**** BEGIN LOGGING AT Wed Jul 27 23:59:56 2005 Jul 28 00:12:06 hi Jul 28 00:26:51 morning Jul 28 00:27:11 and brb Jul 28 00:29:14 morning again Jul 28 00:30:02 morning hrw|work Jul 28 00:30:33 hmm.. I must reread what exactly VIRT column in top means.. Jul 28 00:30:41 Xorg has 311m there Jul 28 00:59:52 why do_fetch is nostamp? Jul 28 01:10:10 hi BigAl Jul 28 01:22:38 moin Jul 28 01:22:57 hi CoreDump|home Jul 28 01:26:20 hey hrw|work Jul 28 01:26:34 good morning Jul 28 01:26:49 hi florian_kc Jul 28 01:32:00 morning hrw Jul 28 01:33:21 morning florian_kc CoreDump|home Jul 28 01:37:13 morning all Jul 28 01:37:27 hey RP Jul 28 01:37:48 morning RP Jul 28 01:37:54 hi RP Jul 28 01:39:27 RP: 2.6.13-rc-mm2 is available Jul 28 01:40:57 do13: Ok, thanks - I'll upgrade soon, although maybe tomorrow as I'm in the middle of a patch sorting session :) Jul 28 01:43:56 RP: tosa compiles with this patch http://www.do13.de/openzaurus/temp/linux-2.6.13-rc3-mm2.patch - One prob: preempt_nwfpe1-r0 patch doesn't apply Jul 28 01:51:09 Is there some simple way to make OE compile natively only? Jul 28 01:52:08 target-arch=native or something Jul 28 01:53:07 MACHINE=native Jul 28 01:53:14 cool Jul 28 01:53:39 someone know why do_fetch is nostamp? Jul 28 01:58:34 iirc the .md5 is used like a stamp Jul 28 01:59:32 it is - but if I fetched all then why bitbake has to do_fetch() again.. Jul 28 02:04:47 hey koen, you wouldn't happen to have an already build OE tree on that host I get 6mbit to? :P Jul 28 02:04:57 :) Jul 28 02:06:05 this is going to take fscking for ever to build on my laptop :P Jul 28 02:06:38 i just want to add a few OE packages, so shell acess would be another option ;P Jul 28 02:06:40 * koen waits for zecke to come online to explain how to fix h*00.conf Jul 28 02:08:45 mithro: be sure to 'merge' after you 'sync' Jul 28 02:11:21 * mithro does a -c fetch task-bootstrap first Jul 28 02:14:12 do13: Ok, I need to rework the nwfpe code anyway - rmk has asked me for an updated version! :) Jul 28 02:17:41 "richard, or something, could you send me, as the original maintainer an update to the nwfpe code?" Jul 28 02:18:28 "What is the status of this patch? Ralphs recalls that the only final objection was all the reporting of inexact results, though this can probably be solved with a FAQ entry. If it is to be applied, could an updated patch be supplied please?" Jul 28 02:18:54 No mention of my name ;-) Jul 28 02:20:42 you could mention 'EABI' in your patch ;) Jul 28 02:21:00 but that will render RMK unuseable for a week Jul 28 02:21:00 More than my life's worth ;-) Jul 28 02:31:20 re Jul 28 02:31:30 re zecke Jul 28 02:31:31 moin zecke__ Jul 28 02:32:45 zecke__: how do I fix h3*00.conf? Jul 28 02:33:49 zecke__: OVERRIDES .= "${KERNEL}" ? Jul 28 02:37:46 yes Jul 28 02:40:17 ~lart sf.net for being lagged Jul 28 02:40:17 * ibot executes killall -TERM sf.net for being lagged Jul 28 02:48:18 Where was the list of monotone backup servers? Jul 28 02:48:25 * koen is a bit puzzled with monotone add and branches Jul 28 02:48:28 I remember seeing it but can't find it atm Jul 28 02:48:50 oe.handhelds.org Jul 28 02:48:55 koen: thanks Jul 28 02:49:37 Yeah.. In the front page.. *tries to hide* Jul 28 02:57:57 koen: I was playing with a monotone server and viemtn, which version of mt do you run for viewmtn to work? Mine borks with the "depth" option with v0.20 Jul 28 02:58:41 depth was replaces Jul 28 02:58:51 which line was that? Jul 28 02:59:02 umm checking Jul 28 02:59:42 hmm.. I want opie/x11 - many apps will be removed then Jul 28 03:00:16 http://oz.hentges.net/viewmtn-0.03/revision.psp?id=93b2c42bded24d8cdc33b9d4388d8caa90b22fd1 Jul 28 03:01:39 s/depth/last/ Jul 28 03:03:09 ~hail pb_ Jul 28 03:03:10 * ibot bows down to pb_ and chants, "I'M NOT WORTHY!!" Jul 28 03:03:29 CoreDump|home: change --depth into --last Jul 28 03:03:41 koen: thx :) Jul 28 03:03:42 hail zecke of the many underscores Jul 28 03:05:01 #debian.de hate | in the names so I decided to add a infinite number of underscores as indicator Jul 28 03:05:13 I've not yet settled on a proper encoding though Jul 28 03:06:24 pb_: how is recieva going? Will you attend the IFA in Berlin? Jul 28 03:07:23 CoreDump|home: make sure you only serve official branches: monotone --db=OE-koen.db serve 130.89.193.158 "org.openembedded.{dev,dreambox,nslu2-linux}" Jul 28 03:07:45 we don't want zecke and koen branches spreading again ;) Jul 28 03:07:53 heh Jul 28 03:08:03 * zecke__ started using push/pull instead of sync Jul 28 03:08:18 i wasn't intending to serve the OE branches :) Just want to get some practice Jul 28 03:10:48 haha someone broke the tinderbox Jul 28 03:11:32 /data/tinderbox2/data_dir/OpenEmbeddedBuild/db/Build.Update..OpenSlug.1122490090 Jul 28 03:11:35 now he hides Jul 28 03:12:02 koen: please remove that one Jul 28 03:12:29 done Jul 28 03:14:30 prcoessmail should not write that file... Jul 28 03:14:56 zecke__: if you log in and out you should be able to write to the tinderbox dirs Jul 28 03:16:19 zecke__: gzipping the http post might help too Jul 28 03:16:34 but I'm not sure it post likes binary data Jul 28 03:17:12 values can be gzipped Jul 28 03:18:14 koen: gzipping could work Jul 28 03:18:37 koen: it is more important to get more tinderclients running and regulary running Jul 28 03:18:48 koen: then the error parser needs to be fixed Jul 28 03:19:01 koen: OT: one can do httpz that should gzip data transparently Jul 28 03:19:25 yeah, I also thought of modgzip Jul 28 03:20:04 and make the server more robust about starttime=0 Jul 28 03:20:59 and convince someone to write a VC_monotone Jul 28 03:21:23 fetching source for whole OE takes time... Jul 28 03:21:47 especially with slow cvs servers Jul 28 03:23:13 BB>> getvar BBMASK Jul 28 03:23:14 (nylon-*|xscreensaver|minimo|mozilla|firefox|thunderbird|squeak-*|micro-emacs|ypserv|ifupdown|mailx|python-inotify|wrt-utils|microcom|bitbake|gumstix|ion3|dhclient) Jul 28 03:31:26 when finish I will work on it Jul 28 03:33:36 03koen 07unknown * r1e753c67... 10/packages/libgc/libgc_6.5.bb: first try at libgc Jul 28 03:33:40 03rwhitby 07org.oe.nslu2-linux * rd3e36f00... 10/packages/ (3 files in 3 dirs): Added /dev/video to openslug device_table.txt Jul 28 03:33:43 03koen 07unknown * rdd6071f2... 10/packages/inkscape/inkscape_0.42.bb: first try at inkscape 0.42 Jul 28 03:33:47 03rwhitby 07org.oe.nslu2-linux * r8d77e5cf... 10/packages/initscripts/initscripts-openslug_1.0.bb: Bumped the initscripts-openslug PR Jul 28 03:33:51 03koen 07unknown * rd2902082... 10/packages/libgc/libgc_6.5.bb: Fill in some blanks for gc Jul 28 03:33:59 rwhitby-away: hey Jul 28 03:34:02 03rwhitby 07org.oe.nslu2-linux * rc233def8... 10/ (3 files in 3 dirs): Added setpwc for controlling Philips webcams. Jul 28 03:34:06 03koen 07unknown * r27e14340... 10/packages/libgc/libgc_6.5.bb: make libgc compile Jul 28 03:34:12 koen: unknown is org.openembedded.koen? Jul 28 03:34:14 ah, mickey|zzZZzz restarted vanille.de Jul 28 03:34:15 03rwhitby 07org.oe.nslu2-linux * rbb229e5f... 10/packages/initscripts/initscripts-1.0/openslug/device_table.txt: Changed /dev/video to /dev/video[0-4] for openslug Jul 28 03:34:18 hrw|work: yes Jul 28 03:34:33 hrw|work: or any other branch Jul 28 03:34:40 03koen 07unknown * r1dd748dc... 10/packages/libgc/libgc_6.5.bb: stage libgc headers and libs Jul 28 03:34:50 03koen 07org.oe.dev * r4dadf63b... 10/conf/machine/h3900.conf: Attempt to fix the overrides situation in h3900.conf Jul 28 03:34:50 03mickeyl 07org.oe.dev * raaea884d... 10/packages/linux/openzaurus-pxa27x_2.4.20-rmk2-embedix20050228.bb: openzaurus-pxa27x: fix CMDLINE constructing, fix module_autoload for pxa27x ohci module Jul 28 03:34:53 03mickeyl 07org.oe.dev * r45798617... 10/packages/base-files/base-files/spitz/fstab: spitz fstab: reorder for more clarity Jul 28 03:34:57 03mickeyl 07org.oe.dev * r40dca932... 10/.mt-attrs: update attributes Jul 28 03:35:01 03mickeyl 07org.oe.dev * rf53aad80... 10/conf/machine/ (poodle.conf tosa.conf zaurus-clamshell.conf): s/EXTRA_BOOTSTRAP_DEPENDS/BOOTSTRAP_EXTRA_DEPENDS/ thanks Justin Jul 28 03:37:58 03koen 07unknown * rbb1a38be... 10/packages/inkscape/inkscape_0.42.bb: disable inkscape 0.42 for now Jul 28 03:39:12 treke|home: ping ;) Jul 28 03:40:23 * koen merges Jul 28 03:46:09 * koen reads the note on the tinderserver Jul 28 03:47:32 zecke__: Jul 28 03:47:32 OTE: exceptions.SyntaxError:EOL while scanning single-quoted string (line 1) while evaluating: Jul 28 03:47:35 kernel ipaq-boot-params ${@linux_module_packages('${H3900_MODULES}', d)} pcmcia-cs apm apmd network-suspend-scripts ipkg-collateral ipkg ipkg-link Jul 28 03:47:37 * florian_kc curses at bitbake Jul 28 03:51:13 zecke__: hey Jul 28 03:51:21 koen: updatet bitbake? Jul 28 03:51:32 have any OpenSlug or Unslung tinderclient posts got through yet? Jul 28 03:51:54 rwhitby: no, you seem to send a broken start time Jul 28 03:52:58 zecke__: TINDER_START := "${@time.strftime('%m/%d/%Y %H:%M:%S', time.localtime())}" Jul 28 03:53:22 zecke__: r294 Jul 28 03:53:32 zecke__: TINDER_TZ = "Europe/Berlin" Jul 28 03:53:33 zecke__: after that berlios went hostile on me Jul 28 03:53:59 rwhitby: how recent is your bitbake? Jul 28 03:54:10 I'm in preparation for a new bitbake release anyway Jul 28 03:54:24 zecke__: since before berlios.de went down last Jul 28 03:54:33 rwhitby: you send '0' as starttime Jul 28 03:54:38 rwhitby: try to update please Jul 28 03:54:57 does current bitbake trunk work with current OE head? Jul 28 03:55:12 rwhitby: it should Jul 28 03:57:50 ok, will update the org.nslu2-linux.bitbake branch now Jul 28 03:58:49 Hmm "svn co svn://svn.berlios.de/bitbake/trunk/bitbake" is not responding Jul 28 03:58:54 what a surprise .... Jul 28 03:59:09 rwhitby: it was back... Jul 28 03:59:19 what, for five minutes? Jul 28 03:59:34 * rwhitby is not impressed by berlios.de's uptime record ... Jul 28 04:00:06 zecke__: can you put a snapshot somewhere for me to download? Jul 28 04:02:17 sadly not Jul 28 04:02:30 is it small enough to email? Jul 28 04:03:00 lol I can not login to berlios as well Jul 28 04:03:49 http://ewi546.ewi.utwente.nl/OE/bitbake.tar.bz2 Jul 28 04:03:52 grab it fast Jul 28 04:04:00 koen: is it right I can write to that directory? Jul 28 04:04:12 got it, thx Jul 28 04:04:26 zecke__: correct Jul 28 04:04:40 rwhitby: http://dominion.kabel.utwente.nl/koen/pda/files/bbr294.tar.bz2 Jul 28 04:04:45 rwhitby: it is almost recent, it lacks the .= =. stuff but you probably do not need it anyway Jul 28 04:04:59 which one, koen's or zecke__'s ? Jul 28 04:05:22 r294 should be the latest and greatest Jul 28 04:05:26 but it shouldn't matter Jul 28 04:05:48 florian_kc: gkdial is not fetchable Jul 28 04:07:18 hrw|work: http://familiar.handhelds.org/source/v0.8.2/${PN}-${PV}.tar.gz is missing? Jul 28 04:07:31 SRC_URI = "http://www.kernelconcepts.de/~fuchs/gkdial-${PV}.tar.gz \ Jul 28 04:07:36 is missing Jul 28 04:07:46 that isn't gkdial :) Jul 28 04:07:56 that's gkdial-gnomeless_1.8.1.bb Jul 28 04:08:01 13:05 hrw@home:gkdial$ less gkdial-gnomeless_1.8.1.bb Jul 28 04:08:14 ok.. Jul 28 04:08:14 gkdial_1.8.1.bb fetches ok Jul 28 04:08:25 (don't know what the difference is) Jul 28 04:08:27 florian_kc: gkdial-gnomeless is not fetchable Jul 28 04:08:30 me too Jul 28 04:08:57 anyway florian_kc is maintainer for both Jul 28 04:09:02 koen: your bitbake had so much extra stuff in it that it scared me, so I used zecke__'s r293 :-) Jul 28 04:09:18 :) Jul 28 04:09:38 zecke__: did I update h3900.conf the right way? Jul 28 04:14:41 koen: the syntax is right Jul 28 04:14:49 ~lart ftp.easysw.com Jul 28 04:14:49 * ibot burns ftp.easysw.com to a crisp with a laser Jul 28 04:14:53 koen: you probably mean "${KERNEL}:" thougj Jul 28 04:15:20 koen: the operator looks right Jul 28 04:15:46 zecke__: ":foo" or "foo:" ? Jul 28 04:16:20 does it prepend or append? Jul 28 04:16:51 koen: good question, but pretty much like += and =+ Jul 28 04:17:23 what would happen if I do ":foo:" Jul 28 04:17:26 ? Jul 28 04:18:23 you would that append to OVERRIDES Jul 28 04:19:03 so OVERRIDES .= ":${KERNEL}" Jul 28 04:19:09 is the correct thing Jul 28 04:22:20 koen: no you want O =. "${K}:" Jul 28 04:24:41 someone has that rev in? e34e002afa05c5fa5be23cd743d8ca5ad2d2b90b hrw@openembedded.org 2005-07-28T09:02:26 Jul 28 04:26:13 ok. now it is pulling something Jul 28 04:26:25 hrw|work: that's obsolete... it is replaced by the 1.8.1 which includes a newer and better gpe patch. Jul 28 04:27:21 florian_kc: if gkdial-gnomeless is obsolete then why it is in repo? Jul 28 04:28:36 hrw|work: because noone removed it so far :-) Jul 28 04:29:00 florian_kc: you have r/w, you are maintainer of it so it is a job for you ;) Jul 28 04:29:09 03koen 07org.oe.dev * rcf227662... 10/conf/machine/h3900.conf: add : before KERNEL in h3900.conf Jul 28 04:29:21 zecke__: yes, I screwed up the message Jul 28 04:29:44 ok, did an Unslung tinderclient post come through? Jul 28 04:29:52 hrw|work: right Jul 28 04:30:35 rwhitby: no errors yet Jul 28 04:31:57 * koen crosses fingers Jul 28 04:33:11 checking whether m4 supports frozen files... no Jul 28 04:33:12 | configure: error: GNU M4 1.4 is required Jul 28 04:33:12 | FATAL: oe_runconf failed Jul 28 04:33:19 Is this a known one? Jul 28 04:35:18 rwhitby: yes it came through Jul 28 04:35:30 florian_kc: broken configure.ac? Jul 28 04:35:43 florian_kc: or Makefile.am requiring version 1.4 Jul 28 04:38:16 zecke__: It just built m4-native, it detects the correct one and seems to use it. Jul 28 04:38:45 zecke__: | checking for m4... /home/fuchs/bdisk/oe/sbox_nokia770/staging/arm-linux/bin/m4 Jul 28 04:38:54 zecke__: that's the line before... Jul 28 04:40:25 koen: you have viewmtn on ewi? Jul 28 04:40:33 yes Jul 28 04:40:39 uri? Jul 28 04:40:43 hrw|work: see http://oe.handhelds.org :) Jul 28 04:40:58 ;) Jul 28 04:40:59 ok Jul 28 04:41:24 ugly version.. Jul 28 04:41:46 my last changesets looks funny on graph Jul 28 04:48:45 03florian 07org.oe.dev * r0bd531ea... 10/packages/gkdial/gkdial-gnomeless_1.8.1.bb: Remove obsolete file. Jul 28 04:51:51 silo is not fetchable too Jul 28 04:56:00 03hrw 07org.oe.dev * r04f3645e... 10/packages/dash/dash_0.5.2.bb: updated to Debian 0.5.2-6 as 0.5.2-4 is not fetchable Jul 28 04:56:05 03hrw 07org.oe.dev * re34e002a... 10/packages/memtester/ (memtester_4.0.4.bb memtester_4.0.5.bb): updated memtester to 4.0.5 Jul 28 04:56:08 03hrw 07org.oe.dev * r6b7fc0bd... 10/packages/python/python-imdbpy_2.0.bb: fixed SRC_URI for python-imdbpy Jul 28 04:57:01 zecke__: looks like updating bitbake fixed the tinderclient. now the timezone just needs fixing :-) Jul 28 04:57:15 rwhitby: yes Jul 28 04:57:56 I'll wait for that before I start building periodically. Jul 28 05:15:40 vanille.de refuse ;( Jul 28 05:18:22 03mickeyl 07org.oe.dev * r9b27578c... 10/packages/hostap/ (files/hostap-fw-load.patch hostap-utils_0.3.7.bb): hostap-utils: fix firmware path for prism3 fw 1.8.4 Jul 28 05:18:26 03mickeyl 07org.oe.dev * r222a8ce2... 10/packages/curl/ (5 files in 3 dirs): add curl 7.14.0 and remove 7.12.2 Jul 28 05:18:30 03mickeyl 07org.oe.dev * rd04c0208... 10/packages/python/ (python-pycurl_7.12.2.bb python-pycurl_7.14.0.bb): upgrade python-pycurl to 7.14.0 Jul 28 05:38:08 03hrw 07org.oe.dev * r8e805ab2... 10/packages/dgen/ (dgen-sdl-1.18/fix-everything.patch dgen-sdl_1.18.bb): drop dgen-sdl 1.18 - it is unfetchable and whole project is non maintained by upstream Jul 28 05:38:13 03hrw 07org.oe.dev * rf73e4aee... 10/packages/slotsig/ (slotsig_0.4.bb slotsig_0.6.bb): updated slotsig to 0.6 Jul 28 05:44:48 heh.. 2heads :( Jul 28 05:49:43 merged Jul 28 06:13:59 so, does OE support ucLinux? Jul 28 06:14:29 sort of Jul 28 06:14:40 it support uclibc Jul 28 06:15:29 hmm.. my working copy has some missing files (which was removed by one of older changesets) and I'm on HEAD now.. whats wrong? Jul 28 06:16:38 but nobody has built it against a ucLinux kernel? Jul 28 06:16:40 http://pastebin.ca/18705 Jul 28 06:17:10 monotone list missing | xargs monotone co Jul 28 06:17:51 mithro: right Jul 28 06:18:30 koen: error Jul 28 06:18:53 checkout cannot get single files Jul 28 06:19:57 zecke__: hrm, somebody should kick these uclinux people in the butt so they start using oe too :P Jul 28 06:20:17 they seem pretty attached to their buildroot afaik Jul 28 06:20:32 looks like my working copy is b0rken now Jul 28 06:20:33 yes Jul 28 06:22:12 * zecke__ considers insulting mallum and starts Jul 28 06:22:29 mallum: your southpark charachter looks like big gay al ;) Jul 28 06:22:36 * zecke__ stops insulting nice guys Jul 28 06:23:02 mallum: happen to have any uclinux guy around? Jul 28 06:23:30 zecke__: no, we ate them Jul 28 06:23:52 zecke__: where's your southpark char? Jul 28 06:24:11 mithro: happen to have any uclinux guy around? Jul 28 06:24:40 koen: hehe. I'm that big gene manipulated chicken, I'm nothing and everything... Jul 28 06:25:17 he is talking out the front now Jul 28 06:25:39 I hope that I dont fscked something in repo today... Jul 28 06:28:36 mithro: http://wiki.whatthehack.org/index.php/Using_linux_for_embedded_devices ? Jul 28 06:33:45 florian_kc: did you manage to convince nilsf that iCal support in gpe-calendar is cool? Jul 28 06:35:00 03hrw 07org.oe.dev * r5813eabf... 10/packages/libssh/ (libssh_0.1.bb libssh_0.11.bb): updated libssh to 0.11 Jul 28 06:37:25 koen: yes Jul 28 06:37:57 sorry for spamming oe@handhelds.org. is it possible to submit a bug with attached patch in one step? Jul 28 06:38:45 pH5: no, bugzilla is a bit strange Jul 28 06:42:33 perl-native fails Jul 28 06:43:17 koen: is this really necessary? the only problem here is time... Jul 28 06:50:06 ~hail Debian Jul 28 06:50:07 * ibot bows down to Debian and chants, "I'M NOT WORTHY!!" Jul 28 06:50:39 one good thing in Debian 'stable' being old before next release... source archive ;) Jul 28 06:51:30 * france is away: Away Jul 28 06:57:17 ~lart perl-native-5.8.7 for making my tinder red: http://ewi546.ewi.utwente.nl/OE_qa/OpenEmbeddedBuild/status.html Jul 28 06:57:17 * ibot takes out a cattle prod and gives perl-native-5.8.7 a good jolt for making my tinder red: http://ewi546.ewi.utwente.nl/OE_qa/OpenEmbeddedBuild/status.html Jul 28 06:59:08 03hrw 07org.oe.dev * rbec0de74... 10/packages/atftp/atftp_0.7.bb: switched to fetching source from Debian 'stable' + patch from Debian to get it build with gcc 3.4 Jul 28 06:59:52 koen: is there a planet.hh.org now ? Jul 28 07:00:41 mallum: as soon as jamey reruns the fetcher you can view it at http://handhelds.org/~jamey/planet/ Jul 28 07:00:51 so: "not yet" Jul 28 07:06:24 03rwhitby 07org.oe.nslu2-linux * r9df4aaab... 10/packages/linux/ (nslu2-kernel_2.6.12.2.bb openslug-kernel-2.6.12.2/defconfig): Enabled INPUT and USB_HID modules in the OpenSlug kernel Jul 28 07:08:04 NAiL: is perl-native parallel-make safe? Jul 28 07:08:26 koen: no idea. I haven't messed with perl. DaKa2 is prolly the one to ask Jul 28 07:08:35 ah, ok Jul 28 07:08:39 DaKa2: is perl-native parallel-make safe? Jul 28 07:09:03 * koen tries -b for a third time Jul 28 07:10:08 planet updated Jul 28 07:10:51 jamey: earth? Jul 28 07:11:27 planet handhelds Jul 28 07:11:42 http://handhelds.org/~jamey/planet/ currently Jul 28 07:11:55 still experimental, though content seems pretty relevant now Jul 28 07:12:25 :-) Jul 28 07:12:42 koen: uhm.. no idea... Jul 28 07:12:52 I liked the lisp machines, they had a 'save world' command Jul 28 07:13:00 DaKa2: it isn't :) Jul 28 07:13:16 I didn't really touch the -native package.. just added a 5.8.7.. Jul 28 07:13:19 dammit.. :-) Jul 28 07:13:20 DaKa2: I've pushed (well, as soon as vanille responds again) a fix for it Jul 28 07:13:30 PARALLEL_MAKE="" Jul 28 07:13:58 ah, good :) did it work in the old package? Jul 28 07:14:01 re Jul 28 07:15:05 DaKa2: it seems it worked in the previous package Jul 28 07:15:15 DaKa2: but you never know with perl Jul 28 07:16:01 koen: strange.. :-) well, I looked at the native package and decided I better not start messing with it :) Jul 28 07:17:13 03koen 07org.oe.dev * re91215db... 10/packages/perl/perl-native_5.8.7.bb: Disable PARALLEL_MAKE for perl-native Jul 28 07:17:17 03hrw 07org.oe.dev * r5888828b... 10/packages/inputpipe/inputpipe_svn.bb: fixed SRC_URI for inputpipe, changed PV to something more normal Jul 28 07:17:59 * koen merges Jul 28 07:20:19 morning Jul 28 07:21:11 1re Jul 28 07:23:17 fsck.. I have to stop merging.. too much merging now Jul 28 07:24:19 03hrw 07org.oe.dev * r85ab954b... 10/packages/arpwatch/ (8 files in 2 dirs): Jul 28 07:24:19 added arpwatch 2.1a13 with some patches taken from Debian Jul 28 07:24:19 it needs some work - adding/removing "arpwatch" user in post(inst/rm) Jul 28 07:26:13 there, everything is merged again Jul 28 07:26:40 yep Jul 28 07:26:57 last changesets looks like war 'who will merge this time' Jul 28 07:27:55 03hrw 07org.oe.dev * r332ff521... 10/packages/linux/linux-geodegx_2.4.24.bb: fixed SRC_URI for linux-geodegx Jul 28 07:29:48 Errr - um Jul 28 07:30:07 What was wrong with the SRC_URI? Jul 28 07:30:19 Just so I know... :) Jul 28 07:30:52 2.4 vs v2.4 Jul 28 07:30:57 http://www.handhelds.org/hypermail/oe-commits/current/0438.html Jul 28 07:36:12 Eewww,,, Jul 28 07:36:36 This is why you should periodically test with an empty source repo, boys and girls Jul 28 07:36:43 :) Jul 28 07:37:47 CosmicPenguin: what do you think I'm doing last days? :) Jul 28 07:38:08 hrw|work: yeah, but *all* of us should do that Jul 28 07:38:12 no fair making you do it.. :) Jul 28 07:38:37 yep Jul 28 07:41:11 CosmicPenguin: zecke is setting up a fetching tinderbox Jul 28 07:41:39 good Jul 28 07:41:42 * CosmicPenguin prepares himself for more embaressement Jul 28 07:44:02 currently my dl_dir is 1.6G Jul 28 07:44:16 not counting cvs_dir ofcourse Jul 28 07:44:44 Still having trouble with the nunome package. The last problem was that nunome uses recursive make. mickey suggested "do_compile() { for i in dir; oe_runmake ...; done )" and I tried all kinds of variations on that but it still fails. Jul 28 07:45:33 I tried to graphically represent the make dependencies in http://rafb.net/paste/results/q6ud6c96.html As a step-by-step approach I tried to only compile the ui target with "do_compile() { cd ui;oe_runmake TARGET_DIR=${D} USR_DIR=${D}/usr;cd .. }" but OE still iignores the TARGET_DIR variable and tries to install into /plugins/inputmethods Jul 28 07:45:51 03hrw 07org.oe.dev * r27ecd8d7... 10/packages/rsync/ (rsync_2.6.4.bb rsync_2.6.5.bb): updated rsync to 2.6.5 Jul 28 07:47:21 do we have an EXTRA_OEMAKE ? Jul 28 07:47:29 morning Jul 28 07:47:34 hey mickeyl Jul 28 07:47:58 hi mickeyl Jul 28 07:48:07 moin mickeyl Jul 28 07:48:08 yes, EXTRA_OEMAKE we have Jul 28 07:48:10 mickeyl: isn't it afternoon? :P Jul 28 07:48:12 hmm.. debian has offlineimap 4.0.10 but upstream has 4.0.8 Jul 28 07:48:17 hi mickeyl Jul 28 07:48:27 mithro: Good time to get out of bed, isn't it? ;-) Jul 28 07:48:39 mickeyl: so I guess the slug distro's might be mentioned in the OOO newsletter now ;-) Jul 28 07:48:59 rwhitby: if you will write that part ;) Jul 28 07:49:07 rwhitby: *nod*. could I talk you into submitting a section about it? Jul 28 07:49:14 i plan to release the next one in mid-august Jul 28 07:49:36 mickeyl: I figured you might be writing about your experience as a distro user instead of developer :-) Jul 28 07:49:46 rwhitby: heh, good idea :) Jul 28 07:49:55 Ciao all Jul 28 07:50:10 mickeyl: You'd get to complain about someone else's distro for once :-) Jul 28 07:50:20 rwhitby: what a feeling! :) Jul 28 07:50:33 i will take off the OE-developer hat :) Jul 28 07:51:55 FYI: I've managed to write a lua hook (yay!) which talks to a fifo triggering CIABOT Jul 28 07:52:12 that means... in theory (when it doesn't crash) cia bot notifications should be more or less instantanoues now Jul 28 07:52:18 ibot: spell instantaneous Jul 28 07:52:19 'instantaneous' may be spelled correctly Jul 28 07:52:22 heh Jul 28 07:52:24 I do the same thing Jul 28 07:52:42 Never spell the word the same way twice, thats my rule Jul 28 07:52:44 mickeyl: awesome Jul 28 07:56:38 03koen 07org.oe.dev * r4bfb43e8... 10/packages/xauth/xauth/autofoo.patch: Apply patch from Philpp Zabel to make xauth compile again, closes #178 Jul 28 07:58:10 03rwhitby 07org.oe.nslu2-linux * r51d2dac7... 10/ (2 files in 2 dirs): Added setpwc to openslug-packages, and re-enabled the fixed atftp Jul 28 08:04:26 uh no... looks something is broken again :-/ Jul 28 08:07:27 mickeyl: http://gakusei.sf.net/nunome_1.0.bb still fails to compile although the ui target is the end of the recursion AFAICT, cf. http://rafb.net/paste/results/q6ud6c96.html. Can you take a look? Jul 28 08:07:47 florian_kc: in which sense? Jul 28 08:08:20 hrw|work: monotone: misuse: merge of 'packages/perl/perl-native_5.8.7.bb' : '17a9c3ad3eabd05aa0d0387505f0627fa7c70a4a' vs. 'aa90f26186f213f50cfbd3cf1a9bc946bd16340c' (no common ancestor) failed Jul 28 08:08:45 use explicit_merge Jul 28 08:09:32 Laibsch: yeah, i'll take a look Jul 28 08:09:45 mickeyl: Thanks. Jul 28 08:10:02 mickeyl: I'm off IRC now but will be back shortly. Jul 28 08:10:03 koen|ewi: how does this work? Jul 28 08:10:24 monotone explicit_merge Jul 28 08:10:29 florian_kc: monotone: misuse: branch 'org.openembedded.dev' is merged Jul 28 08:11:28 Laibsch: it may be necessary (and wise) to split the package up Jul 28 08:12:06 * mickeyl tinkers some more Jul 28 08:13:02 mickeyl: Along what lines? What are the parts? Trying to understand. I still have a steep learning curve. Jul 28 08:13:42 What is a good example to look at for a split package? Jul 28 08:14:13 perl :) Jul 28 08:14:20 hmm Jul 28 08:14:20 858 subpackages Jul 28 08:14:24 oh and the source tarball is ridicolous Jul 28 08:14:32 you need to remove all the binaries Jul 28 08:14:44 otherwise you can't really see if the .bb file works or not Jul 28 08:15:15 mickeyl: I need to remove the binaries from the source? Or where you referring to perl? Jul 28 08:15:27 yes, the binaries need to be removed or the Makefile gets confused Jul 28 08:15:29 or even does nothing Jul 28 08:15:40 with "split package" i mean more than one .bb file Jul 28 08:15:46 i.e. one bb file for the lib, one for the server etc. Jul 28 08:16:09 then again, it may just work when the source has been cleaned Jul 28 08:16:18 OK. I was thinking about changing the source directly sometimes. nunome appears to be pretty stale. No changes for the last two years AFAIK. Jul 28 08:16:53 i hate when people include binaries in sources Jul 28 08:16:56 makes our life hard Jul 28 08:17:49 OK, I'll remove the binaries as a first step. Jul 28 08:18:30 hrw|work, koen|ewi: No chance... Jul 28 08:18:42 uhrgs Jul 28 08:18:47 * mickeyl looks at the Makefile and pokes Jul 28 08:18:57 this is a _hand coded_ Makefile in a Qt/Embedded app Jul 28 08:19:01 *sigh Jul 28 08:19:04 throw that away as well Jul 28 08:19:10 write a .pro file instead Jul 28 08:19:47 .pro for professionals and .am for amateurs? Jul 28 08:19:51 hehe Jul 28 08:19:53 you said that Jul 28 08:20:22 it's pretty unwise to not use qmake when you're targetting Qt/Embedded Jul 28 08:20:35 and you don't interface with some special libs or other systems Jul 28 08:20:41 * koen|ewi targets qt/E Jul 28 08:20:44 * koen|ewi shoots Jul 28 08:20:52 hehe Jul 28 08:21:04 ~lart monotome Jul 28 08:21:04 * ibot makes a balloon animal out of monotome Jul 28 08:21:40 Laibsch: a) remove all .o and libs b) remove all the Makefiles and write .pro files c) be happy because it will be dead simple to build it then Jul 28 08:23:15 Hell... can't we use CVS? I can really do ithout all these cool features, but a RCS needs to work properly. Jul 28 08:23:34 03hrw 07org.oe.dev * r65ba11a9... 10/packages/offlineimap/ (offlineimap_4.0.3.bb offlineimap_4.0.9.bb): updated offlineimap to 4.0.9 from Debian 'stable' Jul 28 08:23:40 florian_kc: feel free to maintain your stuff in CVS and import/export from time to time Jul 28 08:24:08 you might even reuse one workdir Jul 28 08:24:29 There's lots of options out there - but somebody other then mickey / koen / zecke needs to take the bull by the horns and set up a test repo Jul 28 08:24:29 mickeyl: sounds like a plan... do we have any experience if this works properly? Jul 28 08:24:36 no we don't have Jul 28 08:24:47 CosmicPenguin: *nod* Jul 28 08:25:32 Not fair to just complain - for my part, I want to try Mercurial, just need to find some time to set it up Jul 28 08:25:39 ~lart AMD for not letting me play nice behind the proxy Jul 28 08:25:39 * ibot slams AMD against a large cement Tux for not letting me play nice behind the proxy Jul 28 08:26:24 hmm... as soon as i have some time i can do that, that might be a good way to handle this bad behaviour. Jul 28 08:26:51 mickeyl: I will do that as soon as I understand what you are talking about ;-) Jul 28 08:26:51 /me goes off reading about build systems Jul 28 08:26:51 mickeyl: There are no .o files, just .po. Do you know off-the-top of your head about a really easy intro on how to write .pro files? Jul 28 08:27:32 I wonder what is going on there.... i never touched this perl package it is using to merge. Jul 28 08:28:42 koen|ewi: How do i get the revisions to use for explicit_merge? Jul 28 08:29:32 Laibsch: http://doc.trolltech.com/3.3/qmake-manual.html Jul 28 08:29:47 Laibsch: cheet hint: cd $mydir; qmake -project . Jul 28 08:30:43 florian_kc: that stuff typically occurs when you don't update the working directory. you then create a fork without knowingly doing something Jul 28 08:30:58 florian_kc: use viewmtn to find an ancestor Jul 28 08:31:09 (and the other revids) Jul 28 08:32:24 great :-/ Jul 28 08:32:35 Thanks mickeyl. I'm really off now. Will rejoin you guys later. Jul 28 08:32:41 k, goodluck Jul 28 08:35:02 1,6G dl_dir Jul 28 08:38:04 bye Jul 28 08:38:15 cu Jul 28 08:38:20 bbl, housework Jul 28 08:38:33 florian_kc: you see the revisions in http://ewi546.ewi.utwente.nl/tmp/viewmtn/revision.psp?id=4bfb43e89e28f30568b8ae7af6bedc4854219177 Jul 28 08:38:56 florian_kc: divergence happens when someone commits in the time that it takes to do 'monotone update' Jul 28 08:41:11 koen|ewi, mickey|housewrk: One of the revisions i try to merge seems to be unknown to viewmtn. Jul 28 08:42:17 florian_kc: that would be your commit Jul 28 08:43:06 * florian_kc tries to sync Jul 28 08:45:06 hrm... no external 2-way merge command found Jul 28 08:45:14 apt-get install meld Jul 28 08:46:02 * koen|ewi -> pancakes Jul 28 08:46:27 03florian 07org.oe.dev * rdab6767a... 10/packages/gpe-dm/gpe-dm_0.47.bb: Start in gpe-dm in runlevel 2 too. Jul 28 08:47:02 very good Jul 28 08:53:22 hrm... i hope for some free minutes to set up a cvs for oe :-/ Jul 28 08:56:32 that'd be excellent Jul 28 08:56:38 or subversion, or something Jul 28 08:59:05 I only don't know where to do this... on hh.org we would need power to et up an experimental cvs, here the network link is too slow... Jul 28 09:00:41 maybe Greg could host something on treke.net Jul 28 09:02:35 pb_: hmm... that would rock. Jul 28 09:03:06 I should talk with the rest of the crew here... we should rent an external server. Jul 28 09:03:30 heh Jul 28 09:03:35 oesources.org doesn't have any room? Jul 28 09:07:37 florian_kc: yeah, that's another option. you can rent a virtual linux server (or even a real linux server) for virtually no money these days. Jul 28 09:09:38 maybe a company sympithetic to the cause would donate one? Jul 28 09:09:51 I would if I had more bandwith. Jul 28 09:11:11 pb_: i'll try to arrange this... maybe not this week.... but in not too far future. Jul 28 09:13:42 florian_kc: awesome Jul 28 09:18:08 "morning" Jul 28 09:19:22 isn't oesources treke.net? Jul 28 09:20:58 I just like the name Jul 28 09:21:02 it tickles my fancy Jul 28 09:23:50 is there any way to expand a disk image? I have an ext2 disk image I want to increase from 100MB to 200MB Jul 28 09:29:30 sure. use dd to stick some extra zeros on the end, then use resize2fs to expand the filesystem into that space. Jul 28 09:35:37 pb_, thanks :) Jul 28 09:37:35 pb_, I would use oflag=append option, right? Jul 28 09:39:47 mickeyl: thanks for the kernel patch, I'll try it soon Jul 28 09:40:00 mickeyl: did you get your mount -o bind thing working? Jul 28 09:40:57 bipolar: yah, something like that. or let dd write to stdout, and use a >> redirect in the shell. Jul 28 09:41:06 pb_, gotchya Jul 28 09:41:22 I also think that we should tell people to "pull, merge, update, push" instead of "sync, merge, sync". The sync method causes problems when multiple people try to merge the same 2 heads. Jul 28 09:44:48 monotone.vanille.de seems to be down again... Jul 28 09:57:23 zecke__: did you ever hear anything back from Paul Brook? Jul 28 10:11:33 hey ljp Jul 28 10:23:14 * mithro builds OE on his poor laptop Jul 28 10:27:18 * mickeyl puts on the pedant hat Jul 28 10:27:21 you can't build OE Jul 28 10:27:24 OE is not a distribution Jul 28 10:27:27 * mickeyl puts off the hat Jul 28 10:28:29 :) Jul 28 10:28:52 so were you able to sneak some hardware from the guys? Jul 28 10:29:17 * koen compiles bitbake Jul 28 10:30:01 * koen waits for mickeyl to find his hat Jul 28 10:30:18 * mickeyl puts it on again Jul 28 10:30:22 you can't compile Python apps Jul 28 10:30:29 unfortunately Jul 28 10:30:36 are there important things missing that would stop a bitbake release? Jul 28 10:30:37 * mickeyl puts it off again and throws it away Jul 28 10:30:59 is the overrides situation cleaned up? Jul 28 10:31:28 it seems to, but reenoo said something about preferred_providers Jul 28 10:31:46 and I'm not sure ?= works as expected Jul 28 10:32:24 ?= has been broken since long Jul 28 10:32:31 that wouldn't qualify as a showstopper to me Jul 28 10:32:43 let me check if my preferred providers test case works Jul 28 10:33:02 zecke and I want it out soon so we can persuade people to run tinderboxes Jul 28 10:34:09 uh oh Jul 28 10:34:17 i need to update my home workstation tree more often Jul 28 10:34:22 new revs : 433 Jul 28 10:34:25 that'll take a while... Jul 28 10:36:13 mickeyl: do you have some uni-sponsored box you could transform into a tinderbox? Jul 28 10:36:42 i don't think so - i'm already using more resources here than I'm allowed to :/ Jul 28 10:36:52 :) Jul 28 10:41:18 whats a tinderbox? Jul 28 10:42:26 http://ewi546.ewi.utwente.nl/OE_qa/OpenEmbeddedBuild/status.html and http://www.mozilla.org/tinderbox.html Jul 28 10:44:09 ibot: nslookup 82.75.118.233 Jul 28 10:44:11 it's a system to collect logfiles and match them to checkins so you know who to blame Jul 28 10:44:20 I can't find the address 82.75.118.233 in DNS Jul 28 10:44:37 233.118.75.82.in-addr.arpa domain name pointer cc444495-a.hnglo1.ov.home.nl Jul 28 10:44:47 hmm Jul 28 10:44:51 that's approx 15 km from my present location Jul 28 10:45:01 someone's still using org.openembedded.dev to read things from monotone.vanille.de Jul 28 10:45:12 no Jul 28 10:45:15 org.openembedded Jul 28 10:45:20 which - of course - fails Jul 28 10:46:35 hrm Jul 28 10:46:41 perl-5.8.7 fails too Jul 28 10:47:07 my appreciation for perl just dropped below zero Jul 28 10:49:57 * JustinP fires up monotone-viz to find the correct ancestor for merging... Jul 28 10:56:08 * JustinP shakes his head.....Why the hell can't users just answer the specific questions you're asking in an intelligent way? Jul 28 11:00:27 at least they answer the question Jul 28 11:14:31 03rpurdie 07org.oe.dev * racd8b0ec... 10/packages/linux/linux-openzaurus_2.6.13-rc3-mm1.bb: Prepare a load of patches for upstream. Apply temp fix for keyboard switch handling. Jul 28 11:16:58 insmod: init_module: usb_ohci_pxa27x: Operation not permitted Jul 28 11:17:01 any guesses ? Jul 28 11:17:07 yes, i am root Jul 28 11:17:28 when is this happening? Jul 28 11:17:44 i want to use usb host on my 3000 Jul 28 11:17:50 which doesn't quite work Jul 28 11:18:07 I *thought* it worked a little while ago...perhaps I was using Sharp ROM still then... Jul 28 11:18:14 on the 6000 it works flawlessly Jul 28 11:18:19 that's 2.4.18-hell though Jul 28 11:18:20 not 2.4.20-hell Jul 28 11:18:25 heh Jul 28 11:19:34 is it already loaded? Jul 28 11:19:48 no Jul 28 11:20:24 strangem I don't seem to have that module... Jul 28 11:22:01 perhaps you have other usb modules inserted which make it a client? Jul 28 11:22:07 no Jul 28 11:22:16 * JustinP really isn't helping.... Jul 28 11:22:19 hehe Jul 28 11:22:28 seems i've checked all potential failures Jul 28 11:22:44 Operation not permitted is not a pretty helpful message either Jul 28 11:22:59 "Which operation again?" Jul 28 11:23:02 I'll be trying the new kernel once my slow machine gets it compiled Jul 28 11:23:18 not that there's *really* much changed.... Jul 28 11:24:34 perhaps Sharp/Lineo also patch modutils Jul 28 11:24:38 patched Jul 28 11:25:28 hi Jul 28 11:25:33 hey Jul 28 11:26:28 whats going? Jul 28 11:26:56 zinga_ was apparently Jul 28 11:27:08 :) Jul 28 11:27:19 ~usbdmonitor Jul 28 11:27:32 was loaded? Jul 28 11:27:35 no Jul 28 11:27:41 usbd suxx Jul 28 11:27:44 it's just that we have a nice lart for that Jul 28 11:28:00 guess i need to add some printk into the module :/ Jul 28 11:28:04 *sigh* Jul 28 11:28:14 there we are... fscking with embedix kernels again Jul 28 11:28:29 isn't embedix great? Jul 28 11:28:36 ibot: pester bigal for 2.6 on spitz Jul 28 11:28:37 bigal for 2.6 on spitz: Are we there yet? .. Are we there yet? .. Are we there yet? see http://beverlys.net/LJ/BuggingYou.swf Jul 28 11:28:40 my idols Jul 28 11:28:49 it makes you appriciate everything else Jul 28 11:28:57 yes Jul 28 11:29:01 indeed Jul 28 11:29:12 the handhelds.org folks may have a hard time to get their kernels ported at all Jul 28 11:29:21 but they don't have to fsck with embedix crap Jul 28 11:29:37 just with completely undocumented chips Jul 28 11:29:38 heh Jul 28 11:30:01 ooh nice: Jul 28 11:30:09 udc_connect: usb-device is not active. Jul 28 11:30:09 Internal error: Oops - undefined instruction: 7 Jul 28 11:30:09 CPU: 0 Jul 28 11:30:09 pc : [] lr : [] Tainted: P Jul 28 11:30:09 sp : c03f1f64 ip : a0000013 fp : c03f1f80 Jul 28 11:30:10 r10: c03f0000 r9 : c03f1fbc r8 : c023f28c Jul 28 11:30:12 r7 : c3add800 r6 : c3add808 r5 : c379a660 r4 : c379a760 Jul 28 11:30:32 how'd you manage that? Jul 28 11:30:48 modprobing moduels Jul 28 11:30:52 nothing more Jul 28 11:31:13 eek Jul 28 11:31:14 ah Jul 28 11:31:15 3 heads Jul 28 11:31:16 ~lart RP Jul 28 11:31:17 * ibot dumps 42 tons of dirt, manure, and fish heads on RP Jul 28 11:31:34 pull, merge, push Jul 28 11:32:11 ~lart sync Jul 28 11:32:11 * ibot executes killall -KILL sync Jul 28 11:32:56 ~seen Bernado Jul 28 11:33:00 i haven't seen 'bernado', mickeyl Jul 28 11:33:07 judging from http://openzaurus.org/wordpress/development/c1000-support/ it looks like he managed to use usb host Jul 28 11:33:28 cu Jul 28 11:33:32 mickeyl: using coredump's image, usb host works on my akita Jul 28 11:34:12 * koen votes for autoinstalling ${PN}-dev packages into staging Jul 28 11:34:13 looks like a spitz-only failure then Jul 28 11:34:36 DaKa2: perl-5.8.7 fails on do_install for me Jul 28 11:35:04 koen: dammit.. Jul 28 11:35:07 i was just poking through coredump's stuff and don't see anything special he's doing. just a normal insmod Jul 28 11:35:13 koen: whats the error? Jul 28 11:36:04 something with man.... Jul 28 11:36:14 I'll look after this buildrun is completed Jul 28 11:37:02 he does unload a boatload of other modules before insmod usb_ohci_pxa27x Jul 28 11:37:15 (pxa27x_bi usb-storage net_fd storage_fd usbdcore usb_ohci_pxa27x) Jul 28 11:41:43 fsck Jul 28 11:41:53 mickeyl: newest kernel still isnt'w working for me Jul 28 11:43:01 * koen welcomes the green: http://ewi546.ewi.utwente.nl/OE_qa/OpenEmbeddedBuild/status.html Jul 28 11:43:32 JustinP: darn. can you check which differences in CMDLINE are in? Jul 28 11:44:01 mickeyl: CONFIG_CMDLINE is still showing in the "old" order in the .config. I didn't check the serial output but I'm willing to bet it said 2005, not 2006 Jul 28 11:44:16 perhaps I should do a new build from scratch.... Jul 28 11:45:00 mickey|dinner: same operation not permitted error if i insmod, but when i connect a device it works fine and lsmod shows the module Jul 28 11:45:54 * JustinP starts a fresh build Jul 28 11:51:49 mickey|dinner: any objections to the udev tty fix? Jul 28 11:51:56 mickey|dinner: ~/OE/build/tmp/familiar/staging/arm-linux/include Jul 28 11:52:00 aargh Jul 28 11:52:06 mickey|dinner: http://bugs.openembedded.org/attachment.cgi?id=109 Jul 28 13:08:35 Ciao all Jul 28 13:19:29 anyone tried gpe-image + akita? just started a fresh build, crossing my fingers :) Jul 28 13:22:01 I've successfully built a gpe image for spitz Jul 28 13:22:18 I don't think akita is all that different Jul 28 13:22:34 ~akita Jul 28 13:22:34 hmm... akita is the Sharp SL-C1000, or a dog Jul 28 13:23:13 gpe is upside down in the default build Jul 28 13:23:49 it's the wrong orientation you mean Jul 28 13:23:56 you need to change 90 into 270 (or vice versa) somewhere Jul 28 13:24:17 I altered mine to be "correct" but the portrait is upside down Jul 28 13:24:40 if I hit rotate it goes the right way Jul 28 13:25:48 koen|ewi: ok with me (udevtty) Jul 28 13:25:53 ok Jul 28 13:26:05 Is there any reason the bootstrap-image does not copy over the System.map to the /boot directory? Jul 28 13:26:08 will look at that after the movie Jul 28 13:26:13 How du I install busybox-httpd into the rootfilesystem? I get a busybox-httpd and a hostap-modules-cs, but I can't install it into the rootfs the way I install busybox and hostap-modules. bitbake returns a 'nothing provides busybox-httpd'. Jul 28 13:26:50 * JustinP is now running two new builds....lets see which one wins ;-) Jul 28 13:26:59 can we place bets? Jul 28 13:27:26 sure, as long as I'm not the one paying them out Jul 28 13:27:36 heh Jul 28 13:27:50 two different machines, different speeds, the faster was started when the slower got to glibc Jul 28 13:28:02 and I'm bitbaking task-bootstrap Jul 28 13:28:14 probably the slower one is going to win....we'll see Jul 28 13:37:18 mickeyl: patch applied Jul 28 13:39:19 03koen 07org.oe.dev * rb5a7cc16... 10/packages/udev/ (udev-063/tty-symlinks.patch udev_063.bb): Add tty patch, courtesy Philpp Zabel, closes #177 Jul 28 13:40:02 mickeyl: is cia running on lua yet? Jul 28 13:40:23 Anyone have any idea why the System.map file does not end up in my image? Jul 28 13:42:47 koen|tv: it's connected via a lua trigger, yes Jul 28 13:42:57 i had to implement a debounce algorithm though Jul 28 13:43:04 because the lua triggers once for every key received Jul 28 13:43:07 and we don't want that Jul 28 13:43:11 s/key/rev/ Jul 28 13:43:21 so it comes 1 minute later Jul 28 13:48:48 anybody here? Jul 28 13:49:29 ~ Jul 28 13:50:19 I upgraded the kernel on my collie but it oopses when I try to use my Linksys WCF12 Jul 28 14:01:52 martin__: i guess because there's no arm boot loader that is making use of it Jul 28 14:02:08 s/arm/embedded system/ Jul 28 14:03:26 mickeyl: but it seems that lots of things besides bootldrs might use it... Jul 28 14:06:00 i can't think of anything except ksymoops Jul 28 14:11:41 koen, anyone: could you please put up md5sums for the collie upgrade zImages? Jul 28 14:14:09 JustinP: que? Jul 28 14:14:23 iirc I md5 summed all familiar images Jul 28 14:14:29 I thought you were the one....oh, no it was hrw, sorry Jul 28 14:14:44 hrw: could you please put up md5sums for the collie upgrade zImages? Jul 28 14:14:56 OZ images Jul 28 14:15:01 :) Jul 28 14:15:19 yay, emacs is compiling on my zaurus Jul 28 14:15:40 I just had to disable all of the texinfo stuff in the Makefile Jul 28 14:29:15 it would rock to have a working tetex dist in oe Jul 28 14:31:14 It would rock to have a working OE Jul 28 14:31:17 * koen|tv hides Jul 28 14:31:41 ~lart berlios just because Jul 28 14:31:42 * ibot executes killall -KILL berlios just because Jul 28 14:31:46 It would rock to have a working RCS Jul 28 14:32:26 we have proven we can't handle the degree of freedom monotone enforces Jul 28 14:32:55 * koen|tv buys biggers cluebats Jul 28 14:33:07 kergoth: how did the interview go? Jul 28 14:33:37 mickeyl: I have a bb file for texinfo but it doesn't compile. Fails due to an empty "tools" directory Jul 28 14:34:26 Are both CONFIG_ARCH_PXA_SPITZ and CONFIG_ARCH_PXA_AKITA defined for the Akita? There's some *strage* "#if defined" stuff in here....(I'm looking at a diff of spitz to akita kernel code) Jul 28 14:35:28 does the CE-RH2 remote even work on the Akita? Jul 28 14:35:43 * JustinP wonders about these people.... Jul 28 14:36:23 welcome to the wonderful world of lineo Jul 28 14:36:27 yes, akita defines : Jul 28 14:36:35 CONFIG_ARCH_PXA_CORGI=y Jul 28 14:36:35 # CONFIG_CORGI_TR0 is not set Jul 28 14:36:35 CONFIG_ARCH_PXA_SHEPHERD=y Jul 28 14:36:35 CONFIG_ARCH_PXA_HUSKY=y Jul 28 14:36:35 CONFIG_ARCH_PXA_BOXER=y Jul 28 14:36:36 CONFIG_ARCH_PXA_SPITZ=y Jul 28 14:36:38 CONFIG_ARCH_PXA_AKITA=y Jul 28 14:36:43 ~lart lineo just because we didn't do it enough today Jul 28 14:36:43 * ibot beats lineo just because we didn't do it enough today senseless with a 50lb Unix manual Jul 28 14:37:12 ~lart lineo for being insane Jul 28 14:37:12 * ibot pours gasoline all over lineo, ignites the fire, and then enjoys some toasty marshmallows with the glorious blaze for being insane Jul 28 14:37:22 akita is the mega-dog Jul 28 14:37:25 heh Jul 28 14:38:58 this is just so mind-boggling Jul 28 14:39:11 * JustinP closes the diff before *he* goes insane Jul 28 14:40:02 hmmm....it seems ppp fails on the 64-bit AMD but not on my puny 800Mhz 32-bit.... Jul 28 14:40:22 hmm... /usr/lib/ipkg/info/module-init-tools.postinst seems to be missing an 's' ob bin/lsmod .... Jul 28 14:40:25 Thats because ppp isn't nearly l33t enough to be on 64 bit Jul 28 14:41:52 however if I manually run ranlib on two .a files it compiles fone Jul 28 14:42:24 re Jul 28 14:46:14 looks like hostap may need to be recompiled for the collie upgraded kernel.... Jul 28 14:47:57 yes, it is official, the Opteron has passed the Athlon up, despite being several hours behind starting its build Jul 28 14:48:14 and having a build problem with ppp ;-) Jul 28 14:49:06 oh, man you have to post a more intellegent bug report then that Jul 28 14:49:14 hmmm opteron, yummie Jul 28 14:49:38 * JustinP is lucky to have gotten an account on this server...it's so freaking fast Jul 28 14:49:57 and is now done Jul 28 14:50:52 CosmicPenguin: who? Jul 28 14:52:23 There aren't that many people with 64 bit - its unlikely that those bugs will be fixed by somebody else Jul 28 14:52:48 CosmicPenguin: well, that's all I know. Jul 28 14:52:58 I'm just putting up what I kno Jul 28 14:52:59 w Jul 28 14:53:09 What I'm saying is - start debugging.. :) Jul 28 14:53:24 I'll try to alter the bb file / Makefile to "fix" this Jul 28 14:53:38 yes, I will Jul 28 14:53:50 I just put up the bugs as I find them...solutions come later Jul 28 14:54:03 I didn't know at the time that it would work on my other server Jul 28 14:55:58 JustinP, you are aware the largest issue with a 64bit system right? Jul 28 14:56:58 any code that specifically depends on a 32bit short,long, double, etc will need to be changed Jul 28 14:57:12 nite all Jul 28 14:57:20 night Jul 28 14:57:52 emte: lovely. No, I don't really know anything about 64-bit systems. This is someone else's server. Things have worked fine, though. And as long as the native apps compile the cross-compiled stuff should be fine Jul 28 14:57:53 altho i guess a python patch now exists so that python will work on a 64bit os Jul 28 14:58:14 emte: must be since 1) this is Gentoo and 2) bitbake *is* running on it Jul 28 14:58:50 could be running in 32 then Jul 28 14:59:45 i cant remember but the opteron is a hybrid is it not? Jul 28 14:59:59 32/64 Jul 28 15:00:53 tre 64bit w/ fast 32bit emulation AFAIK Jul 28 15:00:53 no idea... Jul 28 15:01:02 true even Jul 28 15:01:05 yeah its not a true 64 Jul 28 15:01:22 What part is not true? Jul 28 15:01:26 ::shrug::....anyway, this bug happens on this machine but not my other Jul 28 15:01:32 so you will avoid quite a lot of the 64bit issues Jul 28 15:02:25 CosmicPenguin, in the way it stores integers etc Jul 28 15:02:39 it does math differently than a true 64 Jul 28 15:03:33 How does a true 64 do math, then? Seriously, I had no idea the Opterons wern't a real 64 bit machine Jul 28 15:05:02 i am trying to remember, had a professor go over this 3 months ago Jul 28 15:05:19 * france_ loves his alpha systems Jul 28 15:05:45 france_: ev7? ev68? or slower? Jul 28 15:05:49 has something to do with the fact in a 32/64 all the registers are still 32 unless implemented as 64 Jul 28 15:06:07 martin__: I have one of each model Jul 28 15:06:23 as a 32bit int will still be a 32 int in a 32/64 Jul 28 15:06:41 france_: the quad ev68? dual ev7? Jul 28 15:06:49 but a 32bit int will NOT be the same in a full 64bit system Jul 28 15:06:59 martin__: es45 Jul 28 15:07:01 france_: oh... just saw your IP... makes sense you would have those. :D Jul 28 15:07:10 martin__: hehe Jul 28 15:07:24 then there was a big explination of comp math Jul 28 15:07:29 emte: I don't understand - 32 bits is 32 bits - are you saying that opterons can't handle vaariables bigger then 32 bits? Jul 28 15:07:34 because thats just wrong Jul 28 15:07:38 no Jul 28 15:07:44 you missed what i typed Jul 28 15:08:00 emte has something to do with the fact in a 32/64 all the registers are still 32 unless implemented as 64 Jul 28 15:08:21 i guess it pairs registers or enables extra hw Jul 28 15:08:29 specifically i dont know Jul 28 15:08:34 ANd then you said a 32bit int will be different between a 32 bit system and a 64 bit system Jul 28 15:08:40 which makes no sense - 32 bit is 32 bit Jul 28 15:08:52 ahh Jul 28 15:08:56 think placement Jul 28 15:09:03 and direction Jul 28 15:09:06 no, a 32-bit int on a true 64 system would be 64-bits in the register Jul 28 15:09:15 (right....?) Jul 28 15:09:23 or am *I* off-base here? Jul 28 15:09:28 right - but thats the same as a 16 bit int on a 32 bit system Jul 28 15:09:54 yes Jul 28 15:09:55 except taht a 32 bit OS knows how to handle 32-1bit Jul 28 15:10:03 I think what emte is trying to imply is that the Opteron does trickery to simulate a 64 bit value, because it has to combine two 32 bit registers Jul 28 15:10:36 okay ... take a 32 bit short Jul 28 15:10:39 no, he's saying it runs in 32-bit mode unless specifically asked to do 64 Jul 28 15:10:48 of any value Jul 28 15:10:58 "32-bit mode" is probably not the right wording, but... Jul 28 15:11:15 if you take that value and try to access it as a 64bit short you will get garbage Jul 28 15:11:52 untrue - it will be zero extended out Jul 28 15:12:00 just like if you had a 8 bit value and tried to access it as a 32 bit value Jul 28 15:12:18 that would be true if 32 and 64 both used the same algoritthm for storage Jul 28 15:12:27 they dont Jul 28 15:12:52 The endian doesn't change - bytes are still stored in the same order Jul 28 15:13:48 * emte isnt explaining this right ... Jul 28 15:14:29 okay, you know how a number will have a mantissa right? Jul 28 15:14:34 CosmicPenguin: you're thinking of 8 and 32-bit storage on the same system. Those can be read any way, but the actual *storage* is different on a 64 bit system. So a 32-bit value won't be the same as it would be on a 32-bit system. Jul 28 15:14:51 unless you got one of those alien variable endianess processors :) Jul 28 15:15:08 lol Jul 28 15:15:25 emte: am I still saying all this right? Jul 28 15:15:43 JustinP, yeah i am following you but i dont think it helps CosmicPenguin Jul 28 15:16:26 I'm trying to figure out what you all mean by "storage" - you mean in the registers? Jul 28 15:16:48 03mickeyl 07org.oe.dev * r78fbed1d... 10/packages/e2fsprogs/ (4 files): remove nonworking e2fsprogs 1.36, add e2fsprogs 1.38 Jul 28 15:16:48 in the menory, yes Jul 28 15:16:53 no Jul 28 15:16:54 memory* Jul 28 15:16:57 memory Jul 28 15:16:58 ok - memory Jul 28 15:17:25 in other words the bits are stored differently Jul 28 15:17:41 in another format Jul 28 15:18:00 mantissa is a different length and in a different location Jul 28 15:18:20 there you go Jul 28 15:18:32 but that doesnt help if you dont know what the mantissa is ... Jul 28 15:18:45 or where/how long it is Jul 28 15:20:25 http://www.ee.unb.ca/tervo/ee6373/IEEE64.htm Jul 28 15:20:43 Oh - but you're talking floating point Jul 28 15:20:44 if you look at the very bottom there is an example for the value if +10 Jul 28 15:21:06 it is an issue with all number storage Jul 28 15:21:15 short, long, double, etc Jul 28 15:22:19 So you are saying that all values in a Opteron are stored like this in memory? Jul 28 15:22:45 * JustinP quietyly submits a patch for ppp Jul 28 15:23:49 CosmicPenguin, on an opteron, no because its 32/64 Jul 28 15:25:19 the opteron spec overview states you "enable" 64bit on a process basis Jul 28 15:25:30 whatever that means Jul 28 15:25:47 So what you're saying is that a true 64 bit machine stores memory in the way you mentioned Jul 28 15:25:52 then it talks about extra threads and piplelines etc Jul 28 15:25:57 tes Jul 28 15:26:01 yes* Jul 28 15:26:07 but that since the Opteron isn't a true 64 bit machine, it stores memory the old fashioned x86 way? Jul 28 15:26:15 i guess so Jul 28 15:26:24 which would make it a 32/64 Jul 28 15:26:32 and x86 compatible Jul 28 15:26:44 Ok - one final question - what advantages does your mantissa method hold so that the true 64 bit processors would prefer to use it? Jul 28 15:26:52 BTW, my emacs compile on my Zaurus is still going ;-) Jul 28 15:26:59 faster math Jul 28 15:27:08 JustinP: compiling *on* the Z? Jul 28 15:27:18 64bit systems are traditionally crunching machines Jul 28 15:27:35 accounting and db Jul 28 15:27:45 Faster floating point math - obviously the integer math would be just as fast either way Jul 28 15:27:56 jh___: yes. I couldn't get it to compile in OE so I figured I'd try on the Z. It needs to compile itself, then use itself to compile parts of itself or something...will likely be a large pain to get it working in OE Jul 28 15:28:18 ~lart emacs' buildsystem Jul 28 15:28:18 * ibot gives emacs' buildsystem an extra strength ACME sleeping pill, sending emacs' buildsystem to sleep for 150 years, and awakening to seven strange dwarfs and a large apple Jul 28 15:28:24 well trying to find the point at which PI shows a pattern is more float than int :P Jul 28 15:28:47 JustinP: nice. yeah, that might take a while :) Jul 28 15:29:00 but with some magic you can split float math into int math Jul 28 15:29:26 * koen|tv calls harry potter Jul 28 15:29:30 lol Jul 28 15:29:53 * emte watches Harry Pothead stumble over Jul 28 15:30:30 wonder if that cartoon is still on the net ... Jul 28 15:30:42 JustinP: post the loading time when it's done... should be interesting. i use microemacs on the z now, would be nice to have the full version Jul 28 15:31:53 I have mec but things like "Ctrl-Space" don't work for me so it's kind of useless... Jul 28 15:32:10 It's compiling .el files now so I think it will actually finish. :-) Jul 28 15:33:21 * florian doesn't get rid of "41: line 445: syntax error near unexpected token `'INHIBIT_PACKAGE_STRIP','" Jul 28 15:33:26 cool. yeah, the ctrl-space thing is annoying. plus its macro lanuage looked crazy Jul 28 15:33:49 florian: do you inherit ipk somewhere? Jul 28 15:35:01 florian: that error usually means 'I want to strip a package, but I don't know what format that package is suppossed to be' Jul 28 15:35:44 hmm... /me investigating Jul 28 15:36:03 jh___: what system do you have? Jul 28 15:36:17 collie & akita Jul 28 15:37:04 ah. well I can try to package this up if it works....maybe.... Jul 28 15:37:34 * koen realizes that akita wouldn't really sell in .nl : http://c1000.nl/ Jul 28 15:37:41 that'd be great if it's not too much trouble Jul 28 15:37:53 being named after a cheap supermarket Jul 28 15:38:09 lol Jul 28 15:38:22 or a dog Jul 28 15:38:24 :P Jul 28 15:38:36 koen: that isnt what the error means. it means "oe is being a dumbass and appending python code to a nonexistant function which defaults to 'shell'" Jul 28 15:38:51 koen: when you inherit package, the do_package variable gets the python flag set. Jul 28 15:38:59 which fixes that unholy union Jul 28 15:39:14 iirc. Jul 28 15:39:16 heh Jul 28 15:40:01 The package causing this failure ia a kernel bb, it inherits kernel. Jul 28 15:40:09 how was the interview? Jul 28 15:40:16 florian: doesnt matter. Jul 28 15:40:24 florian: it will fail on all packages until you inherit a package bbclass. Jul 28 15:40:24 kergoth: aha Jul 28 15:40:30 florian: set DISTRO or inherit package_ipk manually Jul 28 15:40:39 * koen stores the 'dumbass' quote Jul 28 15:40:44 hehe Jul 28 15:40:56 how was the interview? Jul 28 15:41:17 oh, pretty good. i'm blind, didnt see it the first time Jul 28 15:41:18 heh Jul 28 15:41:24 lol Jul 28 15:41:25 np Jul 28 15:41:53 he tell you your overqualified? Jul 28 15:41:57 the cto guy, when he was going to bring me over and introduce me to the lead developer, goes "well.. why dont you talk to her and get a feel for whether hte job is a good fit for you.. and .. well *glances at resume*, we wouldnt have brought you in here if we didnt think you were a good fit" Jul 28 15:42:07 hehe, no, but it was clear that i was Jul 28 15:42:07 kergoth: DISTRO is set to "familiar-0.8.2" Jul 28 15:42:10 they're getting a deal Jul 28 15:42:25 lol Jul 28 15:42:39 florian: then find out why things arent happy. check sanity. for some reason familair-0.8.2.conf isnt being included, or some other .conf file along that parse chain Jul 28 15:42:46 sounds like you have more qualifications than thier current lead Jul 28 15:43:06 emte: indeed. it doesnt seem too bad, mostly application level stuff on their network appliances, iptables, qos, etc Jul 28 15:43:10 pretty easy Jul 28 15:43:13 at least i'd have some cash flowing Jul 28 15:43:19 yeah thats easy stuff Jul 28 15:43:20 * kergoth crosses fingers Jul 28 15:43:40 * JustinP crosses kergoth's fingers Jul 28 15:43:51 then you will have to exercise that OE muscle to stay in shape :P Jul 28 15:44:16 hehe Jul 28 15:44:30 as long as it doesnt get too boring.. it seems like theres some variety, just not much challenge Jul 28 15:44:34 we all know your itching to finish bb2 :P Jul 28 15:44:43 that works for me. if i dont get _too_ bored, i can at least find some motivation to do open source work again :) Jul 28 15:45:00 yeah Jul 28 15:45:20 you could also develope a device of your own ... on the side Jul 28 15:45:54 kergoth: thank you... that's it. Somehow familiar-0.8.2.conf was lost. Jul 28 15:45:58 np Jul 28 15:46:34 koen: eh... why is the conf for our stable familiar distribution missing?! Jul 28 15:46:38 everyone wants an appliance for something Jul 28 15:47:19 florian: 0.8.2 doesn't build with the current OE Jul 28 15:47:31 koen: ! Jul 28 15:47:40 koen: how could that happen?! Jul 28 15:47:55 lol Jul 28 15:47:57 updating patches for cvs versions, removing old packages, etc Jul 28 15:48:31 * JustinP whispers "tinderbox" Jul 28 15:48:43 lol Jul 28 15:49:21 if you *really* want to build 0.8.2 you can use the bk cvs snapsnot and checkout the 0.8.2 tag: http://familiar.handhelds.org/source/ Jul 28 15:49:59 anyone know if we have an IR scope in packages? Jul 28 15:50:15 i cant remember the name of the linux one Jul 28 15:54:21 good thing I have screen on my Z...I may need to leave it compiling while I'm driving home ^_^ Jul 28 15:55:02 * koen tries to image life without screen Jul 28 15:55:20 * koen sees zillions of terminals cluttering the desktop Jul 28 15:56:02 I hated not having screen on my Zaurus Jul 28 16:02:17 DaKa2: before I go to sleep: Jul 28 16:02:18 | make[1]: Nothing to be done for `install.man'.| make[1]: Leaving directory `/home/koen/OE/build/tmp/familiar/work/arm-linux/perl-5.8.7-r3/perl-5.8.7'| mv: cannot stat `/home/koen/OE/build/tmp/familiar/work/arm-linux/perl-5.8.7-r3/image//usr/lib/perl5/5.8.7/arm-linux/CORE/libperl.so': No such file or directory Jul 28 16:02:32 Is monotone.vanille.de down ? Jul 28 16:02:33 'night all Jul 28 16:02:45 I get a connection refused Jul 28 16:02:53 gerwinin: could be, see http://oe.handhelds.org for mirrors Jul 28 16:03:11 koen|sleep: ok.. Ill look at that......... Jul 28 16:05:00 It seems there only one mirror Jul 28 16:10:28 http://oe.handhelds.org/oe_wiki/ is Forbidden Jul 28 16:11:07 yep, it is Jul 28 16:12:39 justinp I am getting a connection refused on monotone.vanille.de Jul 28 16:13:00 Is it down or are am I doing something wrong ? Jul 28 16:13:51 kergoth: it's listed on the Wiki main page. It says you can get to the old wiki....but obviously you can't. Jul 28 16:14:25 gerwinin: it's probably down Jul 28 16:15:55 :( than I can only build from the db file and I don't know how to :( Jul 28 16:15:55 JustinP: then the wiki main page needs fixing Jul 28 16:16:54 kergoth: I assumed that someone could get the old wiki back up....oh well Jul 28 16:17:10 gerwinin: I don't get what you're saying. You always have to us eyour DB file to do checkouts and such... Jul 28 16:17:44 JustinP: afraid not. it being down isnt an accident. the hh.o admin folks took it down because it was a spam sink. Jul 28 16:17:57 iirc. Jul 28 16:18:26 JustinP: I ment I don't know how to use the snapshot Jul 28 16:19:02 gerwinin: the snapshot is a normal monotone DB file. Just skip the pull step and start using it. First create a private key. Jul 28 16:23:50 * emte wonders if monotone really is harder to use than BK Jul 28 16:24:32 not harder - just less understanding of stupid users Jul 28 16:24:50 hmm Jul 28 16:25:05 wonder if i should setup an Hg mirror Jul 28 16:25:17 its pretty userfriendly Jul 28 16:26:54 * emte contemplates it Jul 28 16:27:07 It needs more cpu power... Jul 28 16:28:25 monotone? Jul 28 16:28:42 i have heard that Jul 28 16:33:38 emte: yes really... do you have a fast machine? Jul 28 16:34:04 apperently by some people's perspectives Jul 28 16:34:14 by my own ... not really Jul 28 16:34:38 Hg runs quite a lot lighter than monotone Jul 28 16:35:09 so it would be an Hg mirror of the repo... maybe a few times a day update Jul 28 16:35:18 Hg? Jul 28 16:35:33 another scm Jul 28 16:35:49 emte: Do you have a link for me? Jul 28 16:35:58 imports/exports monotone,git,bk,etc Jul 28 16:36:31 http://www.selenic.com/mercurial/ Jul 28 16:37:41 works well for the most part ... needs a vuser system tho Jul 28 16:39:02 emte: looks pretty interesting Jul 28 16:39:40 what generates the files in /usr/lib/ipkg/info ? Jul 28 16:39:43 works good so far Jul 28 16:40:07 with exception to my vuser annoyance Jul 28 16:40:30 i dont like having to make physical accounts just for others to have access to it Jul 28 16:41:15 that does sound like a bit of a deal breaker Jul 28 16:41:46 its one of those issues that are sometimes overlooked Jul 28 16:42:07 i am hoping it gets implemented soon, or i'll get annoyed enough to fix it Jul 28 16:42:47 but its relitivly cross-platform Jul 28 16:42:59 which is something i need Jul 28 16:47:20 martin__: ipkg... a part of these files are extracted files from the packages control section. Jul 28 16:50:23 hmm Jul 28 16:50:47 anyone have a good memory of which SMD/T cap is which? Jul 28 16:51:04 good night Jul 28 16:51:38 night Jul 28 16:54:35 kergoth, i have the perfect deveice for you to develop Jul 28 16:54:39 device* Jul 28 16:55:49 a master electronics parts inventory database device- it would be remote accessible, standalone, and store all the datasheets for the parts Jul 28 16:55:59 and usable Jul 28 16:56:09 man i'd kill for one of those Jul 28 16:56:22 make it a small cube with blinky lights Jul 28 16:59:58 hmm Jul 28 17:00:04 is this a resistor or a cap ... Jul 28 17:04:37 what does it look like? Jul 28 17:05:13 black, two silver caps and ~ 0.3mm x 0.1xx Jul 28 17:05:15 mm* Jul 28 17:05:32 i'll have to try it on my meter to tell Jul 28 17:05:43 doesnt seem to have any markings Jul 28 17:05:56 smd? Jul 28 17:06:07 i have some EMI suppressors if anyone is interested Jul 28 17:06:08 yeah Jul 28 17:08:30 k ... who is k ... Jul 28 17:09:18 smd caps tend to be marked. You can usually tell by the metal on the ends. Jul 28 17:10:16 if it's plastic or ceramic I would bet it's a cap Jul 28 17:11:05 pretty sure its a cap Jul 28 17:11:22 its ceramic and a "block" as opposed to flat Jul 28 17:11:39 doubt its the 1uF i need tho Jul 28 17:14:28 where did i put them ... i know i have some 1uF areound ... Jul 28 17:14:29 it's probably more like 1pF Jul 28 17:17:42 nah Jul 28 17:17:50 was written on the case they were in Jul 28 17:17:52 :P Jul 28 17:17:57 thats why i know Jul 28 17:18:05 but where i put the case .... Jul 28 17:20:05 i KNOW if i order more i'll find em :( Jul 28 17:21:03 fk it Jul 28 17:21:26 you can always use more parts ... isnt that how the addage goes? Jul 28 17:22:40 well, nothing is worse than missing a few components when working. Better with plenty than none. Jul 28 17:23:44 That's what I thought when I spent $250 on resistors.. :P Jul 28 17:25:07 lol Jul 28 17:25:31 thats a lot of resistors ... concidering you can usially get a thousand for ~$5 Jul 28 17:25:37 usually* Jul 28 17:25:42 unless its an odd resistor Jul 28 17:26:24 hmm 16 or 25 ... Jul 28 17:27:20 yeah its like 10 of preatty much each of every one from 1ohm to 1M. Buying plenty of one kind is obviously much cheaper. Jul 28 17:28:41 yeah Jul 28 17:28:47 1ohm :P Jul 28 17:29:41 3.5 cents each ... how many do i want ... Jul 28 17:30:27 should order more crystals while i am at it i guess Jul 28 17:30:48 what are you making? Jul 28 17:31:24 working with PICs doind some embedded automation stuff Jul 28 17:31:35 hence the crystals and 1uF caps Jul 28 17:31:54 doing* Jul 28 17:32:12 1uF makes a handy decoupling cap Jul 28 17:32:23 for most applications Jul 28 17:32:41 or a quick and dirty filter Jul 28 17:34:20 doesn't any of the PIC's have internal oscillators? Some of the AVR's have. Jul 28 17:34:45 5MHz is a bit low for some applications Jul 28 17:35:16 especially after you divide that by 4 Jul 28 17:35:36 since PICs are a 4cycle device Jul 28 17:36:18 fetch-read-exicute-write Jul 28 17:47:17 I need help with the touchscreen library. Is there anyone here who is familiar with qt and /dev/ts? Jul 28 17:48:37 AVR's are single cycle right? Jul 28 17:49:15 would have to be a min of three no? Jul 28 17:49:22 hi, how can I make opengl work on my collie? Jul 28 17:49:38 fetch-read/exec-write Jul 28 17:49:39 actually it says that X11 driver are not configured with OpenGL Jul 28 17:50:10 EnceloAMD, ask in the right channel? Jul 28 17:50:17 emte: pipelining Jul 28 17:50:18 this is not a distro channel Jul 28 17:50:39 gb2, i ment per instruction Jul 28 17:50:50 emte: mmm, ok Jul 28 17:51:01 i think you want #oz Jul 28 17:51:05 emte: lots of architectures are single instruction per clock Jul 28 17:51:08 or is it #openzaurus Jul 28 17:51:47 I think most, but not all of the instructions on avr are single cycle. Jul 28 17:51:48 gb2, yes but how many clock cycles does it take to process a full instruction? Jul 28 17:52:03 emte: #openzaurus Jul 28 17:52:16 emte: a guy suggest that I ask here too :D Jul 28 17:52:19 emte: sure, you may have latency, but you still have 1:1 for throughput Jul 28 17:52:26 EnceloAMD, aha, i was guessing since i dont own a zaurus device Jul 28 17:52:49 EnceloAMD, i take it that guy hasnt met kergoth then :P Jul 28 17:53:03 yes, I have...::sigh::... Jul 28 17:53:09 lol Jul 28 17:53:13 bad you JustinP Jul 28 17:53:30 opengl would be distro specific Jul 28 17:53:32 I figured this is more of a "how to I enable GPE for the X server". This *is* a compiling question after all.... Jul 28 17:53:34 oh well Jul 28 17:53:57 for a device and distro Jul 28 17:54:16 not a "generic" question Jul 28 17:55:17 I suppose. However #openzaurus is more of a user-help forum while this is more of a compiling help forum.... Jul 28 17:55:22 Generic Answer; refer to your distribution for further instructions Jul 28 17:55:31 anyway, I'll stop debating it Jul 28 17:55:35 :) Jul 28 17:55:44 GPE for X is just enabling QVFB. Jul 28 17:56:19 we are talking about OpenGL... Jul 28 17:57:02 You are trying to compile OpenGL? Jul 28 17:57:41 I can compile gl example with mesa Jul 28 17:57:45 but I can't make them work Jul 28 17:57:56 because "X11 driver not configured with OpenGL" Jul 28 17:58:26 Makes sense. X needs to be recompiled to use OpenGL. It is a compile time option. Jul 28 17:59:54 doh :) Jul 28 18:00:12 hence the reason it is a distro issue Jul 28 20:09:25 anyone here ever use AC? Jul 28 20:09:36 for pcb Jul 28 20:10:03 http://www.4pcb.com Jul 28 20:19:28 * france is back (gone 13:27:58) Jul 28 20:19:49 what happend to france? the germans get it again? :P Jul 28 20:19:59 sorry .. bad humor Jul 28 23:03:11 * france is away: Away **** ENDING LOGGING AT Thu Jul 28 23:59:56 2005