| 00:01.43 | treke- | alright. ckermit is gone |
| 00:01.45 | CosmicPenguin | What if I grab a video of kermit using kermit? |
| 00:02.09 | CosmicPenguin | I find it interesting that the kermit project actually got permission to use the name |
| 00:02.39 | Pendalar | my wife is misusing my son's Casey the Kinderbot as prerecorded annoyance to telemarketers.. |
| 00:03.11 | Pendalar | I suppose it works but .. freaks me out |
| 00:07.15 | Pendalar | hrm hh.org doesn't have a rss |
| 00:10.24 | *** part/#oe IRC_noobee (~IRC_noobe@p548310F3.dip0.t-ipconnect.de) |
| 00:10.25 | treke- | http://www.handhelds.org/geeklog/index.rdf |
| 00:12.45 | CIA-5 | 03ggilbert 07 * r1.2963 10openembedded/BitKeeper/deleted/ (2 files): |
| 00:12.45 | CIA-5 | .del-qte-for-qtopia_2.3.9.bb~304327447618863f: |
| 00:12.45 | CIA-5 | <PROTECTED> |
| 00:12.45 | CIA-5 | .del-qte-for-opie-2.3.9.bb~9371dfe1bb97a47c: |
| 00:12.45 | CIA-5 | <PROTECTED> |
| 00:18.00 | Pendalar | ooohhh |
| 00:20.18 | *** join/#oe kiatoa (~matt@rich-dsl3-66.greenmountainaccess.net) |
| 00:29.19 | *** join/#oe Neo|Home (~neotron@c-24-17-200-89.client.comcast.net) |
| 00:59.35 | *** join/#oe acidware (~acidware@cable-24-158-210-137.sli.la.charter.com) |
| 01:00.15 | acidware | do they have a howto for restorin your ipaq from the windows mobile 2003 cd, if you dont have a backup.gz image for it |
| 01:00.42 | CosmicPenguin | I'm sure that other people in other rooms that actually deal with that sort of stuff would know |
| 01:01.33 | acidware | i have my old ipaq with linux on it but i want to sell it |
| 01:05.26 | Pendalar | I saw the hh post about getting a semi boot for ipaq hx4700 |
| 01:05.39 | Pendalar | pretty neat, was looking at the hx4705 today |
| 01:24.12 | kergoth | re |
| 01:24.40 | kergoth | ah, ckermit is finally killed? good good |
| 01:24.55 | Pendalar | everyone hates poor kermy |
| 01:25.51 | keturn | I dunno, linux-mips seems to like kermit lots better than minicom. |
| 01:26.10 | kergoth | cant blame em |
| 01:26.13 | kergoth | minicom is crap |
| 01:26.47 | kergoth | days like today remind me why i like texas |
| 01:26.53 | kergoth | was like 85 degrees out (F) today |
| 01:26.57 | kergoth | in january |
| 01:27.50 | Pendalar | 47 highest I believe here today |
| 01:27.57 | Pendalar | didn't watch it that close however |
| 01:28.06 | Pendalar | 85 is good but texas summer.. sucks |
| 01:28.20 | kergoth | no doubt |
| 01:28.29 | kergoth | mad dashes from one building with AC to another |
| 01:29.37 | Pendalar | carrying around backpacks filled with ice just incase you need to dump it on your head.. |
| 01:29.43 | kergoth | heh |
| 01:31.08 | Pendalar | I should try out kismet with this ambicom |
| 01:31.22 | Pendalar | guess I could pull the old kismet and kismet qt/e and pray |
| 01:43.48 | kergoth | jeeze, mambo at openzaurus.org -really- relies on the openzaurus.org domain specifically |
| 01:43.52 | kergoth | it makes a lot of flawed assumptions |
| 01:44.00 | kergoth | openzaurus.sf.net is hardly even usable |
| 01:56.28 | *** join/#oe mrz80 (~scotsman@ip24-136-33-61.ga.at.cox.net) |
| 02:02.25 | *** join/#oe mithro (~tim@dsl1-83.gw1.adl1.airnet.com.au) |
| 02:06.29 | *** join/#oe duffolonious (~bryan@63-226-146-73.mpls.qwest.net) |
| 02:22.55 | kiatoa | I'm trying to make a .bb file for widestudio libs. ws instructions: cd src;./configure;make devfb |
| 02:23.14 | kiatoa | I put cd src in do_configure_prepend which seems to work. |
| 02:23.45 | kiatoa | but |
| 02:24.16 | kiatoa | cd src followed by make devfb doesn't work. |
| 02:24.21 | *** join/#oe micropal_ (ident@139.80-202-82.nextgentel.com) |
| 02:25.08 | kiatoa | It trys to install into /usr/local/. Suggestions? |
| 02:25.55 | *** join/#oe raster (~raster@vagw.valinux.co.jp) |
| 02:27.30 | *** join/#oe raster (~raster@vagw.valinux.co.jp) |
| 02:27.32 | kiatoa | Seems like DESTDIR isn't supported and also the binaries generated are x86. |
| 02:29.39 | kiatoa | I'll take the deafening silence as a sign that I'm on my own :-) |
| 02:29.49 | kergoth | how you exert control over a build is completely dependent on the buildsystem in question. |
| 02:30.10 | *** join/#oe noclouds (~mhfan@61.191.206.159) |
| 02:30.11 | kergoth | ./configure implies autoconf/automake, so make sure you're inheriting autotools |
| 02:30.43 | kiatoa | Hi kergoth. Ok. Fair enough. Yup. "inherit autotools" is in there. |
| 02:31.29 | kergoth | its a case by case thing. |
| 02:31.36 | kergoth | read its makefiles |
| 02:31.41 | kergoth | get a feel for what variables it obeys |
| 02:31.45 | kiatoa | How does bitbake run configure to set CC etc? |
| 02:31.49 | kergoth | make sure its using the CC, CFLAGS, etc from autoconf |
| 02:31.52 | kergoth | they're set in the environment |
| 02:31.57 | kergoth | autoconf ./configure picks them up |
| 02:31.59 | kergoth | and uses them |
| 02:32.05 | kergoth | when it does the Makefile.in -> Makefile replacements |
| 02:32.09 | kergoth | @CC@ .. etc |
| 02:32.10 | kiatoa | Ok. So ws build system must be broken. |
| 02:32.23 | kergoth | most buildsystems are, one way or another |
| 02:32.41 | kergoth | read the makefiles and see if you can figure out why it isnt using the configure variables |
| 02:33.01 | mithro | hey people |
| 02:33.05 | mithro | whats up? |
| 02:33.07 | kergoth | hey |
| 02:33.30 | kiatoa | Suggestions on how to run the configure/make process? Is there a way to run bitbake to start up an xterm or shell with the environment exactly as it would be during a build? |
| 02:33.35 | kergoth | ? |
| 02:33.45 | kergoth | let oe run the configure phase. |
| 02:33.53 | kergoth | bitbake -c clean blah |
| 02:33.58 | kergoth | bitbake -c configure blah |
| 02:34.07 | kergoth | now cd into the sources and check out the makefiles post-./configure |
| 02:35.39 | mithro | treke-: how do you do the source mirror? |
| 02:36.09 | treke|home | right now? It's completely broken, but it just called out to oebuild -cfetch :p |
| 02:36.23 | treke|home | in the future things will be getting a bit more complicated |
| 02:36.29 | treke|home | since I'll be doing source aging |
| 02:38.52 | mithro | how do you only get the gpl etc source? |
| 02:38.59 | treke|home | I don't |
| 02:39.07 | treke|home | there is code in bitbake to handle that |
| 02:39.14 | mithro | ahh okay |
| 02:39.16 | treke|home | but I wasn't using it yet |
| 02:39.38 | treke|home | it wasn't useful unless everything in oe had a license declared |
| 02:39.55 | treke|home | now we're down to 11 packages without licenses, so I can set it up |
| 02:41.40 | mithro | i need more ram in this laptop :/ |
| 02:41.49 | mithro | 256mb just isn't enough |
| 02:42.39 | *** join/#oe emte (~emte@d66-183-163-211.bchsia.telus.net) |
| 02:43.07 | treke_ | yeah |
| 02:49.23 | keturn | except lots of stuff, from binutils on down, uses FILESDIR, so it must not be totally worthless. |
| 02:50.10 | kergoth | FILESDIR is the old way |
| 02:50.12 | kergoth | deprecated |
| 02:50.15 | kergoth | FILESPATH is the new way |
| 02:52.53 | kiatoa | kergoth: FYI: What I just did is to insert "xterm" after the "cd src". Gives me the exact environment that bitbake creates. Seems great for debugging. |
| 02:53.51 | keturn | Okay. Can you confirm what I'm seeing in fetch.Local.localpath? path = url.split()[1]; newpath = path; if not newpath: filesdir = stuff # is never reachable |
| 02:59.46 | kergoth | hm? |
| 03:00.01 | kergoth | it uses FILESDIR if it wasnt able to find the file its looking for in FILESPATH. |
| 03:00.11 | kergoth | <PROTECTED> |
| 03:00.11 | kergoth | <PROTECTED> |
| 03:00.34 | keturn | but newpath was set to path previously, so newpath is never false at that line |
| 03:00.41 | kergoth | what? |
| 03:00.44 | kergoth | yes, it is |
| 03:00.48 | kergoth | <PROTECTED> |
| 03:00.48 | kergoth | <PROTECTED> |
| 03:00.54 | kergoth | if FILESPATH is set, newpath is overridden |
| 03:01.01 | kergoth | and if which returns None, it didnt find it. |
| 03:01.17 | kergoth | so if FILESPATH isnt set, it uses FILESDIR, or if it couldnt find it in FILESPATH, it uses FILESDIR, otherwise it uses path |
| 03:01.22 | kergoth | get it? |
| 03:01.58 | keturn | "which" returns empty, right, okay |
| 03:02.41 | kergoth | its rather convoluted, but that way FILESDIR still works |
| 03:02.46 | kergoth | as expected |
| 03:02.53 | kergoth | i should probably add a warning there |
| 03:04.43 | kergoth | oe.note("FILESDIR is deprecated, please switch to FILESPATH.") |
| 03:04.43 | kergoth | anyone here setup cups printing via samba? |
| 03:04.43 | kergoth | i cant get cupsaddsmb to authenticate :\ |
| 03:05.53 | keturn | if FILESPATH isn't set, what does getVar return? |
| 03:06.27 | kergoth | None |
| 03:06.47 | kergoth | which is false when used in an if statement |
| 03:08.41 | keturn | in that case, newpath is still set to path, since 'which' is never called to set it to empty. |
| 03:10.03 | kergoth | yep, looks like a flaw in the logic |
| 03:10.08 | kergoth | not that anyone cares, since FILESPATH is always set |
| 03:10.14 | kergoth | its set in bitbake.conf |
| 03:10.20 | kergoth | patches welcome |
| 03:11.02 | kergoth | ooh |
| 03:11.05 | kergoth | i think stabbing worked |
| 03:11.43 | kiatoa | kergoth: if I update the .bb file does bitbake see that? It seems not but I wish it did. |
| 03:12.37 | kergoth | ? |
| 03:12.41 | kergoth | of course it does |
| 03:12.55 | kergoth | there's a cache, but the cache for that file will be dirtied on modification |
| 03:13.01 | kergoth | and you can bypass the cache by using -b |
| 03:13.10 | kergoth | alternatively, do a bbread /path/to/filename |
| 03:13.18 | kergoth | and confirm that it looks correct. bbread doesnt use the cache either. |
| 03:13.55 | kiatoa | Ok. Maybe I forgot to save my changes. twice. Ah. Ok. this time it did pick it up. |
| 03:14.05 | kergoth | heh |
| 03:21.12 | keturn | So apparently there is a way to set FILESPATH -- base_set_filespath. Where should that be documented? Not in bitbake documentation, since it's provided by an OE class. In wiki ["bb file"]? |
| 03:22.05 | kergoth | we need an _OE_ users manual. |
| 03:22.12 | kergoth | that documents our classes and stuff |
| 03:24.46 | duffolonious | yes, that would be very nice. |
| 03:25.05 | duffolonious | currently I make conceptual guesses from what I know of portage and ebuilds. |
| 03:29.37 | kiatoa_ | I swear there is something I don't understand going on with dependancies. If a .bb file is changed will do_configure and do_compile get called on running bitbake again? |
| 03:31.12 | treke|home | no |
| 03:32.30 | treke|home | it will only run them if the tasks haven't been completed yet |
| 03:32.44 | kergoth | ls tmp/stamps/ |
| 03:33.12 | kiatoa_ | Oh! Ok. Well, that explains it. Thanks. So even if the task is modified it won't pick it up? |
| 03:33.20 | treke|home | right |
| 03:33.40 | kiatoa_ | Ok. |
| 03:33.41 | treke|home | if bb has finished running the task, it doesn't seen any reason to rerun it |
| 03:33.56 | treke|home | it assumes you will tell it if it needs to redo something |
| 03:38.38 | *** join/#oe cyn (~cyn@adsl-4-200-252.mem.bellsouth.net) |
| 03:40.38 | kiatoa_ | How can I get a list of all the environment variables that are set? Other than digging through the output of set? |
| 03:41.00 | kiatoa_ | I mean environment variables set by bitbake. |
| 03:41.03 | kergoth | ? |
| 03:41.16 | kergoth | bbread /path/to/blah.bb|grep ^export |
| 03:41.19 | kergoth | heh |
| 03:41.28 | kergoth | only vars flagged as export end up in the emitted shell script |
| 03:42.26 | kiatoa_ | Ok. Now I get it. Thanks. |
| 03:42.50 | kergoth | np |
| 03:43.03 | kergoth | 'export' should really be documented in the bitbake users manual. |
| 03:45.36 | keturn | pyflakes says Fetch.__init__ is broken -- it refers to 'd' but 'd' is not one of the arguments |
| 03:45.51 | kergoth | as i said before, patches welcome |
| 03:46.07 | keturn | what's correct behavior there? remove code, or modify signature? |
| 03:46.16 | kergoth | signature? |
| 03:46.32 | keturn | the constructor's argument list, that is |
| 03:46.36 | kergoth | its a base class. that constructor is never actually called anyway. |
| 03:46.43 | kergoth | compare to the classes based on it. like Wget and Cvs |
| 03:46.52 | kergoth | if they supply 'd' in their constructors, so should that one |
| 03:47.02 | kergoth | (going from memory here) |
| 03:47.07 | keturn | neither of which overrides __init__ |
| 03:47.26 | keturn | Most likely case is that Fetch objects are always constructed with an empty list of urls? |
| 03:47.46 | kergoth | i'd say so the urls are likely passed to the methods directly |
| 03:47.53 | kergoth | s/so/so,/ |
| 03:49.12 | kergoth | its simple enough. if the constructor needs d when urls are passed, then it needs a d argument |
| 03:49.19 | kergoth | send a patch to the list |
| 03:59.47 | kergoth | hehe |
| 04:00.19 | keturn | I am somewhat unsettled by your Python style, I think. |
| 04:00.24 | kergoth | most are |
| 04:00.40 | kergoth | that python is nearing perl levels of obfuscation :D |
| 04:00.50 | kergoth | there's a reason we want to get the rewrite going |
| 04:01.28 | kergoth | what i really hate about the current bb python package is that it cant seem to decide whether or not its object oriented |
| 04:01.32 | kergoth | we use objects here, modules there |
| 04:04.01 | keturn | I see. And lots of the classes -- at least in this module -- make extensive use of static methods, which makes them less like objects after all. |
| 04:05.40 | keturn | I am sorely tempted to apply the chainsaw refactoring tool to fetch.py, but if bitbakeng isn't going to use it, I probably shouldn't spend my energy there. |
| 04:07.49 | kergoth | nod. thats why i'm hesitant to clean things up |
| 04:07.52 | kergoth | its just wasted energy |
| 04:07.56 | kergoth | and i _hate_ wasting time |
| 04:17.19 | keturn | I guess this is the disadvantage of a rewrite. It's hard to make incremental contributions before the foundation is rebuilt, and that takes a lot of someone's time. |
| 04:21.17 | kergoth | yep |
| 04:21.23 | kergoth | and we cant even really get coding on it until the design is right |
| 04:21.30 | kergoth | we dont want to fuck it up this time around :) |
| 04:34.26 | mithro | kergoth: bb's python is pretty horrible in many places :/ |
| 04:34.35 | kergoth | yep |
| 04:34.43 | kergoth | it evolved from portage, which wasnt very good in the first place |
| 04:34.51 | kergoth | coupled with my ignorance of python when i was working on it.. |
| 04:34.58 | kergoth | blech |
| 04:35.49 | kergoth | mmmm, guinness |
| 04:36.44 | *** join/#oe noclouds (~mhfan@61.191.204.226) |
| 04:46.40 | chouimat|Zzzz | night |
| 04:54.49 | emte | moo |
| 04:54.56 | kergoth | hey emte |
| 04:54.59 | kergoth | how goes it? |
| 04:55.13 | emte | decent i think and you? |
| 04:55.23 | emte | what do you know about specifix ? |
| 04:55.50 | *** join/#oe T`|Hel (total@GO.HELI.RI.CMU.EDU) |
| 04:56.07 | kergoth | not bad, and i dont even know what it is |
| 04:56.08 | emte | kinda looks like they are doing an almost parallel project |
| 04:57.02 | emte | from the reading it looks like they can already build for arm, mips, etc |
| 04:57.34 | emte | just stumbled across it lookinf for a more ... "devloper"-ish distro |
| 04:57.46 | kergoth | ah. there are a few similar projects to this one. pdaxrom uses a platform builder device thats somewhat similar to oe |
| 04:57.49 | kergoth | less flexible though |
| 04:57.59 | kergoth | developerish? in what way? |
| 04:58.14 | emte | support, extensibility, etc |
| 04:58.56 | *** join/#oe anomaly_ (~pop@dsl-202-173-142-213.sa.westnet.com.au) |
| 04:58.56 | emte | currently need something that will let me install oregano, quartis, smash, etc without overhauling my system |
| 04:59.22 | kergoth | ah |
| 04:59.40 | emte | and preferably preinsalled with ghc since i cant get it to cooperate |
| 05:00.21 | kergoth | i cant stand any distros for my desktop anymore. even debian is barely tolerable. i'm definately throwing an oe built distro together for it one of these days |
| 05:00.24 | kergoth | heh |
| 05:00.33 | emte | basicly i just want a distro with every language compiler and mainstream libs by default ... that would be great |
| 05:01.55 | emte | hurry up and finish teh nextgen bb so we can build it :P |
| 05:02.04 | kergoth | hehe |
| 05:02.09 | kergoth | we'll need to add a good x86 installer to oe |
| 05:02.13 | emte | we can call it cookiemonster |
| 05:02.28 | kergoth | debian-installer's latest betas perhaps, or whatever the redhat one is called |
| 05:03.03 | emte | lots exist, depends how fancy you want it |
| 05:03.11 | kergoth | nod |
| 05:03.14 | kergoth | lets add them all |
| 05:03.19 | kergoth | its the distros decision what to use :) |
| 05:03.19 | emte | lol |
| 05:03.40 | emte | basicly yeah, providing someone maintains em |
| 05:03.49 | mithro | the new debian installer is pretty nice |
| 05:03.53 | kergoth | we need to get more package maintainers |
| 05:03.56 | emte | tho with a larger x86 distro more developers might be willing |
| 05:04.02 | kergoth | convince some oz and fam users to adopt things or something |
| 05:04.05 | kergoth | yeah, thats true |
| 05:04.35 | kergoth | i like the idea of a desktop distro with granular packaging. i hate distros that install crap i dont need |
| 05:04.38 | emte | who's building teh x86 one now? |
| 05:04.58 | kergoth | some people are targeting x86, but its mostly embedded x86 devices. machines that are a known quantity, so dont need an installer |
| 05:05.01 | kergoth | like epia |
| 05:05.06 | kergoth | and the geode stuff CP is doing |
| 05:05.14 | emte | yeah i miss the old 4-hour rh "select-a-package" installer |
| 05:05.26 | kergoth | lol |
| 05:05.30 | kergoth | i like debian |
| 05:05.36 | kergoth | "here's a small box. install what you need" |
| 05:05.40 | mithro | i like debian quite alot |
| 05:05.55 | mithro | i have a debian install with X down to 59mb |
| 05:05.55 | kergoth | no default bigass image. install what you want when you want it |
| 05:05.59 | kergoth | realy appeals to me |
| 05:06.01 | emte | never used debian |
| 05:06.06 | kergoth | course i'm not most users, but.. |
| 05:06.23 | kergoth | emte: after you're done with the installer, it reboots you into a debian rootfs thats like 30 megs or something, on your hd. |
| 05:06.24 | emte | this lunatix is okay ... missing a lot of things i need tho |
| 05:06.28 | kergoth | emte: you apt-get install everything else. |
| 05:06.35 | mithro | oe is like debian with smaller packages :) |
| 05:06.39 | kergoth | yeah |
| 05:07.10 | emte | honestly i still like netbsd the best |
| 05:07.20 | kergoth | i've never actually used netbsd |
| 05:07.27 | kergoth | i've used openbsd before |
| 05:07.31 | kergoth | on my firewall boxes |
| 05:07.35 | kergoth | (ipless) |
| 05:07.42 | kergoth | thats about it though |
| 05:07.43 | emte | openbsd might as well be freebsd compared to net |
| 05:07.57 | kergoth | never used freebsd either :) |
| 05:08.04 | kergoth | what do you like about netbsd? |
| 05:08.09 | emte | net by far has more stingent security auditing of any of teh bsd distros |
| 05:08.19 | kergoth | ah |
| 05:08.25 | emte | its clean |
| 05:08.44 | emte | install image is ~25 mb if i remember |
| 05:09.17 | emte | you basicly just install a shell with basic console tools |
| 05:09.26 | *** join/#oe luke-jr_ (~luke-jr@207.192.219.246) |
| 05:09.34 | emte | then install audited and patched sources from netbsd's tree |
| 05:09.49 | kergoth | i like that |
| 05:10.03 | emte | problem is its bsd |
| 05:10.05 | kergoth | that sort of model is what i want for my desktop linux distro, really. |
| 05:10.29 | emte | if bsd wasnt so slow in hardware support .. |
| 05:10.36 | kergoth | debian is similar in that regard, minus the stringent security auditing |
| 05:10.43 | kergoth | small base image, install the rest |
| 05:10.54 | kergoth | er, i forgot to eat dinner |
| 05:10.57 | emte | lol |
| 05:11.05 | emte | yeah back to studying with myself |
| 05:11.26 | mithro | studing with yourself is that kinda like playing with yourself but less fun? |
| 05:11.48 | mithro | bah it's to hot here |
| 05:18.53 | *** join/#oe dougt (~dougt@adsl-63-198-216-19.dsl.snfc21.pacbell.net) |
| 05:23.55 | *** join/#oe Mortis (~chatzilla@p14013-ipadfx41osakakita.osaka.ocn.ne.jp) |
| 05:27.18 | *** join/#oe hufnus (~slonsiki@63.93.79.7) |
| 05:29.32 | *** join/#oe zwi (~zwi@216.88.131.43) |
| 05:32.29 | *** join/#oe zedkatuf (~dfutak@82-32-58-45.cable.ubr08.azte.blueyonder.co.uk) |
| 05:46.39 | *** join/#oe Luke-Jr (~luke-jr@207.192.219.246) |
| 05:59.28 | *** join/#oe andersee (~andersen@206.170.30.2) |
| 06:30.15 | *** join/#oe Marsupilami24 (~Marsupila@152.139-60-66-nokia-static.surewest.net) |
| 06:47.08 | *** join/#oe gints (~gints@159.148.213.135) |
| 06:58.52 | emte | mithro yeah pretty much |
| 07:00.05 | emte | wonder if i should try specifix |
| 07:00.36 | emte | need to findout how to get an altera subscription for quartis as well |
| 07:17.34 | *** join/#oe micropal (ident@139.80-202-82.nextgentel.com) |
| 07:21.38 | *** join/#oe Q_Continuum (~Q_Continu@rrcs-67-52-219-78.west.biz.rr.com) |
| 07:26.32 | *** join/#oe andersee (~andersen@206.170.30.2) |
| 07:26.55 | *** join/#oe kurre2 (~tomimo@a84-231-39-238.elisa-laajakaista.fi) |
| 07:31.43 | *** join/#oe cedric__ (~cedric@195.94.59.29) |
| 07:31.46 | cedric__ | ~logs |
| 07:31.47 | ibot | from memory, logs is http://ibot.rikers.org/<channelname>/ where channelname is html encoded ie: %23debian, or stats for some channels at http://ibot.rikers.org/stats/<channelname>.html.gz for apt, ibot, jbot and purl. |
| 07:47.55 | *** join/#oe Gnea (gnea@garson.org) |
| 07:56.33 | *** join/#oe Toi (~pleemans@dD577A871.access.telenet.be) |
| 08:58.20 | *** join/#oe XorA (~dp@81-178-244-41.dsl.pipex.com) |
| 09:15.57 | *** join/#oe rob_w (~roboo@p213.54.155.218.tisdip.tiscali.de) |
| 09:18.44 | *** join/#oe mallum (~mallum@host81-153-52-173.range81-153.btcentralplus.com) |
| 09:19.32 | *** join/#oe schurig (~schurig@p5080BE8A.dip0.t-ipconnect.de) |
| 09:19.52 | schurig | hmm, anyone knows why ckermit has been deleted from OE? |
| 09:23.51 | schurig | what's the IRC name of the BK name "ggilbert" ? |
| 09:25.46 | ljp | treke |
| 09:26.50 | koen | good morning all |
| 09:28.25 | schurig | treke-: any reason you deleted ckermit.bb? |
| 09:29.19 | schurig | treke-: I've not yet used the OE version of ckermit, but I'm using the OZ version of ckermit for low-level hardware debug to my built-in GSM/GPRS modem |
| 09:29.44 | schurig | treke-: so once I got rid of all old OZ packages, I'll need a ckermit anyway :-) |
| 09:38.46 | *** join/#oe cedric (~cedric@195.94.59.29) |
| 09:42.04 | *** join/#oe Ravens (~Ravens@odaiba.imag.fr) |
| 09:55.05 | *** join/#oe tigrux (~tigrux@charybdis-ext.suse.de) |
| 10:05.29 | RP|zZz | schurig: It was a licencing problem. ckermit's licence was pretty restrictive |
| 10:05.56 | schurig | RP|zZz: I know. But the license allows compilation of the program |
| 10:06.07 | schurig | RP: good morning |
| 10:06.17 | RP | morning all :) |
| 10:06.55 | schurig | RP: and so why should one not compile and use this program? At least it would have been nice to make some non-free directory and move it over, just like we have non-working |
| 10:07.33 | RP | schurig: I don't think anyone would be allowed to distribute binaries and therefore its use was massively limited |
| 10:08.05 | RP | schurig: I didn't look at it and you'll have to ask treke- for more info... |
| 10:08.28 | RP | It was said that there are other kermits around with less problematic licences... |
| 10:10.07 | schurig | RP: you think wrong. Even Debian distributes ckermit |
| 10:10.19 | schurig | Package: ckermit |
| 10:10.19 | schurig | Priority: extra |
| 10:10.20 | schurig | Section: non-free/comm |
| 10:10.29 | schurig | (that's from "apt-cache show ckermit") |
| 10:11.17 | schurig | ~lart treke |
| 10:11.17 | schurig | :-) |
| 10:13.43 | schurig | mithro: and, things in non-free ARE distributable |
| 10:13.47 | schurig | The LICENSE says: |
| 10:13.49 | schurig | "This is the new C-Kermit 7.0 and 8.0 license. The intention is to allow |
| 10:13.49 | schurig | C-Kermit to be distributed with "free" operating systems such as GNU/Linux, |
| 10:13.49 | schurig | FreeBSD, NetBSD, OpenBSD, The Hurd, etc |
| 10:14.17 | schurig | It's in non-free because the "you have to send source changes back to us" clause in the LICENSE, which is considered non-free by the 130%ers |
| 10:14.51 | RP | Did the .bb file need to modify the source at all? |
| 10:15.26 | RP | If so, there's your problem... |
| 10:16.00 | schurig | RP: AFAIK I only modified Makefile to get it cross-compiled |
| 10:16.35 | XorA | schurig: Makefile is source if its supplied with the program |
| 10:16.46 | schurig | RP: and this in a section where the Makefile itself writes "You need to modify this to compile it on SUN, that for HPUX, another define for FreeBSD and this LIBS= for Linux" |
| 10:17.01 | XorA | though can you call smtp from bb file? auto email the diff everytime its applied |
| 10:17.01 | schurig | you're all crazy |
| 10:17.24 | schurig | if the Makefile says "Modify me to get it compiled" then this is okay. |
| 10:17.37 | schurig | after all, they didn't use automake/autoconf |
| 10:18.27 | XorA | schurig: the trouble comes when you think its ok for one package, then another comes along, then another, then the lawyers |
| 10:18.45 | XorA | schurig: when we invent a lawyer eating virus we can all relax |
| 10:19.00 | schurig | XorA: don't you think that the Debian project modifies the Makefile? Hey, they even didn't have some "make install DESTDIR=blah" feature, which is what Debian needs |
| 10:19.04 | schurig | you want to be holier than the pope |
| 10:19.23 | schurig | and ckermit is in RedHat, Debian, SuSE etc |
| 10:19.35 | schurig | just some people that don't have brains want to get rid of it in OE |
| 10:19.36 | schurig | wow |
| 10:21.00 | XorA | schurig: well the way bitbake is setup and oe is constructed there is no reason you cant do an overlay :-D if you cant persuade the guru's to allow it |
| 10:21.12 | schurig | I AM A GURU! |
| 10:21.20 | schurig | I was one of the people that brought OE to live |
| 10:21.37 | schurig | I worked 3 months on oebuild/oemake before kergoth took over and I made most of the design |
| 10:22.00 | schurig | So don't point me to some higher being |
| 10:22.53 | mithro | schurig: you wrote patcher right? |
| 10:23.01 | schurig | I'll readd this tonight |
| 10:23.05 | schurig | mithro: yes |
| 10:23.08 | XorA | schurig: so if you have write access just undelete it |
| 10:23.14 | schurig | mithro: but originally not for OE, but for my kernel work |
| 10:24.00 | koen | schurig: or make a non-free dir and put ckermit in there |
| 10:24.27 | mithro | was it properly marked with a license? |
| 10:25.31 | schurig | koen: that would be acceptable |
| 10:25.48 | schurig | mithro: why don't you read? |
| 10:26.07 | schurig | mithro: "# This software comes with the same License as Perl." |
| 10:26.19 | mithro | schurig: i mean in the OE file |
| 10:26.51 | schurig | mithro: the LICENSE= in the .bb file is a hint, the license that is in the software is the one that counts |
| 10:27.25 | schurig | mithro: and patcher-native.bb/patcher-native.oe is way older than our peppering of .oe/.bb files with LICENSE= lines ... |
| 10:27.25 | koen | schurig: untill there is a clear policy people have to move "nonfree" stuff in that dir instead of rm'ing it |
| 10:27.41 | mithro | schurig: but i belive treke was cleaning up files which didnt have a license |
| 10:27.54 | schurig | mithro: so a missing LICENSE= line doesn't say that it is nonfree or removable |
| 10:28.15 | mithro | i was actually talking about ckermit |
| 10:29.57 | *** join/#oe pb_ (~pb@2002:3e03:42cc:2:20a:5eff:fe00:391e) |
| 10:33.02 | RP | schurig: Obviously treke doesn't want to mirror the ckermit source and I can understand why, even if I agree or disagree myself. I don't see the point in a non free directory. You just need to come up with a LICENCE line that everyone is happy with... |
| 10:34.12 | *** join/#oe robtaylor_ (~robtaylor@217.204.121.82) |
| 10:34.12 | schurig | RP: if he doesn't want to mirror something, than this is his decision. Maybe next week he decides that he doesn't want to mirror qt-embedded. That's ok. But it would not be OK to delete the .bb file for things that he doesn't like to mirror. |
| 10:34.54 | schurig | RP: he can't just delete all .bb files for things that pleases him. His mirror is his thing, but OE is something that is not owned by one, it's somethng collaborative. |
| 10:36.10 | XorA | schurig: Youve convinced me actually! sorry! |
| 10:37.14 | RP | schurig: I got the impression it was a long running issue and I have no idea of the background. All I've done it try and say why I think he did it. My opinion remains neutral... |
| 10:37.38 | schurig | RP: that's ok. I'll try to meet him this evening in IRC |
| 10:38.20 | *** join/#oe MasterJim (~MasterJim@lns-vlq-8-tou-62-147-189-97.adsl.proxad.net) |
| 10:38.25 | *** part/#oe MasterJim (~MasterJim@lns-vlq-8-tou-62-147-189-97.adsl.proxad.net) |
| 10:42.52 | schurig | and debian is at version 211, not 209 |
| 11:03.28 | *** join/#oe kjoe (~tody@bantha.ftk.de) |
| 11:08.10 | *** join/#oe cedric (~cedric@195.94.59.29) |
| 11:10.51 | *** join/#oe IRC_noobee (~IRC_noobe@p548315B7.dip0.t-ipconnect.de) |
| 11:11.28 | *** join/#oe zinga (~arnaudb@186.80-203-227.nextgentel.com) |
| 11:11.38 | IRC_noobee | Hi everybody |
| 11:15.16 | *** join/#oe Virusmaster (~Thomas@p5091AAE4.dip.t-dialin.net) |
| 11:15.40 | XorA | ji IRC_noobee |
| 11:31.00 | IRC_noobee | Did my bb file for anthy make it to the ML? |
| 11:31.30 | *** join/#oe reenoo ([nbbzluweq@x18.informatik.uni-bremen.de) |
| 11:32.26 | reenoo | afternoon |
| 11:43.13 | *** join/#oe micropal_ (ident@139.80-202-82.nextgentel.com) |
| 11:48.12 | webmind | hmm sorry, but if I try to do a 'bk pull; bk -r co -q' to update my oe tree.. I get 'pull: cannot find package root' and 'bk: Cannot find package root.' |
| 11:48.19 | webmind | anyone any idea what this can be ? |
| 11:48.55 | schurig | webmind: Do you have a "BitKeeper" directory in the dir where you do "bk pull" ? |
| 11:48.57 | XorA | webmind: oe doesnt exist anymore, the build system is now bitbake and the packages are now openembedded |
| 11:49.14 | webmind | XorA, ok |
| 11:49.18 | schurig | webmind: if yes, what says "bk parent" ? |
| 11:49.46 | schurig | webmind: it should say something like "bk://openembedded.bkbits.net/openembedded" |
| 11:49.50 | webmind | no BitKeeper directory ? |
| 11:51.18 | webmind | hmm, brb (forced to leave terminal) |
| 11:53.55 | schurig | webmind: when you issue a "bk ..." command in a directory that was not created with "bk clone", then there is no BitKeeper/ directory and you see "parent: cannot find package root" as error. |
| 11:54.26 | schurig | webmind: if you have a BitKeeper directory, then probably only your parent is wrong, OE moved to a different location for anonymous access maybe a month ago |
| 11:54.56 | schurig | You might also try to restart from scratch, just follow http://openembedded.org/cgi-bin/moin.cgi/GettingStarted |
| 11:57.10 | IRC_noobee | Good news ;-) Just got off the phone with a good friend who in exchange for me installing Debian on his machine (hey, another potential Win2Lin convert) there |
| 11:58.01 | IRC_noobee | is willing to share his 2.2GHz, 512MB RAM machine with 10 MBit SDSL connection with me for compiling OE stuff. |
| 11:58.08 | IRC_noobee | Is that cool or what? |
| 11:58.32 | IRC_noobee | No more "compiliation" on my 100MHz router. ;-) |
| 11:58.33 | schurig | IRC_noobee: just install distcc there :-) |
| 11:58.55 | IRC_noobee | You wanna share? |
| 11:59.17 | IRC_noobee | Actually I was already thinking about something like that for me at home. |
| 11:59.31 | IRC_noobee | You guys know about openmosix.org? |
| 11:59.59 | IRC_noobee | It is a cluster that does not require changes to source code for programs to run |
| 12:00.02 | schurig | distcc works when on both systems the same compiler exists, just like you can put "ccache arm-linux-gcc" you can do "ccache distcc arm-linux-gcc". |
| 12:00.17 | IRC_noobee | I know |
| 12:00.37 | schurig | IRC_noobee: it will call the precrozessor cpp on your box, so that on the other box you don't have to install all header file (or all of OE) and just let the other box compile |
| 12:00.46 | schurig | AFAIK final linking is done on your box again |
| 12:00.55 | IRC_noobee | schurig: OK, that I did not know |
| 12:01.12 | IRC_noobee | Thank you for the hint. |
| 12:02.01 | schurig | IRC_noobee: just look at OE/conf/bitbake.conf, you can put the distcc call into the CCACHE variable as well, near line 190. |
| 12:02.32 | IRC_noobee | Cool, thank you for the hint. |
| 12:03.29 | RP | schurig / IRC_noobee: That still leaves you with the overhead of bitbake though which was the original problem... |
| 12:03.44 | IRC_noobee | RAM-wise? |
| 12:03.52 | schurig | RP: he just wrote 100 MHz, not 32 MB :-) |
| 12:04.14 | schurig | IRC_noobee: if you are low on RAM, then use ssh :-) |
| 12:04.23 | IRC_noobee | I'll just install the complete bitbake on my friend's machine |
| 12:05.20 | IRC_noobee | I'll ssh anyway to "admin" his machine. |
| 12:06.43 | *** join/#oe mithro (~tim@dsl1-83.gw1.adl1.airnet.com.au) |
| 12:09.25 | IRC_noobee | II do still have a question regarding "make -j #" in conf/local.conf |
| 12:09.58 | IRC_noobee | Will "make -j 2" for example double RAM reqs or will they be about the same |
| 12:10.27 | IRC_noobee | IOW are the 2 compilation processes "complete" compilation processes with full RAM reqs? |
| 12:11.01 | RP | IRC_noobee: It just means more make processes in do_compile. They aren't fully blown bitbake ones... |
| 12:11.17 | RP | So, no, the RAM requirements will be much the same |
| 12:12.01 | IRC_noobee | OK, then bitbake is a good candidate for parallel execution on an openmosix.org cluster and that might make it even feasible to compile here at home |
| 12:12.08 | IRC_noobee | where I want to install such a cluster |
| 12:12.27 | IRC_noobee | Just doubled my RAM but still short of 512MB |
| 12:12.51 | IRC_noobee | But distributed compilation might then work |
| 12:13.18 | *** join/#oe stevenh (~xev@65.167.23.2) |
| 12:14.05 | IRC_noobee | An openmosix cluster is similar to an SMP machine |
| 12:14.30 | XorA | IRC_noobee: cluster all the Z's in the world, call them wintermute, retire in a hurry :-) |
| 12:14.51 | IRC_noobee | Will do ;-) |
| 12:15.10 | IRC_noobee | After I get anthy and CRIM to work in OZ |
| 12:15.33 | IRC_noobee | I will try get the openmosix patch into the OZ kernel |
| 12:15.45 | IRC_noobee | Now that would indeed be super-geeky |
| 12:16.01 | IRC_noobee | via WLAN, of course, that is <lol> |
| 12:18.08 | IRC_noobee | openmosix has the advantage that nodes can leave and join w/o disturbing the cluster |
| 12:20.33 | webmind | schurig, uhm.. I did create the directory with bk clone ? and just found my error.. I was in '/stuff/' instead of '/stuff/openembbed/' :) |
| 12:20.36 | webmind | thnx |
| 12:21.00 | schurig | webmind: yes, "bk clone" automacally create the BitKeeper/ directory |
| 12:21.28 | webmind | I noticed :) |
| 12:21.31 | webmind | thnx |
| 12:32.54 | *** join/#oe anomaly_ (~pop@dsl-202-173-142-212.sa.westnet.com.au) |
| 12:34.34 | IRC_noobee | How much HD space should I set aside for compiling two input methods (and the required dependies, not sure what those are, though. opie maybe)? |
| 12:34.37 | IRC_noobee | Do not want to clog up my friend's machine too much. |
| 12:34.57 | koen | 5-6GB |
| 12:35.09 | IRC_noobee | OK, thank you. |
| 12:46.21 | *** join/#oe koen_ (koen@dominion.kabel.utwente.nl) |
| 12:55.30 | *** part/#oe IRC_noobee (~IRC_noobe@p548315B7.dip0.t-ipconnect.de) |
| 13:02.29 | *** join/#oe Disconnect (~dis@2001:5c0:828b:0:203:6dff:fe1a:c5b6) |
| 13:06.03 | *** join/#oe mrz80 (1000@caledonia.cns.ufl.edu) |
| 13:42.00 | *** join/#oe hufnus (~slonsiki@63.93.79.92) |
| 13:50.13 | *** join/#oe IRC_noobee (~IRC_noobe@G070e.g.pppool.de) |
| 13:51.14 | cedric | stupid question, I've add new dest in my ipkg.conf -> /home |
| 13:51.51 | cedric | when I install something to the dest home, do I need to ipkg-link? |
| 13:52.11 | *** join/#oe robtaylor_ (~robtaylor@217.204.121.82) |
| 13:57.07 | *** join/#oe raster (~raster@YahooBB219198104088.bbtec.net) |
| 13:59.55 | *** join/#oe zwi (~zwi@ewa-denver.com) |
| 14:00.51 | *** join/#oe micropal (ident@139.80-202-82.nextgentel.com) |
| 14:00.53 | *** join/#oe Tygger-Bob (~myname@64.26.144.98) |
| 14:02.48 | *** join/#oe jamie (~jamie@softmodem.org) |
| 14:13.10 | *** join/#oe jacques_ (~jacques@c-67-171-198-193.client.comcast.net) |
| 14:20.53 | *** join/#oe fiam (~foo@80.103.17.7) |
| 14:20.58 | fiam | hi |
| 14:21.13 | fiam | anybody can tell me who's the konqueror-embedded package maintainer? |
| 14:21.23 | koen | the .bb? |
| 14:21.29 | fiam | *.bb file doesn't mention any maintaniner : |
| 14:21.36 | koen | aha |
| 14:21.53 | fiam | koen: :) |
| 14:23.33 | Cwiiis | cedric: Yes, unless you're using unionroot (in which case though, you'd want it in a subdirectory of home..) |
| 14:24.07 | koen | fiam: http://openembedded.bkbits.net:8080/openembedded/related/packages/konqueror/konqueror-embedded_20030705.bb?nav=index.html|src/|src/packages|src/packages/konqueror |
| 14:24.17 | koen | pick one of them to blame :) |
| 14:26.10 | fiam | koen: thanks :) |
| 14:28.55 | fiam | bye |
| 14:31.47 | cedric | Cwiiis: can u explain a bit more because when i do "ipkg-link mount /home" it can't find anything... |
| 14:38.40 | Cwiiis | cedric: I've heard that ipkg-link doesn't check /etc/ipkg.conf - So without editing the script, or doing it manually, you might be a bit stuck |
| 14:39.35 | *** join/#oe chouimat|Zzzz (~dieu@r2351064.cidc.net) |
| 14:44.47 | cedric | Cwiiis: what's your unionroot thingy?? |
| 14:47.08 | Cwiiis | cedric set PREFERRED_VERSION_task-bootstrap = "1.0unionroot" and find out :) hehe (although it may not work with opie and gpe yet, I'm still testing...) - It lets you install to external media without linking back to root by mounting a subdirectory on that media over root via unionfs (that subdirectory is preset to 'packages' at the moment, to correspond to ipkg-collateral) |
| 14:47.15 | *** join/#oe robtaylor_ (~robtaylor@217.204.121.82) |
| 14:48.35 | cedric | Cwiiis: when is that going to be ready because that's sound cool |
| 14:49.51 | Cwiiis | cedric: It *may* make it into oz3.5.3, but it all depends on how much testing it gets and how many bugs remain by that point |
| 14:51.13 | Cwiiis | There are a couple of bugs at the moment that are pretty easy to resolve that I just don't have the time to do at the moment (suspend/resume tends not to work properly, for example)... My top priority is to just get gpe started with it |
| 14:51.22 | Cwiiis | Doing exams at the moment though, so not much time :p hehe |
| 14:52.00 | *** join/#oe joshua_ (joshua@joshuawise.com) |
| 14:52.02 | koen | Cwiiis: it is indeed that time of the year again |
| 14:52.16 | cedric | Cwiiis: so for the time been, where should look to hack the ipk-link |
| 14:52.59 | Cwiiis | cedric: I know that someone made an ipkg-link that used /etc/ipkg.conf a long time ago because I helped (or rather, suggested changes) them with it... I don't know what happened to it though :/ |
| 14:55.49 | *** join/#oe dalibor-lu (~knoppix@natkirchberg.crpht.lu) |
| 15:07.59 | florian_kc | mickey|sick: hi, that doesn't sound good. |
| 15:08.38 | mickey|sick | hi florian_kc. just a fever + headache. nothing serious |
| 15:08.43 | mickey|sick | (at least i hope so) |
| 15:09.43 | *** join/#oe prpplague (~dave@mailhost.amltd.com) |
| 15:11.01 | *** join/#oe stevenh (~xev@65.167.23.2) |
| 15:11.20 | florian_kc | mickey|sick: but nasty :-/ |
| 15:11.23 | Cwiiis | XorA: Because the Z is primarily a linux pda? :p hehe |
| 15:12.20 | XorA | Im sure I just agreed to have my guts ripped out while Im still alive |
| 15:14.50 | *** join/#oe treke|home (~ggilbert@69-175-32-221.ventca.adelphia.net) |
| 15:15.13 | mickey|sick | the problem is just getting a sane bootloader onto the device. bricking a Sl5500 and Sl5000 is almost as easily as bricking an ipaq w/ windows on. once you have the booloader sorted out, you won't touch it anymore and there's not much risk in updating left. |
| 15:16.33 | XorA | mickey|sick: Im not really that bothered, updating works Ipaq, it just struck me as sillyly scary |
| 15:16.53 | XorA | mickey|sick: so much for windows being nice |
| 15:18.02 | chouimat | morning |
| 15:18.29 | mickey|sick | treke: fyi: i just built a fresh image w/ openzaurus-pxa-2.6 as machine and it boots right into Opie. ts working, keys working (thanks to zeckes patch that's not in bk yet but will appear soon) |
| 15:18.53 | kergoth`zzz | nice |
| 15:19.03 | pb_ | g'day kergoth |
| 15:19.17 | pb_ | mickey|sick: very good! |
| 15:19.31 | koen | 'morning kergoth |
| 15:19.36 | mickey|sick | if you want to try, i put it online @ http://vanille.de/temp/ |
| 15:19.37 | treke|home | mornin k |
| 15:19.55 | treke|home | mickey|sick: cool. I was going to start a build last night, but forgot :p |
| 15:20.35 | treke|home | started to sort out the mirroring stuff a bit |
| 15:26.17 | Cwiiis | hmm... A gdm-style 'X keeps restarting' script would be useful for gpe-dm :p |
| 15:45.16 | mickey|sick | anyone ever used 2.6 hotplug firmware downloading? |
| 15:45.46 | pb_ | yah, I use that for my DVB cards |
| 15:45.56 | pb_ | and for my bluetooth card, come to think of it |
| 15:46.04 | mickey|sick | what do i need to do to make that work? |
| 15:46.17 | mickey|sick | putting firmware into a special directory and how to tell hotplug to do it then? |
| 15:46.27 | *** join/#oe zecke_ (~ich@83.169.146.40) |
| 15:46.40 | pb_ | afaik, it should just be a case of dropping the files into /usr/lib/firmware/ |
| 15:46.50 | pb_ | hotplug should already include all the right bits to do what the driver asks for. |
| 15:47.03 | mickey|sick | ah, so hotplug communicates with the driver then |
| 15:47.10 | mickey|sick | don't need to tell it which userland util to use or so |
| 15:47.38 | mickey|sick | interesting. /me goes reading more about that |
| 15:48.56 | mallum | whats the best way of having different bb's for the same software but built with different ./configure options ? |
| 15:49.15 | mickey|sick | factor out a .inc file with everything common |
| 15:49.29 | mickey|sick | then include that and overwrite EXTRA_OECONF in each of the variants |
| 15:50.09 | mallum | mickey|sick, thanks, do you know of an existing package which does that ? |
| 15:50.25 | mickey|sick | mallum: yep, hold on |
| 15:51.27 | koen | mallum: one of the matchbox packages does it |
| 15:51.35 | koen | bluez-untils too |
| 15:51.39 | mickey|sick | right |
| 15:51.40 | koen | utils* |
| 15:53.42 | mallum | koen, it uses a .inc but no EXTRA_OECONF ? |
| 15:54.22 | Cwiiis | argh, why is gpe-dm crashing... |
| 15:54.54 | mallum | aha found the cairo dep ! |
| 15:55.14 | mallum | pb_, libxsettings-client is dependant on libgpe widget, should it be ? |
| 15:56.23 | *** join/#oe marcansoft (1337@80.26.222.198) |
| 15:58.22 | *** join/#oe marcansoft (1337@80.26.222.198) |
| 16:01.07 | mallum | mickey|sick, okey I so it would be like libmatchbox_EXTRA_OECONF="--blah --blah" in my local.conf right ? |
| 16:01.18 | mallum | mickey|sick, but what happens if I want to change depends too ? |
| 16:01.26 | mickey|sick | i thought more along the lines of |
| 16:01.36 | mickey|sick | variant1.bb: |
| 16:01.38 | mickey|sick | include common.inc |
| 16:01.41 | mickey|sick | EXTRA_OECONF = "foo" |
| 16:01.43 | mickey|sick | variant2.bb |
| 16:01.45 | mickey|sick | include common.inc |
| 16:01.48 | mickey|sick | EXTRA_OECONF = "bar" |
| 16:03.26 | mallum | mickey|sick, ok, but then how are the .bb's versioned/selected to be built ? |
| 16:03.57 | Cwiiis | florian_kc: What are the first few things gpe-login does? Is it possible to enable some sort of verbose mode to get output? It's crashing for me and I don't see why :/ |
| 16:04.00 | mickey|sick | well, the common.inc would PROVIDE = "matchbox" |
| 16:04.14 | mickey|sick | mallum: then you can choose PREFERRED_PROVIDER_matchbox = "variant1" |
| 16:04.15 | mallum | also PREFERED_VERSIONS in my local.conf do not overide the distro's ones ? |
| 16:04.16 | mickey|sick | e.g. |
| 16:04.38 | Cwiiis | florian_kc: dmesg says 'unable to handle kernel NULL pointer dereference' :/ |
| 16:04.48 | mickey|sick | mallum: that's right. distro.conf is included after local.conf (of course, because in local.conf you specify which distro to include) |
| 16:05.06 | florian_kc | Cwiiis: yes you can run it from console and make it stay there... |
| 16:05.11 | mallum | mickey|sick, ah of course |
| 16:05.16 | florian_kc | Cwiiis: oh that sounds bad! |
| 16:05.24 | mallum | mickey|sick, any way to work round that ? |
| 16:05.45 | mickey|sick | mallum: use a custom distro |
| 16:05.50 | Cwiiis | florian_kc: No output on the console :/ |
| 16:05.55 | mallum | mickey|sick, okey |
| 16:07.26 | florian_kc | Cwiiis: try gpe-dm -n |
| 16:08.42 | *** join/#oe tomimo (~kurre@a84-231-39-238.elisa-laajakaista.fi) |
| 16:08.58 | pb_ | mallum: I don't think it should be. |
| 16:09.18 | mallum | pb_, no me niether xsettings-client should just need xlibs |
| 16:09.27 | pb_ | right |
| 16:09.32 | mallum | pb_, but gpewidget is listed on DEPENDS |
| 16:10.00 | Cwiiis | florian_kc: Ah, seems like X can't find fonts... |
| 16:11.12 | *** join/#oe reenoo ([sGqMcNZsJ@x23.informatik.uni-bremen.de) |
| 16:11.29 | Cwiiis | florian_kc: Well, I get a lot of failed to initialise font path errors, followed by two segfaults :p hehe |
| 16:11.32 | pb_ | mallum: try removing it from DEPENDS and see what happens. |
| 16:11.41 | pb_ | if it still builds ok, I'll check that change in |
| 16:12.32 | reenoo | re |
| 16:14.28 | Cwiiis | florian_kc: Do you have /usr/lib/X11 in your gpe-image? |
| 16:21.43 | florian_kc | Cwiiis: nope and this is good |
| 16:22.10 | florian_kc | Cwiiis: some complains about missing font paths are ok |
| 16:22.29 | Cwiiis | florian_kc: Right, that's not the problem... Seems I'm just getting segfaults without error messages... Will save the error messages |
| 16:22.51 | Cwiiis | I thought it was >& to redirect stderr? |
| 16:23.26 | reenoo | 2>&1 |
| 16:23.33 | Cwiiis | ah, ok, thanks |
| 16:23.34 | kergoth | bleh |
| 16:23.38 | kergoth | boo work |
| 16:23.57 | reenoo | hey kergoth |
| 16:24.10 | kergoth | mickey|sick: the distro .conf's should really be using ?= .. |
| 16:25.48 | mickey|sick | what does that do? is that a conditional assignment if ! empty? |
| 16:25.55 | kergoth | yep |
| 16:25.58 | mickey|sick | oh cool |
| 16:26.00 | mickey|sick | didn't know we had that |
| 16:26.06 | kergoth | its used already in a number of distro .conf's so we dont override the user |
| 16:26.12 | kergoth | where it isnt, it should be |
| 16:29.17 | kergoth | random thought this morning |
| 16:29.19 | kergoth | i was thinking we should: |
| 16:29.27 | kergoth | include conf/build/${BUILD_ARCH}.conf |
| 16:29.30 | kergoth | and OS, and ARCH |
| 16:29.33 | kergoth | same for target |
| 16:29.37 | kergoth | then they're independent |
| 16:29.48 | kergoth | and we can easily fix BUILD_OS on platforms whose uname isnt what we need |
| 16:29.54 | kergoth | (like freebsd, mac osx..) |
| 16:32.46 | mallum | pb_, okey thanks |
| 16:34.05 | Cwiiis | hmm, xtscal is segfaulting for me :/ |
| 16:38.14 | mallum | mickey|sick, a package version is optionally set in the .bb filename ? |
| 16:38.54 | pb_ | yeah |
| 16:39.27 | pb_ | the filename is parsed into PN{_PV{_PR}}.bb |
| 16:39.46 | pb_ | if it doesn't specify PV and/or PR, they default to "1.0" and "r0" respectively. |
| 16:39.58 | mallum | ok |
| 16:40.18 | mallum | and the .b contents can overide PV |
| 16:40.28 | kergoth | right |
| 16:40.33 | mallum | cool |
| 16:55.30 | mallum | kergoth, is it possible to 'overide' a .bb completely, what happens if another .bb provides the same version etc |
| 16:57.16 | Tygger-Bob | any ideas why this CFLAGS_prepend isn't working? |
| 16:57.16 | Tygger-Bob | http://pastebin.ca/4767 |
| 16:58.02 | Tygger-Bob | or is it that "unset CFLAGS" thing that I just noticed after I pasted that |
| 16:59.36 | kergoth | mallum: if two bb's PROVIDES the same thing, you can select between them based on their PN using PREFERRED_PROVIDERS |
| 16:59.57 | *** join/#oe gints_ (~gints@195.2.115.160) |
| 17:00.04 | kergoth | mallum: if two bb's provide the same thing and have the same PN but different versions, itll use the latest, unless otherwise specified with PREFERRED_VERSION_pn |
| 17:00.09 | pb_ | Tygger-Bob: yah, that'd be it |
| 17:00.14 | mallum | kergoth, yeah but what if the first dont PROVIDES ? |
| 17:00.26 | kergoth | mallum: if two bb's provide the same thing and have the same PN and PV, the behavior isnt well defined, unless you're using package collections |
| 17:00.30 | kergoth | mallum: what do you mean? |
| 17:00.45 | kergoth | mallum: every .bb provides at a minimum, its package name, packagename-version, and packagename-version-revision. |
| 17:00.52 | mallum | kergoth, Im just thinking of a quick way to toy with existing |
| 17:00.54 | kergoth | mallum: as you can see in bitbake.conf |
| 17:01.08 | kergoth | mallum: you want to be using collections for your local package modifications. |
| 17:01.37 | mallum | kergoth, collections ? |
| 17:01.43 | mallum | kergoth, that sounds what I need |
| 17:01.47 | kergoth | its a means of having multiple local package areas |
| 17:01.52 | kergoth | each with a defined priority |
| 17:02.08 | mallum | kergoth, yep, thats what I need |
| 17:02.08 | kergoth | when the same package exists in multiple places with the same version, and you're using that, itll use that priority to choose between them |
| 17:02.09 | treke- | kergoth: Got a bitbake internals question. Would it be possible to load a bb, save the state before overrides are run, and then rerun the overrides multiple times? |
| 17:02.19 | treke- | mallum: one sec I've got an example somewhere |
| 17:02.26 | kergoth | treke-: not without modifying the parsing code. |
| 17:02.36 | kergoth | treke-: the caller doesnt have that degree of control |
| 17:02.39 | treke- | darn |
| 17:02.49 | treke- | mallum: http://treke.net/~ggilbert/local.conf |
| 17:02.58 | kergoth | mallum: http://bitbake.berlios.de/manual/ch03s02.html#id2527092 |
| 17:02.59 | mallum | treke-, thanking you sir |
| 17:03.02 | kergoth | its in the manual :) |
| 17:03.10 | treke- | or read the manual :p |
| 17:03.37 | kergoth | in fact, most of your questions so far are answered there |
| 17:03.38 | kergoth | if not all |
| 17:03.41 | kergoth | so yeah, rtfm |
| 17:03.42 | kergoth | :) |
| 17:03.57 | mallum | hehe |
| 17:03.59 | treke- | no one here is used to having an fm to rt |
| 17:04.02 | kergoth | hehe |
| 17:04.41 | kergoth | pb_: you saw nico's implementation of TLS for arm on l-a-k, i assume? |
| 17:05.15 | pb_ | no, I didn't see that. let me take a look. |
| 17:06.46 | pb_ | grr, our damn firewall is acting up again. taking forever to get to my mailbox. |
| 17:07.42 | *** part/#oe IRC_noobee (~IRC_noobe@G070e.g.pppool.de) |
| 17:07.52 | pb_ | kergoth: picked up my mp70 this afternoon, by the way. |
| 17:07.56 | pb_ | just gotta install linux on it now |
| 17:08.00 | kergoth | cool, how is it? |
| 17:08.13 | pb_ | seems pretty nice so far |
| 17:08.22 | kergoth | how's the construction? seem solid? |
| 17:08.49 | pb_ | yup, feels robust enough. |
| 17:09.05 | mallum | pb_, whats a mp70 ? |
| 17:09.09 | kergoth | cool. that was the only thing i was concerned about. |
| 17:09.18 | kergoth | might have t o order one of those when i get my profit sharing next month |
| 17:10.17 | pb_ | mallum: it's a sharp laptop. |
| 17:10.38 | pb_ | "pc-mp70g" to give it the full title |
| 17:11.26 | mallum | pb_, shiny :) |
| 17:11.37 | mallum | pb_, from japan ? |
| 17:11.47 | pb_ | yup |
| 17:12.25 | mallum | pb_, customs get a cut ? |
| 17:12.41 | pb_ | mallum: they charged me VAT, but apparently no duty. |
| 17:12.57 | mallum | pb_, ah not so bad then |
| 17:13.06 | treke- | Japanese keyboards similar in layout to UK keyboards? |
| 17:13.31 | pb_ | yeah, fairly |
| 17:14.05 | pb_ | lots of extra symbols, but you can mostly ignore those I guess |
| 17:14.12 | kergoth | remap them to something useful |
| 17:14.14 | kergoth | heh |
| 17:14.17 | pb_ | yeah |
| 17:15.25 | kergoth | It indicates the 3d game and streaming image et cetera in smoothly with the ATI corporation make MOBILITYTM RADEON TM 7500 which corresponds to the 3d. |
| 17:15.33 | kergoth | gotta love translating from japanese with google |
| 17:21.43 | pb_ | kergoth: yeah, nico's patch sounds reasonable to me |
| 17:25.19 | kergoth | cool |
| 17:37.46 | treke- | hmm. the bbread code is much more approachable than the bitbake code :p |
| 17:40.26 | kergoth | hehe |
| 17:40.38 | kergoth | simplicity is a Good Thing |
| 17:40.43 | treke- | yeah |
| 17:40.50 | treke- | particularly as a starting point |
| 17:44.39 | *** join/#oe qfh (~qfh@ppp-82-135-0-19.mnet-online.de) |
| 17:44.56 | treke- | i'm thinking I need to tie into bb a bit more than just calling bitbake to do this the way I want to |
| 17:45.12 | kergoth | nod |
| 17:45.23 | kergoth | man |
| 17:45.32 | kergoth | easydns sent the auth code for openzaurus.org to a dead email addy |
| 17:45.36 | kergoth | since tahts whats on the whois |
| 17:45.37 | treke- | suck |
| 17:45.41 | kergoth | but their admin wont let me change teh whois |
| 17:45.45 | kergoth | without paying them to renew |
| 17:45.50 | kergoth | but they already told me i dont need to do that to transfer |
| 17:45.52 | kergoth | dunno whats up |
| 17:45.56 | kergoth | pestering them via email |
| 17:45.58 | treke- | openzaurus.org is being blocked for pornographic content now |
| 17:46.04 | kergoth | hah |
| 17:48.52 | *** join/#oe drw (~drw@c-67-172-218-120.client.comcast.net) |
| 17:51.15 | mickey|sick | hmm |
| 17:52.28 | mickey|sick | any idea for us supporting both prism2 and prism3 wlan cards? the latter needing hostap_fw_load to upload the firmware which we currently accomplish using needs_firmware directive in /etc/network/interfaces. Doesn't look like we can support both non-fw-upload and fw-upload cards in one base image |
| 17:52.48 | mickey|sick | or should we try calling fw-upload in any case? |
| 17:52.55 | kergoth | is there a way to determine whether its necessary? |
| 17:53.02 | kergoth | can we check what chipset it is? |
| 17:53.14 | mickey|sick | no that i know of |
| 17:53.22 | mickey|sick | only by manfid |
| 17:53.28 | kergoth | hmm |
| 17:53.44 | kergoth | can we do that? add that information to the wireless config in /etc/pcmcia/, and have a script yank it out of there? |
| 17:54.14 | kergoth | or better yet, teach /etc/pcmcia/wireless to do it (assuming it gets called before 'net') |
| 17:54.47 | mickey|sick | right. having the wireless script triggering the fw_upload sounds good |
| 17:55.06 | kergoth | thats really where it belongs |
| 17:55.16 | mickey|sick | yep. at least until hostap supports hotplug firmware download |
| 17:55.20 | mickey|sick | which it doesn't yet |
| 17:55.25 | kergoth | right |
| 17:55.37 | mickey|sick | ok, i'll have a look at that. thanks for the pointer |
| 17:56.14 | kergoth | np |
| 17:56.53 | kergoth | easydns fixed the problem with not being able to modify the whois info |
| 17:56.59 | kergoth | should be able to get it xferred shortly |
| 17:56.59 | treke- | oooh. I like the glob module |
| 17:57.02 | kergoth | hehe |
| 17:57.06 | kergoth | glob is nice |
| 17:57.41 | kergoth | :) |
| 17:57.44 | mickey|sick | *cough* |
| 17:58.20 | kergoth | the biggest problem in the bb core is the fact that i never really picked either object oriented or not |
| 17:58.25 | kergoth | so its this ungodly combination of the two |
| 17:58.32 | treke- | this is looking ugly. :) |
| 17:58.45 | treke- | I'm probably going to end up giving each bb its own DL_DIR |
| 17:58.51 | kergoth | classes with all static methods, so it isnt like an object at all, just a module.. etc |
| 17:58.54 | kergoth | heh |
| 17:58.57 | treke- | so I can tell what file came from where |
| 17:59.07 | kergoth | heh, thatd work |
| 17:59.23 | treke- | since I'm guessing bb can't give me that info :p |
| 17:59.38 | kergoth | probably cant, as is |
| 17:59.44 | kergoth | would have to hack it |
| 18:00.00 | treke- | yeah. I want to avoid touching the bitbake code directly |
| 18:00.09 | *** join/#oe [g2] (~g2@rdu57-8-035.nc.rr.com) |
| 18:00.18 | kergoth | hehe |
| 18:00.40 | treke- | hmmm actually I might have a better way of doing this |
| 18:00.54 | treke- | I could just create a "pending-delete" mirror |
| 18:00.56 | *** join/#oe chouimat (~dieu@r2351064.cidc.net) |
| 18:01.03 | treke- | move anything over 30 days into that |
| 18:01.17 | treke- | then if the file is still needed it will be refetched from pending delete |
| 18:01.36 | treke- | and once the run is done the contents of pending-delete can be wiped |
| 18:07.01 | *** part/#oe tigrux (~tigrux@charybdis-ext.suse.de) |
| 18:11.03 | treke- | mickey|sick: Do you know what license the hvsc, manufacturers, and simpad-utilities packages should fall under? |
| 18:11.21 | mickey|sick | yeah |
| 18:11.29 | mickey|sick | simpad-utilities is GPL |
| 18:11.32 | mickey|sick | let me check for the others |
| 18:11.34 | treke- | cool. Those are on the list of un LICENSED packages still |
| 18:11.50 | mickey|sick | hvsc is public domain |
| 18:11.54 | treke- | cool |
| 18:12.27 | mickey|sick | manufacturers as well |
| 18:12.34 | treke- | cool. four more down |
| 18:12.46 | zecke|assignment | treke-: how many apps are missing? |
| 18:12.50 | treke- | 6 |
| 18:12.58 | kergoth | nice |
| 18:13.14 | kergoth | i think we need a new anonymous python function in a bbclass thats always inherited |
| 18:13.22 | kergoth | which does something like the oelint bbclass |
| 18:13.32 | treke- | http://treke.net/~ggilbert/nolicense.txt |
| 18:13.52 | kergoth | i.e. raises an error when the .bb has no LICENSE specified |
| 18:13.56 | kergoth | or warning, or something |
| 18:14.10 | kergoth | since i dont really trust people submitting new bb's to actually lint them |
| 18:14.12 | kergoth | heh |
| 18:14.13 | treke- | the firmware will probably just be ${PN} licenses, but I need to look |
| 18:14.20 | treke- | the fonts I dunno about |
| 18:14.33 | mickey|sick | well, the qpf-fonts are released under QPL/GPL |
| 18:14.41 | mickey|sick | as they're in the same package as qte |
| 18:14.42 | treke- | gah who checked ckermit back in? |
| 18:14.55 | mickey|sick | schurig. this time explaining the license situation in the .bb :) |
| 18:15.20 | treke- | well we don't patch it any more, so we are ok |
| 18:19.19 | treke- | ok, so we're down to the firmware, schedstat-utils, unixbench, and gpe-othello |
| 18:19.32 | CIA-5 | 03ggilbert 07 * r1.2975 10openembedded/packages/ (7 files in 4 dirs): |
| 18:19.32 | CIA-5 | Many files: |
| 18:19.32 | CIA-5 | <PROTECTED> |
| 18:19.43 | mickey|sick | firmware is everything closed, so PN |
| 18:19.52 | mickey|sick | or do we have a special case for that? |
| 18:20.03 | treke- | yeah. I hadn't even looked at them, so they were left over by default |
| 18:20.18 | treke- | unixbench, othello, and schedstat are the only ones I looked for and didn't see a license |
| 18:22.43 | *** join/#oe Luke-Jr (~luke-jr@207.192.219.246) |
| 18:24.45 | florian_kc | bbl |
| 18:35.28 | treke- | gah |
| 18:35.33 | treke- | this is going to be trouble |
| 18:35.44 | *** join/#oe micropal_ (ident@139.80-202-82.nextgentel.com) |
| 18:36.05 | treke- | packages like avr-libc skip on incompatible packages |
| 18:36.26 | treke- | maybe I'll code in a way to specify some variables on a per package basis |
| 18:37.46 | mickey|sick | hmm i still don't get how networks are handled on our distribution |
| 18:37.53 | mickey|sick | isn't there some overlapping functionality |
| 18:38.06 | mickey|sick | pcmcia-cs vs. ifupdown |
| 18:38.16 | pb_ | yeah, there is some overlap. |
| 18:38.24 | treke- | welcome to linux :p |
| 18:38.27 | mickey|sick | can we settle on one? |
| 18:38.31 | pb_ | we prefer ifupdown, because it works transparently with non-pcmcia (i.e. usb) devices. |
| 18:38.50 | mickey|sick | i see |
| 18:38.59 | mickey|sick | is wireless called at all nowadays? |
| 18:39.01 | pb_ | however, if you delete the /etc/network/interfaces entry for a pcmcia device, the system will fall back to using the pcmcia configuration system instead. |
| 18:39.02 | mickey|sick | (pcmcia/wireless) |
| 18:39.18 | pb_ | mickey|sick: not for devices that are mentioned in /etc/network/interfaces, no |
| 18:39.20 | zecke|assignment | ljp: http://www.trolltech.com/company/careers/brisbane.html#position2 |
| 18:39.23 | mickey|sick | pb_: ah |
| 18:39.39 | zecke|assignment | ljp: You seem to need that one as baldy Toll Collect... |
| 18:40.16 | mickey|sick | pb: then i need to find out how to do things manfid dependent in ifupdown |
| 18:40.30 | mickey|sick | because otherwise we can't support all cards out of the box |
| 18:40.58 | pb_ | mickey|sick: ah. which cards need that? |
| 18:41.13 | mickey|sick | all modern prism CF cards need firmware upload |
| 18:41.28 | mickey|sick | prims3 |
| 18:41.40 | mickey|sick | they have RAM instead of flash |
| 18:41.45 | pb_ | oh, right. |
| 18:41.51 | mickey|sick | so need upload on pre-ifup |
| 18:41.59 | pb_ | fwiw, /etc/pcmcia/network _is_ still called, even if ifupdown is in use, so maybe you could do your stuff in there. |
| 18:42.15 | mickey|sick | ah that would be an option right |
| 18:42.24 | mickey|sick | i guess that script has access to the manfid bits |
| 18:43.23 | mickey|sick | another ugly thing is that we need to upload firmware again after resume |
| 18:43.30 | mickey|sick | so either ifdown in suspend.d |
| 18:43.36 | mickey|sick | or something else |
| 18:43.41 | mickey|sick | which i don't know |
| 18:43.50 | treke- | how does redhat/debian/ubuntu do it? |
| 18:44.05 | mickey|sick | they don't |
| 18:44.24 | mickey|sick | they need manual tweakings |
| 18:44.44 | pb_ | yeah |
| 18:45.00 | pb_ | I'm tempted to just add Erik Hovland's scripts to ifdown everything on suspend, and ifup them again on resume. |
| 18:45.22 | mickey|sick | i second that |
| 18:45.25 | pb_ | there are an awful lot of devices that don't work properly unless you do a down/up cycle around suspend, and I don't think it would really hurt anything. |
| 18:45.36 | mickey|sick | lets do it and see who complains :) |
| 18:45.48 | pb_ | heh, good plan |
| 18:45.55 | pb_ | I'll add those bits when I get home. |
| 18:45.57 | mickey|sick | cool |
| 18:47.55 | treke- | sweeeeet |
| 18:48.01 | treke- | bb is doing what I told it to do :p |
| 18:48.10 | koen | eat dirt and die? |
| 18:48.23 | koen | or something more benevolent? |
| 18:48.26 | mickey|sick | heresy! |
| 18:48.39 | treke- | hehe no. fetch. This is the first time I've been able to actually make the bb python modules do something useful |
| 18:49.14 | treke- | usually I just get lost in the pile of code and give up |
| 18:51.12 | kergoth | treke-: heh, re: skipping, that really should never have been part of the parse |
| 18:51.12 | kergoth | it shouldve been a second step after the parse |
| 18:51.12 | kergoth | "alright, can i build you?" |
| 18:52.22 | treke- | hehe yeah |
| 18:52.28 | treke- | something to fix in ng |
| 18:52.35 | kergoth | yep |
| 18:52.41 | treke- | kergoth: for now I'll just let the DB say "pretend you're x86 for now" |
| 18:52.49 | kergoth | all you need is more responses when you ask a .bb to build |
| 18:52.53 | kergoth | instead of only success and failure |
| 18:52.57 | kergoth | we need "nope, cant do" |
| 18:53.03 | kergoth | and "nothing to do" |
| 18:53.07 | treke- | I need something similar for overrid handling anyways |
| 18:53.11 | kergoth | nod |
| 18:53.29 | treke- | I'm just happy to see it fetchin |
| 18:55.29 | ljp | zecke: thats ed's position. he is trying to become perm resident. they need to show no other australian can do that |
| 18:55.56 | treke- | so ed is fucked if an australian applies :) |
| 18:56.17 | ljp | i dont know many people that know qt test |
| 18:56.38 | ljp | naaa. look at the requirements |
| 18:56.55 | ljp | Excellent oral and written communication skills, preferably including written Norwegian and/or German. |
| 18:57.14 | ljp | how many qa people know norwegian AND german? |
| 18:57.42 | treke- | and who have experience with testing tools that are proprietary to trolltech? |
| 18:58.12 | ljp | :) |
| 18:58.19 | treke- | should have just added "4+ years working for trolltech" to the requirements list |
| 18:58.29 | treke- | "Must be named Ed" |
| 18:59.51 | *** join/#oe TyggerBob (1000@CPE0000395fb8ba-CM000f9f7c58fa.cpe.net.cable.rogers.com) |
| 19:00.05 | TyggerBob | quick question, so I don't have to go hunting for it. |
| 19:00.29 | kergoth | quick answer, though it might be rtfm |
| 19:00.31 | kergoth | :) |
| 19:00.39 | TyggerBob | good idea. |
| 19:00.45 | TyggerBob | I'll check there first. |
| 19:01.50 | treke- | kergoth: think it's safe to say that a ${P} will be a unique key? |
| 19:02.16 | TyggerBob | url for the manual again? |
| 19:02.50 | kergoth | there's no OE manual, just the one for bitbake. so whether its useful will depend on the nature of the question |
| 19:02.54 | kergoth | http://bitbake.berlios.de/manual/ |
| 19:03.01 | TyggerBob | that's what I'm looking for |
| 19:03.03 | TyggerBob | Tnaks |
| 19:03.06 | TyggerBob | thanks, even |
| 19:03.25 | kergoth | np |
| 19:04.38 | treke- | I guess filename is safer |
| 19:04.58 | TyggerBob | nope.. not there.. |
| 19:05.16 | TyggerBob | What's the variable that S is normally set to? |
| 19:05.40 | TyggerBob | ${work something} |
| 19:05.41 | koen | WORKDIR/PN-PV |
| 19:05.45 | koen | iirc |
| 19:06.01 | TyggerBob | hehe.. no hyphen/underscore.. |
| 19:06.07 | TyggerBob | tried every combination of them :) |
| 19:06.23 | TyggerBob | thanks |
| 19:06.37 | koen | PN-PV-PR? |
| 19:09.05 | *** join/#oe oe_newbie (~kvirc@195.234.128.72) |
| 19:18.58 | *** join/#oe gremlin[it] (~gremlin@194-185-220-157.f4.ngi.it) |
| 19:22.01 | gremlin[it] | good evening to all ... |
| 19:22.19 | koen | hey gremlin[it] |
| 19:22.44 | gremlin[it] | at now i don't have trouble ;) ... just compiling and listening for easy questions to answer ;) |
| 19:22.47 | gremlin[it] | hi koen ;) |
| 19:33.12 | *** join/#oe mrz80 (1000@caledonia.cns.ufl.edu) |
| 19:41.31 | *** join/#oe jmau (~jmau@pD9E5DFD8.dip.t-dialin.net) |
| 19:43.51 | *** join/#oe jmau (~jmau@pD9E5DFD8.dip.t-dialin.net) |
| 19:50.20 | gremlin[it] | wb kergoth |
| 19:51.28 | *** join/#oe zedkatuf (~dfutak@82-32-58-45.cable.ubr08.azte.blueyonder.co.uk) |
| 19:52.23 | kergoth | gah |
| 19:52.26 | kergoth | <PROTECTED> |
| 19:55.48 | CosmicPenguin | Did you register it with a registar that locked the domains? My registrar did that automagically |
| 19:57.11 | kergoth | i have control over that |
| 19:57.14 | kergoth | i unlocked it already |
| 20:02.53 | kergoth | grr |
| 20:03.01 | kergoth | i should probably just give up and renew the fucking thing at easydns |
| 20:03.04 | kergoth | transfer it next year |
| 20:03.09 | kergoth | heh |
| 20:03.57 | gremlin[it] | mhh question about oe ;;; i have this message >>> NOTE: Couldn't find shared library provider for libgcc_s.so.1 <<< is normal ? |
| 20:04.17 | gremlin[it] | when building many packeges |
| 20:04.47 | kergoth | iirc thats an outstanding bug |
| 20:04.57 | kergoth | but its usually harmless since libgcc1 normally gets pulled into the image at some point anyway |
| 20:05.43 | kergoth | that means that it saw that that package requires libgcc_s.so.1, but it couldnt figure out which package includes it, so couldnt add the dependency to the ipk |
| 20:06.05 | *** join/#oe gints_ (~gints@195.2.115.160) |
| 20:07.39 | gremlin[it] | ok ... i suppose that libgcc will be present in final image :) |
| 20:07.52 | *** join/#oe pb_ (~pb@2002:5168:d330:1:20b:6aff:fe40:e27f) |
| 20:10.00 | CIA-5 | 03pb 07 * r1.2976 10openembedded/packages/network-suspend-scripts/ (files/ifupdown network-suspend-scripts.bb): add network-suspend-scripts |
| 20:10.17 | kergoth | If it's IP protection you're after, please use some other operating |
| 20:10.17 | kergoth | system. Thanks. |
| 20:10.18 | kergoth | hehe |
| 20:10.22 | kergoth | go rmk |
| 20:10.57 | pb_ | heh |
| 20:11.18 | pb_ | mickey|sick: that file I just checked in is the Erik Hovland script I mentioned earlier |
| 20:12.38 | CIA-5 | 03pb 07 * r1.2978 10openembedded/packages/meta/task-bootstrap.bb: bump PR on task-bootstrap |
| 20:12.41 | CIA-5 | 03pb 07 * r1.2977 10openembedded/conf/machine/handheld-common.conf: add network-suspend-scripts to handheld-common BOOTSTRAP_EXTRA_{R}DEPENDS |
| 20:16.12 | kergoth | ah, cool |
| 20:17.55 | RP | kergoth: I was just smiling at that :) |
| 20:18.17 | RP | evening all btw |
| 20:18.26 | pb_ | evening rp |
| 20:19.54 | RP | In the kernel, are you supposed to check for NULL pointers before calling the likes of kfree these days? Which is the current convention? |
| 20:23.55 | *** join/#oe T`|Heli (total@GO.HELI.RI.CMU.EDU) |
| 20:35.55 | gremlin[it] | got something in libvsg-cairo |
| 20:36.00 | gremlin[it] | >>> | svg-cairo-internal.h:91: error: parse error before "va_list" |
| 20:36.00 | gremlin[it] | | svg-cairo-internal.h:91: warning: function declaration isn't a prototype |
| 20:36.00 | gremlin[it] | <PROTECTED> |
| 20:39.08 | koen | ah yes |
| 20:39.09 | gremlin[it] | mhhh if i remember right koen or kergoth was talking about yestarday ... |
| 20:39.16 | koen | that was cairo |
| 20:39.25 | gremlin[it] | no with me with someone else |
| 20:39.29 | kergoth | someone reported libsvg-cairo failing |
| 20:39.33 | kergoth | after the cairo bug was fixed |
| 20:39.40 | koen | libcairosvg need a #include <stdargs.h> in some file |
| 20:40.08 | koen | florian knows what file |
| 20:40.11 | gremlin[it] | ok ... i start to check ... |
| 20:40.44 | koen | my OE machine has XP booted right now, so I can't check |
| 20:40.59 | gremlin[it] | ahhhhh blasfemia !!! |
| 20:41.26 | koen | ILWIS doesn't run on mac or linux, so I had too install XP |
| 20:46.05 | gremlin[it] | ILWIS ??? |
| 20:47.00 | koen | Integrated Land and Water Info System |
| 20:47.07 | koen | see itc.nl for details |
| 20:58.38 | keturn | I tried to subscribe to bitbake-dev in case I wanted to send patches, but I don't seem to have received the confirmation mail. |
| 20:58.52 | kergoth | odd |
| 20:59.10 | kergoth | perhaps berlios.de is lagged? how long has it been? |
| 20:59.39 | keturn | hours and hours. |
| 21:01.46 | treke- | kergoth: is there any way to tell the name of the file that the fetcher is going to return before actually fetching? |
| 21:02.20 | kergoth | treke-: the 'localpath' method is used to determine the local path itll be downloading to |
| 21:02.26 | treke- | cool |
| 21:02.31 | kergoth | that's how the unpack function in base.bbclass works |
| 21:02.41 | kergoth | it iterates through the items in SRC_URI, calls localpath on them |
| 21:02.46 | kergoth | goes from there |
| 21:03.28 | treke- | since I'm downloading to a seperate DL_DIR for each package I need to check for previously obtained dls for things like opie :) |
| 21:03.43 | kergoth | ah |
| 21:04.33 | *** join/#oe HuntsMan (~hunts@CM-lflo3-118-148.cm.vtr.net) |
| 21:05.00 | treke- | looks like from the code that an md5sum should only exist for a file that was successfully downloaded |
| 21:05.12 | kergoth | right, thats basically what it is |
| 21:05.16 | kergoth | a file completion stamp |
| 21:05.51 | treke- | this actually seems mostly sane now that I'm getting used to it |
| 21:08.55 | mickey|sick | RP: hi |
| 21:09.03 | mickey|sick | RP: we have to do something w/ backlight |
| 21:09.06 | mickey|sick | it's still not correct |
| 21:10.29 | mickey|sick | it's nonlinear - which probably means we need to have a mapping table. try for yourself, brightness increases until ~20, then sticks about that brightness for until 31, then when you introduce the 5th bit, it actually gets lower than it was @ 20 |
| 21:10.52 | mickey|sick | so we probably need to go until 18 |
| 21:11.05 | mickey|sick | then kick the 5th bit for everything above and subtract 18 |
| 21:11.12 | mickey|sick | which means we end up with only 40 steps or so |
| 21:11.20 | mickey|sick | but each step increasing the actual brightness |
| 21:11.27 | mickey|sick | (which sounds about correct :) |
| 21:12.19 | mickey|sick | this is probably the reason why sharp limited it to 7 steps |
| 21:12.23 | mickey|sick | because they couldn't figure out :D |
| 21:15.25 | *** join/#oe Pigi (~NoOne@host75-70.pool21345.interbusiness.it) |
| 21:15.29 | Pigi | Ciao all |
| 21:18.40 | CosmicPenguin | Is that what they call it these days? |
| 21:20.54 | treke- | hrmm bbread and bitbake seem to work differently |
| 21:21.15 | kergoth | in what way? |
| 21:21.26 | treke- | bb loadin |
| 21:22.18 | treke- | bbread seems to just call bb.make.load_bbfile |
| 21:22.42 | treke- | bitbake seems to do it differently ( haven't figured out where it's done yet) |
| 21:27.34 | treke- | I guess that doesnt mattter now that I know that I need to do data.getVar('FOO',d,1) if I want the expanded value |
| 21:31.05 | gremlin[it] | koen : mhh seem that stdargs.h wasn't present neither in oe cross-environment neither on my local FC3 /usr/include !!! |
| 21:33.22 | *** join/#oe Virusmaster (~Thomas@p5091AAE4.dip.t-dialin.net) |
| 21:33.47 | gremlin[it] | but info libc (appendix 2) refet to stdarg.h to use variable function argiments ... :( :( :( |
| 21:37.35 | gremlin[it] | koen : i'm tired ... if noone patch till tomorrow i'll try tomorrow evening ... bye bye |
| 21:38.25 | koen | later gremlin[it] |
| 21:38.40 | CIA-5 | 03mickeyl 07 * r1.2973.1.1 10openembedded/packages/pcmcia-cs/ (pcmcia-cs_3.2.8.bb pcmcia-cs-3.2.8/pcic-extra.patch): repair pcmcia init script (was looking for /sbin/pidof instead of /bin/pidof) |
| 21:46.20 | koen | 'night all |
| 21:47.50 | Pigi | pb_ there ? |
| 21:48.24 | *** part/#oe jmau (~jmau@pD9E5DFD8.dip.t-dialin.net) |
| 21:52.10 | kergoth | heh |
| 21:52.14 | kergoth | the gentoo guys have a configure cache now |
| 21:52.15 | kergoth | http://dev.gentoo.org/~stuart/confcache/confcache-portage-2.0.50-r10.patch |
| 21:52.20 | treke- | yeah saw that |
| 21:52.42 | kergoth | that's cool. particularly since we were just talking about that last week |
| 21:52.44 | kergoth | heh |
| 21:53.54 | kergoth | hehe |
| 21:54.07 | treke- | it's breaking my script and we don't want the files anyways |
| 21:54.43 | treke- | is localpath supposed to return things like patch parameters as part of the filename? |
| 21:54.45 | kergoth | technically you should not mirror anything that starts with file://, except for file:/// |
| 21:54.57 | kergoth | (relative is searched in filespath. absolute, if they exist, should probably be mirrored) |
| 21:55.09 | kergoth | yep |
| 21:55.13 | kergoth | because do_patch uses it |
| 21:55.20 | treke- | k. so I'll have to strip them off by hand |
| 21:55.23 | kergoth | guess it could swipe that from SRC_URI |
| 21:55.25 | kergoth | what we normally do is |
| 21:55.52 | kergoth | use decodeurl |
| 21:55.59 | kergoth | then call encodeurl, without the 'parm' option |
| 21:56.08 | kergoth | or just regex the damn thing |
| 21:56.09 | kergoth | heh |
| 21:56.10 | treke- | ah |
| 21:56.11 | kergoth | whichi s probably easier |
| 21:56.13 | treke- | I can do that |
| 21:56.29 | treke- | I think it's safe to say we won't have any semicolons in valid filenames |
| 21:56.51 | kergoth | having the paramteres in localpath means the parameters are in the A variable |
| 21:56.54 | treke- | god knows what all would be broken if a file actually did have one |
| 21:57.02 | kergoth | which means you can write a do_patch or do_unpack in shell |
| 21:57.14 | kergoth | python doesnt need it, since itll iterate over SRC_URI and call localpath directly anyway |
| 21:57.17 | treke- | yeah that's fine. |
| 21:57.41 | treke- | I'm sure most of my code here isn't following the original intent of bb :p |
| 21:58.16 | treke- | at the moment I'm still just executing the do_fetch task for downloading. looked like it would be the easiest route |
| 21:59.31 | *** join/#oe keith_ (~keith@CPE-61-9-204-1.nsw.bigpond.net.au) |
| 21:59.39 | treke- | what I do is load the bb, override DL_DIR, iterate over SRC_URI and copy each already existing uri from pool to DL_DIR, and then execute do_fetch |
| 22:00.15 | treke- | then I'll copy completed items from DL_DIR into pool and move onto the next item |
| 22:01.19 | treke- | and then I'll move onto the fancy bits |
| 22:01.35 | treke- | once I have the equivelant of the old script working |
| 22:05.02 | kergoth | cool, sounds sane enough |
| 22:05.22 | treke- | yeah. Sounds sane so far |
| 22:05.49 | kergoth | random aside: inkscape works great for presentation slides |
| 22:05.53 | treke- | oh? |
| 22:06.02 | kergoth | creating them anyway :) |
| 22:06.14 | kergoth | i need a small svg viewer to actually move through them or something |
| 22:06.26 | kergoth | keith packard and carl worth used xsvg for their presentations at ols |
| 22:06.31 | treke- | why not just use powerpoint? :p |
| 22:06.32 | kergoth | will have to try that, see if itll do |
| 22:06.38 | kergoth | because its powerpoint |
| 22:06.44 | kergoth | besides, i like inkscape, and svg |
| 22:06.47 | treke- | hehe |
| 22:06.51 | treke- | understandable |
| 22:07.20 | kergoth | i dont think i ever have :) |
| 22:07.24 | kergoth | and i'd like to keep it that way |
| 22:07.29 | treke- | good man |
| 22:07.37 | treke- | I really hate powerpoint presentations |
| 22:08.25 | treke- | but that's less about power point and more about people |
| 22:08.45 | kergoth | hehe |
| 22:09.29 | zecke | kergoth: SMIL is kewl as well |
| 22:09.34 | kergoth | smil? |
| 22:09.45 | kergoth | hm, wonder how to emit this set of svgs as a pdf |
| 22:09.53 | zecke | kergoth: http://www.cwi.nl/projects/Ambulant/distPlayer.html |
| 22:10.04 | zecke | it is a w3c standard (xml...) |
| 22:10.25 | kergoth | ah |
| 22:11.21 | treke- | hrmm |
| 22:11.54 | treke- | kergoth: the upside is, after the code I did today, doing an autobuilder really wouldnt be that hard |
| 22:12.12 | kergoth | zecke: probably overkill |
| 22:12.21 | kergoth | zecke: i dont give a shit about screen transitions and animations and crap :) |
| 22:12.26 | kergoth | will check it out though |
| 22:12.35 | zecke | kergoth: I don't get its purpose yet as well |
| 22:12.48 | kergoth | it looks like they're trying to replace flash |
| 22:12.51 | kergoth | more than anything else |
| 22:14.13 | kergoth | hmm, i still havent gotten around to getting the wrt54gs kernel happy in oe |
| 22:14.17 | kergoth | forgot about it |
| 22:14.53 | CIA-5 | 03mickeyl 07 * r1.2980 10openembedded/packages/prism3-support/prism3-support_0.1.bb: add prism3-support (meta) package w/ ifupdown script triggering hostap_fw_load in case it finds a supported card. Add more known manfids as you like... |
| 22:15.09 | kergoth | mickey|sick: got it working? |
| 22:16.26 | mickey|sick | yeah, since pb_ reminded me that wireless doesn't get called at all when the iface is listed in networks, i just added the necessary bits to a minimal script which goes into /etc/network/if-pre-up.d/ that gets called before all others |
| 22:16.34 | kergoth | ahh right |
| 22:16.37 | kergoth | cool |
| 22:17.15 | kergoth | i still like the idea of storing the info about whther it needs a firmware download int he pcmcia .conf, but.. |
| 22:17.40 | mickey|sick | yep - a dedicated directive for that would be nice |
| 22:17.53 | kergoth | one place to maintain that info instead of more than one |
| 22:17.57 | kergoth | keep the maintainance load down |
| 22:17.58 | kergoth | ah well |
| 22:29.32 | RP | mickey|sick: hello |
| 22:29.48 | RP | mickey|sick: Just read the comments about the backlight driver. It was just accepted upstream :-/ |
| 22:30.18 | mickey|sick | heh |
| 22:30.19 | mickey|sick | well, bugs happen |
| 22:30.19 | mickey|sick | or rather |
| 22:30.20 | mickey|sick | improvements :D |
| 22:30.22 | RP | I'll add it to the list of things to look at :) |
| 22:30.28 | mickey|sick | ok |
| 22:30.49 | RP | I've just sent a bugfix for w100 upstream. The fbdev people will be getting sick of me :)... |
| 22:30.59 | mickey|sick | hah |
| 22:31.17 | mickey|sick | i think they will be glad that someone actually submits stuff |
| 22:31.24 | mickey|sick | doesn't happen too often |
| 22:31.30 | mickey|sick | :) |
| 22:31.44 | RP | :) |
| 22:31.52 | RP | I'm trying my best to get this all working... |
| 22:32.02 | RP | My next challenge is the input system people... |
| 22:32.03 | mickey|sick | of course. and you're doing a helluva job! need to go sleeping, c'ya |
| 22:32.13 | RP | 'night! |
| 22:36.48 | *** join/#oe moa_ (~moa@paris.bluebugs.org) |
| 22:37.48 | zecke | mickey|zzZZzz: sleep well |
| 22:38.06 | *** join/#oe hufnus (~slonsiki@63.93.79.84) |
| 22:41.18 | *** join/#oe oggers (~ogger@57.Red-213-96-148.pooles.rima-tde.net) |
| 22:49.46 | *** join/#oe Marsupilami24 (~Marsupila@152.139-60-66-nokia-static.surewest.net) |
| 22:53.44 | mallum | how do I stop union bootstrap getting used over normal bootstrap ? |
| 22:55.10 | *** join/#oe darkschneider (~gab@81-208-36-80.fastres.net) |
| 22:56.17 | kergoth | mallum: bk pull. Cwiiis pushed the fix to lower its default preference. |
| 22:57.01 | mallum | kergoth, it pref is -1 in what I have |
| 22:57.09 | mallum | kergoth, bit its still getting pulled in |
| 22:57.19 | mallum | kergoth, should I clean it ? |
| 22:58.04 | kergoth | if its pref is -1, it wont be used |
| 22:58.08 | kergoth | unless task-bootstrap isnt buildable. |
| 22:58.23 | kergoth | that is, it wont be _built_ |
| 22:58.33 | kergoth | i dont know how its handled, ipk wise |
| 22:58.47 | kergoth | perhaps if the other ipk is sitting in tmp/deploy/ipks, it could interfere with rootfs population |
| 22:58.51 | kergoth | dunno how Cwiiis did it, havent looked |
| 22:59.07 | mallum | kergoth, okey I've cleaned it, will check no rouge ipk either |
| 23:06.05 | *** join/#oe keith (~keith@bh02i525f01.au.ibm.com) |
| 23:08.07 | Pigi | nite all |
| 23:08.56 | pb_ | kergoth: cwiiis and I were discussing that earlier. he made the unionfs thing be a different (newer) _version_ of task-bootstrap, not a different PN. |
| 23:09.35 | pb_ | so, if you have the unionfs version in your deploy/ipk directory, ipkg will always choose that one during image construction, irrespective of DEFAULT_PREFERENCE or anything else. |
| 23:10.30 | kergoth | ahhh |
| 23:10.33 | kergoth | ugly |
| 23:10.48 | pb_ | yeah |
| 23:11.09 | pb_ | I didn't manage to find out exactly why it was done that way, yet |
| 23:38.48 | *** join/#oe mithro (~tim@202.191.111.52) |
| 23:39.29 | zecke | isn't it time to remove iqnotes from meta-opie? |
| 23:39.53 | kergoth | heh, imho nothing that isnt in opie cvs belongs in meta-opie |
| 23:40.00 | kergoth | can create a differnet metapackage for all the qtopia+opie apps |
| 23:40.19 | zecke | kergoth: the issue is the source is not any more the and the app became proprietary |
| 23:40.48 | kergoth | ah |
| 23:40.54 | kergoth | so our version is now maintainerless? |
| 23:41.04 | kergoth | 2.0.4's sources are available |
| 23:41.10 | kergoth | i was talking to someone in #zaurus about it yesterday |
| 23:41.24 | zecke | ah ok then we need to update SRC_URI |
| 23:41.41 | zecke | it could be also that he's behind a broken firewall (not ecn compatible...) |
| 23:42.24 | kergoth | its in a different location |
| 23:42.29 | kergoth | i'll add the new version |
| 23:42.54 | zecke | I'll go to bed now |
| 23:42.57 | kergoth | night |
| 23:48.08 | *** join/#oe zecke (~ich@83.169.146.40) |
| 23:49.51 | *** join/#oe AndyCap (~aoy@host-81-191-64-77.bluecom.no) |
| 23:49.54 | *** join/#oe HuntsMan (~hunts@CM-lflo3-118-148.cm.vtr.net) |
| 23:49.58 | *** join/#oe webmind (~random@217-195-236-172.dsl.esined.net) [NETSPLIT VICTIM] |
| 23:50.47 | *** join/#oe J (~john@cpc4-mfld2-5-1-cust63.nott.cable.ntl.com) [NETSPLIT VICTIM] |
| 23:52.21 | *** join/#oe micropal (ident@139.80-202-82.nextgentel.com) |
| 23:54.56 | *** join/#oe Gnea (gnea@garson.org) |
| 23:55.41 | *** join/#oe robtaylor_ (~robtaylor@217.204.121.82) [NETSPLIT VICTIM] |
| 23:55.41 | *** join/#oe Cwiiis (~cwiiis@user-214-207-151-83.e7even.com) |
| 23:55.41 | *** join/#oe msw (~msw@rdu-nat.specifix.com) [NETSPLIT VICTIM] |
| 23:55.41 | *** join/#oe benpp (~ben@m144.net81-64-227.noos.fr) [NETSPLIT VICTIM] |
| 23:55.42 | *** join/#oe bronson (~bronson@node-40240852.sjc.onnet.us.uu.net) [NETSPLIT VICTIM] |