[00:37] esdentem (n=esdentem@ppp-82-135-72-172.mnet-online.de) joined #rocklinux.
[03:01] esdentem (n=esdentem@ppp-82-135-72-172.mnet-online.de) left irc: Read error: 110 (Connection timed out)
[03:03] kasc_ (n=kasc@dslb-084-060-104-105.pools.arcor-ip.net) joined #rocklinux.
[03:09] kasc (n=kasc@dslb-084-060-096-243.pools.arcor-ip.net) left irc: Read error: 104 (Connection reset by peer)
[03:09] Nick change: kasc_ -> kasc
[04:22] esden_ (i=esdentem@pallas.crash-override.net) joined #rocklinux.
[04:22] esden (i=esdentem@pallas.crash-override.net) left irc: Read error: 104 (Connection reset by peer)
[04:22] th__ (n=th@bozeman.hbsn.de) joined #rocklinux.
[04:22] _toberkel (n=toberkel@toberkel.net) joined #rocklinux.
[04:23] toberkel (n=toberkel@toberkel.net) left irc: Read error: 104 (Connection reset by peer)
[04:24] netrunner (n=andreas@anvame.net) left irc: Read error: 104 (Connection reset by peer)
[04:36] th (n=th@bozeman.hbsn.de) left irc: Read error: 110 (Connection timed out)
[04:37] netrunner (n=andreas@anvame.net) joined #rocklinux.
[06:44] <blindcoder> moin
[07:29] blindcoder (n=blindcod@tor/regular/blindcoder) left irc: Remote closed the connection
[07:34] blindcoder (i=blindcod@tor/session/direct/x-f091a47cec9cdb67) joined #rocklinux.
[08:44] <blindcoder> th__: so, what's the current state? think we got everything covered for a release?
[09:07] Nick change: th__ -> th
[09:08] <th> blindcoder: let's wait a vanilla build after current journals. then we can do a RC
[09:09] <blindcoder> \o/
[09:10] <blindcoder> what's your current journal?
[09:11] <blindcoder> more or less important bugfixes (2006-06-02) Rev.6?
[09:11] <th> $topic or better the rev6-test in same directory it is
[09:11] <th> 20060608-rev6-test_cd1.iso
[09:11] <th> needs testing
[09:13] <blindcoder> ah, so you're not compiling right now?
[09:14] <blindcoder> well, I'll probably be leaving early today
[09:14] <blindcoder> so I could install it
[09:15] <blindcoder> btw: 2006060118383613146 is absolutely necessary, because the current version in trunk won't compile
[09:15] <blindcoder> so is 200606031900521030
[09:15] <blindcoder> 9
[09:15] <blindcoder> 2006060319005210309
[09:17] <th> 2006060118383613146 and 2006060319005210309?
[09:17] <th> ok. noted.
[09:17] <th> strange
[09:17] <blindcoder> pcmciautils and nvidia respectively
[09:17] <th> trunk _did_ compile w/o them for me
[09:18] <blindcoder> nvidia will build, but when you then try to compile the module, it will fail
[09:18] <blindcoder> and pcmciautils is not base IIRC
[09:18] <th> ok
[09:18] <blindcoder> or is it?
[09:18] <blindcoder> it is
[09:18] <blindcoder> hmm
[09:19] <blindcoder> well, nvidia is necessary, pcmciautils nice :)
[09:19] <th> Finished building pcmciautils=pcmciautils.
[09:19] <th> Applying package split: 42 pcmciautils:dev /pkgconfig/
[09:19] <th> Applying package split: 60 pcmciautils:doc /share/doc/
[09:19] <th> uhhh
[09:19] <blindcoder> oops :)
[09:20] <blindcoder> set a message for it, I'll take care of it
[09:20] <th> pcmciautils: usr/usr/share/man/man8/pccardctl.8
[09:20] <th> i guess 2006060118383613146 could fix this one
[09:20] <th> +[O] prefix=/
[09:20] <th> pcmciautils: usr/etc/pcmcia/config.opts
[09:20] <th> insane
[09:20] <th> :doc is ok
[09:21] <th> and there is no :dev
[09:21] <blindcoder> no, because it's just the utils package
[09:21] <blindcoder> lspcmcia, pccardctl and so on
[09:21] <blindcoder> there's nothing to :dev herle
[09:22] <th> jupp
[10:18] blindcoder (i=blindcod@tor/regular/blindcoder) left irc: Remote closed the connection
[10:19] blindcoder (i=blindcod@tor/session/direct/x-e0e1c89521977e8b) joined #rocklinux.
[10:51] <blindcoder> okay, joystick stone module and init script look good
[11:02] <blindcoder> th: oh, and if at all possible, I have a weakness for 2006060911022927386 :-)
[11:03] <blindcoder> th: it _is_ a personal package :-)
[11:03] ija_ (n=ija@ joined #rocklinux.
[11:09] ija (n=ija@ left irc: Read error: 104 (Connection reset by peer)
[13:03] <stf^rocklinux> hi all
[13:04] <stf^rocklinux> blindcoder: in the pcmciautils update, please set prefix in the .conf file.
[13:32] ija_ (n=ija@ left irc: "Lost terminal"
[13:34] <th> blindcoder: consider this one as done.
[13:42] <blindcoder> th: \o/
[14:34] <stf^rocklinux> btw. there's an article about our mission statement in LWN.net Weekly Edition for June 8, 2006 :)
[14:35] <stf^rocklinux> but this edition will be available for free next week only, so I can't read it right now
[14:36] <blindcoder> wow...
[15:12] <SMP> rotfl
[15:12] Action: SMP just returned from "Operating Systems" written exam
[15:12] <SMP> was scheduled to take 45 min. -- I left after 15
[15:13] <stf^rocklinux> too easy?
[15:13] <stf^rocklinux> or too difficult?
[15:14] <SMP> easy. multiple choice test covering the facts of 35 slides
[15:15] <blindcoder> heh
[15:16] <SMP> furthermore it was designed such that you could have absolutely no clue at all, check every possible answer and still pass ;->
[15:21] stf^rocklinux (n=user@heim-032-98.raab-heim.uni-linz.ac.at) left irc: Remote closed the connection
[15:22] stf^rocklinux (n=user@heim-032-98.raab-heim.uni-linz.ac.at) joined #rocklinux.
[15:23] <blindcoder> heh, great
[15:23] <blindcoder> easy grades :)
[15:37] <stf^rocklinux> blindcoder: I've tested 2006060800311910241 now and it happens to work
[15:38] <blindcoder> hmm
[15:38] <stf^rocklinux> readlink -f already returns absolute paths :)
[15:40] <blindcoder> ah
[15:40] <blindcoder> cleared vote :)
[15:40] <stf^rocklinux> thx
[15:49] <SMP> what. the. fuck.
[15:49] <SMP> what's that messing with /etc/conf/hardware supposed to do? I thought ti's clear that udev and only udev does all that work
[15:50] <stf^rocklinux> currently udev doesn't load OSS modules, that's were hwscan is handy 
[15:51] <stf^rocklinux> and hwscan is a better default than hotplug
[15:52] <SMP> gnnn
[15:53] <SMP> OSS modules need to be handled from modprobe.conf
[15:54] <SMP> so they're loaded when snd is loaded
[15:55] <SMP> or maybe rather when sound-slot-0 is requested
[16:02] <stf^rocklinux> ok, but that would require changing modprobe.conf at install time.
[16:02] <SMP> why?
[16:02] <stf^rocklinux> because the OSS modules are aliased to hardware-specific modules
[16:03] <stf^rocklinux> E.g: #alias snd-card-0		snd-trident
[16:03] <stf^rocklinux> maybe we should alias the OSS modules to e.g. the snd module?
[16:04] <SMP> er, wait
[16:05] <SMP> are we talking about pure OSS or rather ALSA's OSS emulation layer (snd-oss-pcm, snd-oss-seq)?
[16:05] <stf^rocklinux> OSS emu. I know for sure that with the default modprobe.conf in ROCK, OSS modules are not loaded...
[16:06] <SMP> then we have these options:
[16:07] <SMP> in modprobe.conf: install snd /sbin/modprobe --ignore-install snd && { /sbin/modprobe snd-pcm-oss; /sbin/modprobe snd-mixer-oss; /sbin/modprobe snd-seq-oss; /bin/true }
[16:08] <SMP> which does not work when these is no snd module because ALSA is linked in statically
[16:08] <blindcoder> we could just put modprobe snd-seq-oss into /etc/conf/kernel
[16:08] <blindcoder> which is sourced by /etc/rc.d/init.d/system
[16:09] <SMP> what we can't do is auto-loading of modules when a process tries to open the device, because with udev the device simply won't be there
[16:09] <SMP> blindcoder: that's a non-solution
[16:10] <SMP> another (probably the best option) would be to trigger loading of the OSS emulation from a udev rule
[16:11] <SMP> i.e. one that matches some ALSA device appearing
[16:13] <stf^rocklinux> I favor the udev rule, maybe I can figure out how to write it...
[16:15] <stf^rocklinux> but you forgot an important option: PANIC! ^^
[16:15] <SMP> stf^rocklinux: that'd be great. maybe this can help you: http://reactivated.net/writing_udev_rules.html -- but be careful, maybe it's outdated
[16:16] <stf^rocklinux> thx
[16:16] <SMP> ah, last revision June, 3. that should be ok
[16:17] <stf^rocklinux> yeah :)
[16:44] <th> back
[16:52] [raphael] (n=raphael@raphael.netpark.at) joined #rocklinux.
[16:53] <[raphael]> hi there
[16:53] <[raphael]> I guess /me has a bug report
[16:53] <blindcoder> shoot
[16:53] <[raphael]> I'm using a generic ROCK linux target with "minimal" package selection
[16:54] <[raphael]> and I am using bash3 instead of bash2 (under bash options)
[16:54] <[raphael]> AFAIK these are the only things I "changed" from the default
[16:54] <[raphael]> so....
[16:54] <[raphael]> build went fine until going to 3rd stage (if I correctly interpret it)
[16:54] <[raphael]> and it simply stopped with the error message:
[16:54] <[raphael]> chroot: cannot run command `bin/bash': No such file or directory
[16:55] <blindcoder> bin/bash should becomea symlink to bash3
[16:55] <[raphael]> so, I assume the bug is somewhat like this: if bash3 is selected as default /bin/sh shell it is still not correctly queued into the build process
[16:55] <[raphael]> because, I checked, it's not there
[16:55] <[raphael]> oh... does the build process try to use the "systems" /bin/bash?
[16:56] <[raphael]> no... because that is present
[16:56] <blindcoder> not in stage 3
[16:56] <[raphael]> well, the thing is that it tries to chroot before the bash3 package is built
[16:56] <blindcoder> hmm
[16:56] <blindcoder> let's see
[16:56] <[raphael]> I can send you the config file + buildlog
[16:56] <blindcoder> the priorities are equal
[16:56] <blindcoder> so it should be build in stage 1
[16:57] <blindcoder> upload it somewhere and send a mail to the mailinglist
[16:57] <[raphael]> By default "bash2" was selected
[16:57] <blindcoder> I don't know how much longer I'll be online today
[16:57] <[raphael]> I'm not registered at the mailinglist, so I'm not sure if it gets through (depends on how carefully the mod requests are looked through)
[16:58] <[raphael]> (your answer is probably: then get registered ;)
[16:58] <blindcoder> then register :)
[16:58] <[raphael]> heh
[16:59] <[raphael]> but... just a question
[16:59] <[raphael]> if I change to bash2 now and rebuild the target, does it simply continue with the already existing packages?
[16:59] <[raphael]> and probably build bash2 before doing a chroot?
[16:59] <stf^rocklinux> raphael: yes
[17:00] <[raphael]> ok, so I'll just try that (bash2 is just as fine, and probably smaller, which is good for the target system)
[17:00] <[raphael]> I'll let you know - because if that works then the bug is really because of the bash selection
[17:00] <[raphael]> (or can be triggered with that)...
[17:01] <stf^rocklinux> thx, it would be a new bug then because I've already done builds with bash3 as default :S
[17:01] <stf^rocklinux> with bash 3.0, to be precise. IIRC some packages' configure didn't like bash 3.1
[17:02] <[raphael]> stf^rocklinux: well, I also have the "minimal" pkg preselection
[17:02] <[raphael]> stf^rocklinux: maybe that's an issue
[17:03] <[raphael]> we'll find out
[17:03] <stf^rocklinux> ahh, I think the 'minimal' only enables bash2
[17:03] <[raphael]> ah... I'll have a look if any bash is selected at all now
[17:04] <[raphael]> hmm... bash2 is selected, but I configured bash3 as default
[17:04] <[raphael]> bash3 is not in the package selection
[17:05] <stf^rocklinux> yep, 'minimal' only enables bash2, use e.g. the custom package selection to correct that
[17:05] <stf^rocklinux> O bash2
[17:05] <stf^rocklinux> X bash3
[17:05] <stf^rocklinux> :)
[17:05] <[raphael]> yes, I'll try exactly this
[17:05] <[raphael]> can this be automated by the option?
[17:06] <[raphael]> or at least the option should be disabled if bash3 isn't even available
[17:06] <stf^rocklinux> I guess bash3 should be automatically enabled if it's the default sh
[17:07] <[raphael]> its not enabled
[17:07] <[raphael]> I selected it as default but it didn't get enabled (packages.txt)
[17:08] <stf^rocklinux> yes, I mean it should be enabled, but it isn't. I'll try to fix this.
[17:08] <[raphael]> stf^rocklinux: ah... sounds great
[17:08] <[raphael]> stf^rocklinux: I restarted the ./scripts/Build-Target
[17:08] <[raphael]> and now it starts with building bash3 right away
[17:09] <[raphael]> so that should work out fine
[17:09] <[raphael]> in stage 1
[17:09] <[raphael]> uh - failed ;)
[17:09] <[raphael]> because I didn't run the download script
[17:17] <[raphael]> bash3 got built in stage 1 and bin/bash points to bin/bash3, build continues, so things seem to work that way
[17:21] <stf^rocklinux> good, but keep in mind that bash 3.1 as default bash might break some packages (it's not well-tested atm.)
[17:21] <[raphael]> oho, I see
[17:22] <[raphael]> I'll revert to bash2 then in the next build (it is a serious production system, so I should be careful)
[17:23] <stf^rocklinux> IIRC some packages' configure scripts didn't work, but I haven't investigated further then.
[17:24] <stf^rocklinux> SMP: udev rule for OSS compat modules is here: 2006060917192322915
[17:24] <SMP> *click*
[17:44] <SMP> stf^rocklinux: I think that's a little too much brute force and it triggers on way too many events (even on the events that are caused by snd-pcm-oss loading) -- the following approach would be better, IMHO: if SUBSYSTEM="sound", KERNEL=="pcmC[0-9]*", RUN+="sh -c 'test -e /sys/class/sound/dsp || modprobe snd-pcm-oss'"
[17:45] <SMP> stf^rocklinux: snd-pcm-oss causes snd-mixer-oss to be loaded, so that doesn't need any extra handling. so a second rule only for snd-seq-oss would be needed
[17:47] <SMP> stf^rocklinux: my rule has bad syntax, but you'll figre that out ;-)
[17:47] <SMP> I'll try more later. bbl
[17:50] <stf^rocklinux> cy
[18:04] [raphael] (n=raphael@raphael.netpark.at) left irc: "using sirc version 2.211+KSIRC/1.3.12"
[18:10] stf^rocklinux (n=user@heim-032-98.raab-heim.uni-linz.ac.at) left irc: "using sirc version 2.211+KSIRC/1.3.12"
[19:11] stf^rocklinux (n=user@heim-032-98.raab-heim.uni-linz.ac.at) joined #rocklinux.
[19:11] stf^rocklinux (n=user@heim-032-98.raab-heim.uni-linz.ac.at) left irc: Client Quit
[19:13] stf^rocklinux (n=user@heim-037-186.raab-heim.uni-linz.ac.at) joined #rocklinux.
[20:35] Nick change: _toberkel -> toberkel
[21:49] daja77_ (n=daja77@dslb-088-072-040-168.pools.arcor-ip.net) joined #rocklinux.
[21:59] daja77 (n=daja77@dslb-088-072-050-255.pools.arcor-ip.net) left irc: Read error: 110 (Connection timed out)
[00:00] --- Sat Jun 10 2006