org.openembedded.dev - no wild cards). Jul 17 11:42:35 * chouimat is back. Jul 17 11:43:30 good evening, all ! Jul 17 11:44:44 i have this error when tryong to launch snes9x : " Xlib : extension "XFree86-DGA" missing on display ":0.0" " Jul 17 11:45:55 That's not an error. Jul 17 11:45:59 libxxf86dga1_0.0cvs20050704-r0_armv5te.ipk is installed, do i have to activate it or something ? Jul 17 11:46:21 pb_: then why does my screen remain blank ? Jul 17 11:46:40 I don't know. But lack of DGA should not be a problem. Jul 17 11:48:00 It sounds like either your SDL library or snes9x itself is broken. Jul 17 11:48:10 well, first, DGA was a dependency for snes9x, second, only the display of the game isn't here. snes9x itself displays infos about the loaded rom... what can it come from SDL ? Jul 17 11:48:22 hehe... Jul 17 11:48:32 pb_: you are much faster than me. Jul 17 11:48:49 snes9x has definitely worked on the iPAQ in the past, and the xserver on those machines has never supported DGA. Jul 17 11:49:05 ok. let's look for SDL. Jul 17 11:51:59 pb_: looks like SDL was not inculded by default in gpe-image for Openzaurus c7x0. Jul 17 11:52:29 that seems likely Jul 17 11:52:37 I don't think any programs in the default set use SDL. Jul 17 11:52:53 ho ... that makes sens... Jul 17 11:57:13 <_chronic> help anyone? after migrating to mt 0.20, from bk, i'm now getting a do_patch() error on one hunk of gnu-config-native-0.1cvs20050717 (for familiar). tried clearing out tmp/ with same results. should i blow away source directory as well to be sure? Jul 17 11:58:32 _chronic: monotone pull && monotone update Jul 17 11:58:53 <_chronic> thx. reenoo_, will try .... Jul 17 12:06:37 <_chronic> hmmm. mt pull && mt update gives me "monotone: misuse: working copy directory required but not found" Jul 17 12:13:43 _chronic: check the oe website for the right way to update Jul 17 12:16:41 <_chronic> JustinP: yup. followed directions on http://oe.handhelds.org/cgi-bin/moin.cgi/MonotonePhraseBook ... Jul 17 12:16:55 <_chronic> might try once more from scratch .... Jul 17 12:18:01 _chronic: you have to run update from within a working copy Jul 17 12:18:32 _chronic: or tell monotone where to find your working copy Jul 17 12:19:23 <_chronic> reenoo_: aha. thanks. that fixed it! :-) Jul 17 12:39:16 hi Jul 17 12:50:54 jbowler: could I get copies of the monotone-5 ipk you got for the C3K? My boost is compiling fine but I'm still getting errors with monotone Jul 17 12:51:45 JustinP: I deleted the whole build tree - no disk space... Jul 17 12:52:58 ::sigh::....oh well Jul 17 12:53:58 ell now I'm getting: Jul 17 12:53:59 | database.cc: In member function `void database::load(std::istream&)':| database.cc:355: error: `sqlite3_complete_last' undeclared (first use this function) Jul 17 12:56:17 What glibc are you using? Jul 17 12:57:15 host machine or in OE? Jul 17 12:57:20 In particular, is it: glibc-2.3.5+cvs20050627-r0 Jul 17 12:57:23 (OE) Jul 17 12:57:52 0726-r17 Jul 17 12:57:54 oops Jul 17 12:58:04 what the? Jul 17 12:58:06 -r17 of what? Jul 17 12:58:26 2004?? Jul 17 12:58:27 glibc-2.3.2+cvs20040726-r17 Jul 17 12:58:46 Dang. Hum.... still. Jul 17 12:58:55 (this was compiled on the 19th of this month too) Jul 17 12:59:04 I didn't set anything as well, this is what OE is using Jul 17 12:59:55 Ok. I figured that maybe you could pick up the ipk from the openslug feed. It might work, but it will object to the old glibc. Jul 17 13:00:53 Ah... no, it's an XScale but it's running little endian and Linux does not support swapping the endianness. Jul 17 13:01:34 03jbowler 07org.oe.nslu2-linux * r6b3cc2b0... 10/packages/ (6 files in 4 dirs): Jul 17 13:01:34 Horrible hacky, maybe working, work round for shutdown -r. Please report Jul 17 13:01:35 any future hangs on shutdown -r in slugbug 145. Jul 17 13:04:27 JustinP: ./sqlite/sqlite3.h:int sqlite3_complete_last(const char *sql, const char **last); Jul 17 13:05:28 And that file is included on line 22 of database.cc Jul 17 13:06:28 And my build, which succeeds, achieves this with the following (from log.do_compile): Jul 17 13:06:31 ccache armeb-linux-uclibc-g++ -march=armv5te -mtune=xscale -DLOCALEDIR=\"/usr/share/locale\" -DHAVE_CONFIG_H -I. -I/home/work-tmp/jbowler/nslu2/osuclibc/work/monotone-5-0.20-r0/monotone-0.20 -I. -I/home/work-tmp/jbowler/nslu2/osuclibc/staging/armeb-linux-uclibc/include -I/home/work-tmp/jbowler/nslu2/osuclibc/work/monotone-5-0.20-r0/monotone-0.20/lua -I/home/work-tmp/jbowler/nslu2/osuclibc/work/monotone-5-0.20-r0/monotone-0.20/sqlite -DNDEBUG Jul 17 13:06:32 -DBOOST_DISABLE_THREADS -Wall -I/home/work-tmp/jbowler/nslu2/osuclibc/staging/armeb-linux-uclibc/include -Os -fpermissive -c -o monotone-database.o `test -f 'database.cc' Jul 17 13:07:38 So I wonder if you have an sqlite subdirectory with an sqlite3.h in staging/arm-linux/include Jul 17 13:08:08 actually...I very well might Jul 17 13:08:50 I have an sqlite3.h in staging/arm-linux/include Jul 17 13:08:54 that's probably it Jul 17 13:09:20 Well... it should match the one in monotone I think... Jul 17 13:10:33 guys : what libsdl is required on GPE ? qpe right ? Jul 17 13:10:38 x11 Jul 17 13:10:50 qpe is for... qpe, I suppose. Jul 17 13:10:52 jbowler: actually it doesn't...one one in staging's only difference is it has no sqlite3_complete_last Jul 17 13:11:01 pb_: ok. Jul 17 13:11:09 jbowler: and it's version 3000000 instead of 3002002 Jul 17 13:11:10 pb_: thanks. Jul 17 13:12:04 JustinP: sounds like you need to either update sqlite.bb to a newer version and make monotone use it, or teach monotone to be more careful with its CFLAGS when using its included copy of sqlite. Jul 17 13:12:36 It might be possible to fix by zapping TARGET_CPPFLAGS Jul 17 13:13:04 I'm not sure where that particular staging -I comes from - there are two (and there don't need to be two...) Jul 17 13:18:21 I moved away that sqlite3.h and am trying again... Jul 17 13:18:51 I suppose upgrading sqlite3 and changing the monotone build to use the lib might be a better solution, though.. Jul 17 13:19:38 The problem seems to be 'CPPFLAGS' in the Makefile, I'm guessing it is TARGET_CPPFLAGS, I think simply setting it to empty will be fine... Jul 17 13:22:19 yeah, that sounds about right Jul 17 13:22:35 I'm not quite sure why we set that variable. Jul 17 13:22:48 afaik, all the -I options are also in TARGET_CFLAGS. Jul 17 13:25:14 Yes - I'm trying it with CPPFLAGS set to empty, since TARGET_CPPFLAGS still contains staging, and this is added to TARGET_CFLAGS that should be ok. Jul 17 13:25:39 The weirdness inside the monotone makefile is that it includes both $(CPPFLAGS) and $(CFLAGS) on the command line. Jul 17 13:26:15 That's pretty standard. I think OE is in error; it shouldn't be putting those -I flags in both places. Jul 17 13:26:51 ok....I'm trying with the moved away sqlite3.h for now... Jul 17 13:27:23 I'm guessing it puts them in CFLAGS in order to work with packages that don't respect CPPFLAGS. Maybe it puts them in CPPFLAGS too to help with packages that override CFLAGS internally. Jul 17 13:27:41 or maybe that is just a bug. who knows. Jul 17 13:28:04 too bad I can't "unstage" sqlite3 Jul 17 13:51:43 well hallelujah. moving away the sqlite3.h fixed my monotone build Jul 17 13:51:47 now to test the ipk Jul 17 13:56:17 dam ! libsdl is not the reson why snes9x won't run... any one has it working ? Jul 17 14:02:29 jbowler: yeah something's buggy here. I removed the zecke branch about 5 times today then when it always came back after someone pulling I stopped. Guess we need to wait until monotone changes the db layout to make that happen Jul 17 14:02:51 db layout changes are always a good chance to make larger changes, so we should use that chance then Jul 17 14:03:15 Building monotone I get "/usr/include/boost/regex/v4/char_regex_traits.hpp:68: error: 'boost::re_detail::c_traits_base::' is/uses anonymous type" Jul 17 14:03:51 time to sleep for me. Good luck gentlemen. Jul 17 14:03:56 I am using boost 1_32_0 Jul 17 14:04:01 mickeyl: if you change the serve list on vanille.de to be just org.openembedded.{dev,nslu2-linux,dreambox} it won't come back. Jul 17 14:04:32 jbowler: good point.... Jul 17 14:04:36 The reason it is coming back is that someone still has it and is syncing. Jul 17 14:05:18 monotone.nslu2-linux.org is already set up with that list, so it shouldn't come down from NSLU2. Jul 17 14:06:03 Ah, 'read_permitted' and 'write_permitted' take a branch... how about if branch == "org.openembedded.zecke" return false? Jul 17 14:06:05 jbowler: how would i do that? Jul 17 14:06:15 jbowler: no, write_permitted no longer takes a branch Jul 17 14:06:41 also ignore_branch didn't work at all here Jul 17 14:07:29 Hum. Ok. montone --db==whatever serve monotone.vanille.de 'org.openembedded.{dev,nslu2-linux,dreambox}' Jul 17 14:07:29 ah, you mean instead directly telling it what branches it servers, instead of using a glob. yeah, i can do that Jul 17 14:08:22 ya, sure. that will limit it spread around, however there's no way to keep it from getting readded to me Jul 17 14:08:30 * mickeyl restarts server Jul 17 14:09:04 If you kill the branch before restarting the server it will never get onto the server. Jul 17 14:09:13 lets see Jul 17 14:09:37 mickeyl: heh, maybe you should set up a cron job to remove the zecke branch every hour. Jul 17 14:09:48 pb_: heh, yeah Jul 17 14:10:04 who knew that zecke could be so persistent. Jul 17 14:10:26 well, actually we could've guessed from his nickname Jul 17 14:10:39 zecke means 'tick' Jul 17 14:11:50 i got this at the end of build rootfs for bootstrap-image: "log_check: There were error messages in the logfile","-e log_check: Matched keyword: [Cannot find package]" but i don't see any such error in log file. is it script bug ? Jul 17 14:12:48 probably. that log checker thing is pretty bogus. Jul 17 14:13:20 mickeyl: aha. hehe. Jul 17 14:13:55 jbowler: starting tests for monotone build Jul 17 14:19:07 * chouimat|ibook is away: food Jul 17 14:20:12 03mickeyl 07org.oe.dev * r71e99617... 10/packages/libpcap/ (libpcap-0.9.3/shared.patch libpcap_0.9.3.bb): add libpcap 0.9.3 Jul 17 14:20:16 03mickeyl 07org.oe.dev * ra2850b44... 10/packages/tcpdump/tcpdump_3.9.3.bb: add tcpdump 3.9.3 Jul 17 14:40:52 03rw 07org.oe.dev * re28a16dc... 10/packages/classpath/classpath_0.16.bb: classpath: upgrade to 0.16 Jul 17 14:40:56 03rw 07org.oe.dev * rd66ab84b... 10/packages/classpath/ (classpath_0.14.bb classpath_0.15.bb classpath_0.16.bb): classpath: fix DEPENDS Jul 17 14:41:00 03rw 07org.oe.dev * r1543b56c... 10/packages/sablevm/sablevm/no-internal-libs.patch: sablevm(-classpath): upgrade to 1.12 and work around internal libs nonsense Jul 17 14:41:11 ha ! i did my first bootstrap-image :) Jul 17 14:41:16 congrats aquadran Jul 17 14:41:24 thanks Jul 17 14:51:25 03mickeyl 07org.oe.dev * r25e31bea... 10/packages/udev/ (12 files in 5 dirs): Jul 17 14:51:25 udev: Jul 17 14:51:25 - remove a couple of obsolete versions Jul 17 14:51:25 - add 063 (default preference = -1, see inside for why) Jul 17 15:00:05 * chouimat|ibook is back. Jul 17 15:04:50 * reenoo_ kills org.openembedded.zecke one more time Jul 17 15:05:11 ~lart udev Jul 17 15:05:11 * ibot gets a hotmal account and SPAMs udev Jul 17 15:05:13 3079 root Z < [udevd] Jul 17 15:05:17 *sigh* Jul 17 15:07:04 * reenoo_ eyes at CIA-2 Jul 17 15:07:17 03rw 07org.oe.dev * r815c5bbb... 10/packages/jamvm/ (jamvm-1.3.0 jamvm): jamvm: use a single dir for patches Jul 17 15:07:20 heh Jul 17 15:07:21 03rw 07org.oe.dev * r72a10d79... 10/packages/jamvm/ (jamvm jamvm-1.3.0): disapproval of revision 815c5bbbed38e5cb537b9d7103dfcafe0689861b Jul 17 15:07:25 03rw 07org.oe.dev * r13209ce0... 10/packages/jamvm/ (jamvm-1.3.1/size-defaults.patch jamvm_1.3.1.bb): jamvm: upgrade to 1.3.1 Jul 17 15:07:39 reenoo_: we're still using a 5 minutes poll timer for CIA Jul 17 15:07:55 ah, ok Jul 17 15:08:17 i'm planning to use the new netsync notify hook in 0.20, but only when i have time Jul 17 15:08:36 sure. no problem. Jul 17 15:09:08 I just didn't know it was poll based Jul 17 15:11:21 did anything important improve in mt 0.20? (thinking about speed or the propagate issue) Jul 17 15:11:53 i think both have been adressed. it seems a bit faster and rwhitby-away told me that it's now possible to have private branches Jul 17 15:12:02 for details you need to ask him Jul 17 15:13:17 pull (and especially the intial one) is still as slow as before here. otherwise it feels a bit faster but that may be due to the addition of progress info to the console output Jul 17 15:13:59 s/otherwise/other than that/ Jul 17 15:14:12 hm ok Jul 17 15:14:15 mickeyl: the private branch thing is a bugfix which means that if you have a private branch on your local db, then it doesn't need to be pushed to remote DB's for the revisions that are merged to other branches to appear in the remote db Jul 17 15:14:41 rwhitby-away: but propagate still merges with a ancient revision, over and over? Jul 17 15:15:32 any udev experts in da house? Jul 17 15:17:42 tmbinc: dunno Jul 17 15:22:44 jbowler: tests still running....some have failed but they seemed to either be CVS related or "expected failures" Jul 17 15:24:36 and 2 unexpected passes...hmm... Jul 17 15:26:22 Yes, I got the unexpected passes - I believe the unencode of a b64 file fails and thereby, somehow, causes the test to appear to pass. Jul 17 15:26:48 The CVS is expected if you are running as root (CVS objects to doing CVS stuff as root). Jul 17 15:28:48 oh yeah, forgot about that Jul 17 15:29:02 there was also a netserve error I think Jul 17 15:29:09 on #137 right now Jul 17 15:29:21 but it might have been expected Jul 17 15:33:05 ~lart t-online Jul 17 15:33:05 * ibot cuts off t-online's head with a halberd that could have been a little bit sharper Jul 17 15:36:07 ~lart unholy autoconf recipes Jul 17 15:36:08 * ibot chops unholy autoconf recipes in half with a free Solaris 7 CD Jul 17 15:51:28 g'night Jul 17 15:51:36 'night mickey|zzZZzz Jul 17 15:57:54 mickey|zzZZzz: night night Jul 17 16:00:04 ::sigh::...an initial sync against a server on a cable modem...this should take...oh, only a few days to finish Jul 17 16:00:53 holy crap....I'm running out of space due to these monotone tests... Jul 17 16:02:06 Eh, yes, some of them are pretty big. There's a couple of hundred. Jul 17 16:02:48 Still, if you've got a substantial proportion of passes you can be pretty sure it's ok. Jul 17 16:05:52 I'll let it finish....I have a little I can remove Jul 17 16:19:49 gmp-native 2.4.2 currently doesn't compile with gcc4, whereas gmp 4.1.4 does (and at least here, on 32bit x86, it passes all tests). is anybody against using the newer version? (there's no maintainer in that package) Jul 17 16:19:57 otherwise i'll commit Jul 17 16:28:47 'night all Jul 17 16:29:33 i'll take that as a "no, just do it." Jul 17 16:32:29 oh well, org.openembedded.zecke is back Jul 17 16:33:04 03tmbinc 07org.oe.dev * r905c8d45... 10/packages/gmp/ (gmp-4.1.4/amd64.patch gmp-4.1.4/configure.patch gmp_4.1.4.bb): upgrade gmp to 4.1.4: old version did not build with gcc4 Jul 17 17:04:38 * JustinP looks into why parted is failing.... Jul 17 17:04:56 hmmm, it seems to all be warnings Jul 17 17:05:02 why is it erroring out... Jul 17 17:06:31 staging/arm-linux/include/wchar.h:336: warning: comparison of unsigned expression >= 0 is always true (in function wctob) Jul 17 17:09:27 Harmless warning. The code is written that way so that it works regardless of signedness. Jul 17 17:11:30 ok, then why does OE stop compiling.... Jul 17 17:11:42 make is giving Error 1 Jul 17 17:13:08 hmmmm... Jul 17 17:14:42 that's the only non-standard thing I can see....those warnings Jul 17 17:18:18 Someone has switched on the warnings-are-errors flag to gcc Jul 17 17:18:35 aha Jul 17 17:21:19 yep, -Werror... Jul 17 17:22:08 Bet there's a Wall in there too, and probably a whole load of extra warnings turned on... Jul 17 17:23:29 probably Jul 17 17:24:43 yeo, -Wall Jul 17 17:24:45 yep Jul 17 17:27:28 This is very fine programming but you have little choice but to kill the Werror because the GNU folks ship header files which won't compile with their own compiler... Jul 17 17:28:20 Not that they're alone in this. Jul 17 17:29:44 I'm trying --diable-Werror Jul 17 17:29:47 disable Jul 17 17:31:25 I think last time I encountered the problem I just added something to the end of CFLAGS - the last setting gcc receives takes effect. Jul 17 17:31:49 hmmm...I'm trying EXTRA_OECONF Jul 17 17:32:20 there's --enable-Werror in the configure script so.... Jul 17 17:33:55 Ah, I guess it must work somewhere then... Jul 17 17:34:19 yes, it worked :-) Jul 17 17:42:40 another bb file down :-) just have to test it Jul 17 17:52:04 does anyone knows about the state of the "madwifi" package? Jul 17 17:54:30 and is there a reason to use 'printf "" ""' instead of 'echo -n'? Jul 17 17:54:44 or even just > ? Jul 17 18:08:47 * JustinP has no idea.... Jul 17 18:09:49 madwifi seems broken (or just outdated?), and "> ${D}...resolv.conf" seems to do it fine :) Jul 17 18:18:15 cool, everything required for my "mobile kismet machine" (based on a wgt634u access point) compiled and packaged well, and it does even run :)) Jul 17 18:18:33 openembedded is cool :) Jul 17 20:01:42 hi Jul 17 20:04:35 hi Jul 17 20:11:07 hi Jul 17 20:11:34 you said that before Jul 17 20:12:06 how are u Jul 17 20:12:11 how are u Jul 17 20:12:13 how are u Jul 17 20:12:14 how are u Jul 17 20:12:14 how are u Jul 17 20:12:25 lol Jul 17 20:12:27 lol Jul 17 20:12:32 lol Jul 17 20:12:33 lol Jul 17 20:12:34 lol Jul 17 20:12:42 takl Jul 17 20:12:43 takl Jul 17 20:12:44 takl Jul 17 20:12:45 takl Jul 17 20:12:46 takl Jul 17 20:12:50 takl Jul 17 20:12:50 takl Jul 17 20:12:51 takl Jul 17 20:12:53 takl Jul 17 20:12:53 takl Jul 17 20:12:53 takl Jul 17 20:12:55 takl Jul 17 20:14:54 stop spamming us Jul 17 23:25:39 good morning. Jul 17 23:44:09 hi all! Jul 17 23:44:13 hi koobla Jul 17 23:44:46 alan: hi! Jul 17 23:45:05 good to see someone else has to work too ;) Jul 17 23:46:38 i have this message error when tryng to compile micro-emacs : "NOTE: Task failed: Fetch failed: http://www.jasspa.com/release_040301/jasspa-memacros-20040301.tar.gz". It's not my first error when trying to fetch from www.jasspa.com. Is it an error from them or should i update something ? Jul 17 23:47:07 koobla: i'm on hollidays, but i always get up early... Jul 17 23:47:56 maybe hte site changed - look with a browser if the files are still there Jul 17 23:48:14 no they are not where they should be. Jul 17 23:48:29 should i update something ? and what ? Jul 17 23:48:54 ok- - tried to paste teh package into mozilla - got a 404 error Jul 17 23:49:39 yeah... i downloaded the previous file (from the previous error) and putted it in /stuff/oe/sources... Jul 17 23:49:56 it solved the problem, but that is not a long term solution... Jul 17 23:50:10 okay -- for the cache, i think you also need an md5 sum file Jul 17 23:51:35 else, copy the bb to reflect the new version number and update the download url Jul 17 23:52:07 good idea. i'm going to check this. Jul 17 23:53:58 * koobla slaps pxa devboard Jul 17 23:55:38 koobla: you were right. the two needed files url's were coded in hard in the .bb file. They were only two, so i downloaded those files from somewhere else, but the .bb file needs to be updated. Jul 17 23:56:11 i'll update my branch and see if the .bb file is correct. Jul 17 23:56:13 is there a maintainer stated in the bb Jul 17 23:56:35 Jason Haslup Jul 17 23:56:59 well, mail it to him if it works Jul 17 23:57:08 ok. Jul 17 23:57:49 koobla: thanks for the infos. I'm a newbie in all this. Jul 17 23:58:48 alan: no problem. spent some time learning all this myself, and got plenty of help on this board Jul 17 23:59:00 ^^ Jul 17 23:59:19 alan: but i am by no means a pro - just fiddled around with the code a bit, and got my first small images running **** ENDING LOGGING AT Sun Jul 17 23:59:57 2005