T2 IRC Log: 2005-01-22

This is the log as captured by an IRC bot in the channel. The statements are those of the individual people and might not neccessarily reflect the policy and legal rules as set forth by the T2 SDE Project.

« prev | next »

--- Log opened Sat Jan 22 00:00:13 2005
00:01 < rxr> [5] security/libksba [5] multimedia/libdvb
00:01 < rxr> 1393 builds total, 363 completed fine, 2 with errors.
00:02 < jsaw> libksba?
00:07 < rxr> looks like PARALLEL problem ...
00:10 < rxr> since libdvb built in the past maybe it is PARALLEL trouble, too
00:13 < rxr> hm - interesting
00:13 < rxr> mkdosfs work (on powerpc) on a file - but not on a usb stick partion ? (unable to get drive geometry for '/dev/discs/disc1/part1')
00:18 < mnemoc> where is drive geometry?
00:19 < rxr> my stick has none ...
00:19 < rxr> s/my/any/
00:19 < mnemoc> oh
00:20 < rxr> looks like a mkdosfs bug ...
00:21 < rxr> what is this:
00:21 < rxr> uba: device 3 capacity nsec 512000 bsize 512
00:21 < rxr> uba: was not changed
00:21 < rxr> /dev/ub/a: p1 p2
00:21 < rxr> usbcore: registered new driver ub
00:21 < rxr> ub ?
00:22 < mnemoc> my stick is 'scsi'
00:22 < rxr> mine normally too
00:22 < rxr> maybe this is this crippled usb storage boot driver ?
00:23 < rxr> CONFIG_BLK_DEV_UB=m
00:23 < mnemoc> that matches dev/ub/ :)
00:24 < mnemoc> i have to go, cu later
00:24 < rxr> cu
00:24 < rxr> later?
00:24 < rxr> == monday ?
00:25 < _Ragnar_> rehi
00:26 < rxr> CONFIG_BLK_DEV_UB:
00:26 < rxr> This driver supports certain USB attached storage devices
00:26 < rxr> uch as flash keys.
00:27 < rxr> and in 2.6.10 it additionally contains:
00:27 < rxr> Warning: Enabling this cripples the usb-storage driver.
00:27 < rxr> and the config option is named:
00:27 < rxr> Low Performance USB Block driver
00:29 < _Ragnar_> *lol*
00:29 < rxr> http://linux.usc.edu/pipermail/usclug-chat/2004-December/000111.html
00:32 < CIA-12> rene * r5768 /trunk/package/base/linux26/disable-broken.lst:
00:32 < CIA-12> * added CONFIG_BLK_DEV_UB to the linux26 disable-broken list
00:32 < CIA-12> (the driver is work in progress, slow and contains a warning not to
00:32 < CIA-12> be enabled - especially not by distributions)
00:32 < rxr> too bad I already have it on my production athlon - and do not feel like recompiling the kernel right now ..
00:32 < _Ragnar_> isn't a module?
00:33 < rxr> yes - and activating it cripples usb-storage
00:33 < rxr> at compile time ...
00:34 < rxr> like removing device ids ...
00:34 < _Ragnar_> oh
00:36 < rxr> hm - at least mkdosfs does not work on my athlon neither ...
00:36 < rxr> although tested with ub intead of usb-storage for obvious reasons ...
00:38 < rxr> SPACE.com is reporting that the White House has eliminated funding for servicing the Hubble Space Telescope from its 2006 budget request ?
00:54 < CIA-12> rene * r5769 /trunk/package/filesystem/dosfstools/reinstall.patch: * really fixed dosfstools rebuild - by adding the patch - sigh
00:55 < _Ragnar_> again?
00:55 * _Ragnar_ sighs
00:55 < rxr> well - it is not my night - forgot the svn add ...
00:55 < _Ragnar_> yes, bushs friends do need that money much more then scientific progress ....
00:56 < _Ragnar_> ;)
00:56 < CIA-12> rene * r5770 /trunk/package/ (3 files in 3 dirs): * marked libdvb, gpm and libksba NOPARALLEL
00:57 < rxr> yes - bush definetly needs the money ....
00:57 < rxr> this way we get a nice overview which authors could fix their Makefiles ...
01:01 < rxr> eiiii
01:01 < rxr> damn - on my iBook this damn ub driver us used, too
01:01 < rxr> it pushes amazing 90kB/s thru my USB port ...
01:01 < _Ragnar_> wow ;)
01:02 < rxr> looks like time to rebuild the kernel ...
01:02 < _Ragnar_> yeah
01:02 < rxr> too bad the 800Mhz iBook needs over an hour for all the t2 enabled modules ...
01:02 < _Ragnar_> ouch
01:04 < rxr> getting 16MB from the stick:
01:04 < rxr> real 3m10.653s
01:04 < rxr> user 0m0.000s
01:04 < rxr> sys 0m0.001s
01:08 < rxr> ok - that looks like now way without a kernel rebuild ...
01:09 < jsaw> ooo does not want to be build... the python error is gone, new error appeared.
01:10 < _Ragnar_> ooo?
01:10 < jsaw> OpenOffice.org
01:10 < rxr> yeah - as OpenOffice.Org is ...
01:10 < rxr> == 01:10:25 =[9]=> Building base/linux26 [2.6.10 2.1.0-beta3].
01:10 < rxr> so - I bett grab s.th. to eat ...
01:11 < _Ragnar_> 5-indent 5-esound 5-libexif fail in my build
01:12 < rxr> uclibc ?
01:12 < _Ragnar_> yes
01:12 < _Ragnar_> what is sinf ?
01:12 < rxr> some sinus function?
01:12 < rxr> sine even ..
01:13 < rxr> I bet forgotten -lm problem ..
01:13 < rxr> if uclibc has s.th. like that ...
01:13 < rxr> for glibc it sometimes works without -lm ...
01:13 < rxr> (libm(ath))
01:14 < CIA-12> jsaw * r5771 /trunk/package/develop/scons/parse-config: * one minus in front of -jobs missing
01:14 < rxr> oh
01:15 < _Ragnar_> no, it has -lm
01:15 < rxr> jsaw: thanks
01:16 < _Ragnar_> oh, sin(float)
01:17 < jsaw> == 01/22/05 01:19:38 =[5]=> Finished building package blender.
01:20 < rxr> ;-)
01:22 < _Ragnar_> now what turns this sin(x) into sinf(x) ?
01:23 < rxr> many the compiler when you throw a float into it?
01:23 < rxr> (if there is no macro that does it in the pkg source)
01:24 < rxr> you could work around the former (if you have no sinf) by casting it to double
01:24 < _Ragnar_> good idea
01:26 < _Ragnar_> doesn't work tho :/
01:27 < rxr> are you sure is it is the sin you are looking at?
01:27 < rxr> maybe there is one somewhere else ...
01:27 < _Ragnar_> positive
01:27 < _Ragnar_> the line number matches
01:28 < rxr> there seems to be sinf in uclib
01:28 < rxr> accordingl google on the topic ..
01:29 < rxr> try putting -lm to the end of the command line
01:32 < mnemoc> re
01:32 < rxr> hi mnemoc !
01:33 < mnemoc> hi rxr
01:34 < mnemoc> that thing of putting -l at the end is a feature or a bug?
01:35 < rxr> feature
01:35 < rxr> _Ragnar_: did you try it ?
01:38 < _Ragnar_> it was already there
01:38 < rxr> uhm
01:39 < mnemoc> what readelf thinks about libm.so? does it has sinf?
01:40 < mnemoc> does it has or does it have?
01:40 < _Ragnar_> 5-gawk: Found shared files with other packages:
01:40 < _Ragnar_> usr/share/locale/locale.alias: gawk gettext grep sed
01:40 < _Ragnar_> mnemoc: it does not
01:40 < mnemoc> uhm
01:43 < mnemoc> i don't see sinf on uClibc source
01:45 < _Ragnar_> me neither
01:47 < mnemoc> what package wants sinf?
01:49 < _Ragnar_> esound
01:49 < _Ragnar_> (not important to my build, but the error likely will manifest somewhere else, too)
01:51 < mnemoc> only sin (double) is implemented
01:52 < mnemoc> and very lightlu
01:52 < _Ragnar_> hmm why doesn't it work when I do sin( (double)( float...) ) ?
01:52 < mnemoc> doesnt?
01:53 < rxr> _Ragnar_: maybe the compiler is too smart and deduces it again ...
01:53 < rxr> could you declare the initial variable as double ?
01:53 < _Ragnar_> doesn't
01:53 < _Ragnar_> uhm
01:55 < _Ragnar_> I'll try
01:56 < _Ragnar_> btw why doesn't it remove the old error logs
01:56 < _Ragnar_> ?
01:56 < rxr> that is only don up to the abort on error up to stage X setting
01:59 < _Ragnar_> close ... had [ ] rebuild broken packages off
02:01 < _Ragnar_> what do we do with pkgs that include argp.h, which seems to be glibc only?
02:01 < rxr> well - nontheless I think it removes error logs for the abort on stage up to X option ...
02:01 < rxr> depends if you need that particular packages
02:02 < _Ragnar_> rng-tools ... probably not
02:02 < rxr> if you need it you might patch it to build without - or add support to uclibc ...
02:03 < _Ragnar_> what about the shared files error?
02:03 < rxr> no idea why it happens - take a look into the packages and see what they do to the file ...
02:04 < _Ragnar_> kay
02:04 < _Ragnar_> another thing: expat and bdb33 fail with:
02:04 < _Ragnar_> checking build system type... Invalid configuration `i686-unknown-linux-uclibc': machine `i686-unknown-linux' not recognized
02:04 < rxr> hm - normally they do not fail because of unkknown here ...
02:04 < rxr> guess it is only about the uclibc that confuses them
02:05 -!- mnemoc [~amery@200.75.27.59] has quit [Read error: 60 (Operation timed out)]
02:05 < rxr> altough we might to move from unkonwn to t2 ...
02:10 < _Ragnar_> hm
02:15 < jsaw> g'night
02:16 < rxr> n8 jsaw - cu
02:18 -!- mnemoc [~amery@200.75.27.20] has joined #t2
02:20 < _Ragnar_> I think it objects to the i686-unknown-linux-uclibc <-- 4 parts
02:21 < mnemoc> ?
02:21 < mnemoc> what did i miss?
02:23 < rxr> 02:04 < _Ragnar_> another thing: expat and bdb33 fail with:
02:23 < rxr> 02:04 < _Ragnar_> checking build system type... Invalid configuration
02:23 < rxr> `i686-unknown-linux-uclibc': machine `i686-unknown-linux' not
02:23 < rxr> recognized
02:23 < rxr> 02:20 < _Ragnar_> I think it objects to the i686-unknown-linux-uclibc <-- 4
02:23 < rxr> parts
02:23 < mnemoc> read uclibc/parse-config-9
02:24 < mnemoc> three diferent kind of hacks are there for that
02:24 < mnemoc> i think bdb is already there
02:25 < mnemoc> if the file is the same you can do bdb|bdb33) ...
02:27 < mnemoc> and, if you have your mind clear, see if there is a real difference between fix_configsub and add_cases
02:28 < _Ragnar_> uclibc/parse-config-9 <-- where?
02:28 < mnemoc> pacakge/base/uclibc/parse-config-9
02:29 < _Ragnar_> oh
02:29 < _Ragnar_> duh
02:29 < mnemoc> that's where uclibc-magic lives
02:29 < mnemoc> i hope it's easier to read than gcc.conf :p
02:29 < _Ragnar_> ;p
02:33 < _Ragnar_> the sed statements are bemuddling
02:34 < mnemoc> bemuddle
02:34 < mnemoc> \Be*mud"dle\, v. t. To muddle; to stupefy or bewilder; to confuse.
02:34 < mnemoc> they inject linux-uclibc where they find linux-gnu references
02:35 < mnemoc> add_case2 complements add_case where foo) is multiline
02:35 < _Ragnar_> yeah
02:35 < mnemoc> but now i'm not really sure if add_case and fix_configsub have really different effects
02:37 < _Ragnar_> might well be;)
02:37 < _Ragnar_> expat works with fix_configsub
02:37 < mnemoc> btw, they output the patch so you can 'validate' it reading the log
02:38 < mnemoc> exposed magic :)
02:39 < mnemoc> your patches to make foo build against uclibc shall go to uclibc/patches/foo-
02:39 < mnemoc> and .conf hacking goes to uclibc/patches/foo.conf
02:40 < mnemoc> uclibc/patches/foo.conf is loaded before $confdir/foo.conf
02:40 < _Ragnar_> nifty:)
02:45 < rxr> when have I mentioned the last time the need for a faster SPARC ?
02:45 < _Ragnar_> bdb33 still fails, due to using -o .libs/ as output name
02:45 < _Ragnar_> (for a .so)
02:48 < rxr> cool - just haved bdb to link the cxx bindings on sparc64
02:49 < rxr> /haved/hacked/ ...
02:49 < rxr> so - new tea
02:49 < _Ragnar_> do I still need bdb33 anyway?
02:50 < rxr> yes, there was s.th. with some minor utility packages ...
02:50 < _Ragnar_> dang ;p
02:51 < rxr> Error logs from reference-2.1.0-beta3-x86-athlon-32-reference:
02:51 < rxr> 1393 builds total, 404 completed fine, 0 with errors.
02:51 < _Ragnar_> ;p
02:54 < rxr> well - at the end some toy stuff will show up
02:54 < rxr> + the audacity and xmule wxwidgets regresions ...
03:00 -!- rxr_ [~rene@p213.54.198.121.tisdip.tiscali.de] has joined #t2
03:00 -!- Topic for #t2: T2 | 2.1.0-beta2 RELEASED | The next generation of System Development Enviroments (SDE) | http://www.t2-project.org/ | CIA, don't ever leave us again!
03:00 -!- Topic set by mnemoc [] [Wed Jan 12 03:04:43 2005]
03:00 [Users #t2]
03:00 [ _Ragnar_] [ daja77] [ mnemoc] [ Postal ] [ rxr ] [ valentin]
03:00 [ CIA-12 ] [ jsaw ] [ nzg ] [ praenti] [ rxr_]
03:00 -!- Irssi: #t2: Total of 11 nicks [0 ops, 0 halfops, 0 voices, 11 normal]
03:00 -!- Channel #t2 created Sun Aug 8 21:15:33 2004
03:00 -!- [freenode-info] why register and identify? your IRC nick is how people know you. http://freenode.net/faq.shtml#nicksetup
03:00 -!- Irssi: Join to #t2 was synced in 10 secs
03:01 < rxr_> rehi
03:01 -!- sparc-kly [~mubex@65-23-206-237.prtc.net] has joined #t2
03:01 < rxr_> hi sparc-kly
03:01 < sparc-kly> hi all
03:02 < _Ragnar_> how do I do this when each file has to belong to one specific package?
03:02 < _Ragnar_> and what the packages add is this:
03:02 < _Ragnar_> # Packages using this file: gettext-runtime gettext-tools grep sed gawk indent libexif
03:02 * _Ragnar_ hits head on table
03:03 < rxr_> what do the packages add ?
03:03 < _Ragnar_> their name to this line
03:03 < rxr_> oh my god
03:03 < rxr_> what was the filename ?
03:03 < _Ragnar_> yes
03:03 < _Ragnar_> usr/share/locale/locale.alias
03:04 < rxr_> they do not do this here ..
03:05 < rxr_> var_append flistdel "|" "usr/share/locale/locale.alias"
03:06 < rxr_> to uclibc parse-config
03:06 < rxr_> as a workaround ...
03:06 < rxr_> or whatever you want to call this ..
03:07 < _Ragnar_> okay
03:08 < rxr_> poste a note if it does what you/we, I would commit it then ...
03:09 < _Ragnar_> hmm that way it won't ever install that file?
03:09 < rxr_> oh - yes, indeed
03:10 < rxr_> huh?
03:10 < rxr_> normally it comes from :
03:10 < rxr_> /var/adm/flists/glibc:glibc: usr/share/locale/locale.alias
03:10 < rxr_> does uclibc install it ?
03:11 < _Ragnar_> negative
03:12 < rxr_> is there any content in that file on your side ?
03:12 < rxr_> except the # comment ?
03:12 < _Ragnar_> yes
03:13 < mnemoc> re
03:14 < _Ragnar_> deutsch de_DE.ISO-8859-1
03:14 < _Ragnar_> dutch nl_NL.ISO-8859-1
03:14 < _Ragnar_> for example ...
03:14 < rxr_> _Ragnar_: could you find out which package installs it first with usable content ?
03:14 < mnemoc> usr/share/locale/locale.alias <--- del_flist ?
03:14 < _Ragnar_> I think gettext does
03:14 < rxr_> mnemoc: do you see this too ?
03:14 < mnemoc> uhm, i was scrolled
03:15 < mnemoc> no, i haven't experienced that
03:15 < mnemoc> _Ragnar_: you may send patches to apply on trunk to fix the stuff you have fixed
03:15 -!- rxr [~rene@p213.54.194.221.tisdip.tiscali.de] has quit [Read error: 110 (Connection timed out)]
03:15 < _Ragnar_> you're not compiling with locale support, aren't you?
03:16 < mnemoc> no :p
03:17 < _Ragnar_> I think the first package to run will create the file ... I'd suggest adding it to del_flist only if it already exists
03:17 < rxr_> hm - _Ragnar_ best find out why other packages beside gettest add a note to it ...
03:17 < rxr_> _Ragnar_: yep - that would be a possibility - best kept in the uclibc parse-config for now
03:17 < _Ragnar_> they seem to all carry a local copy that gets copied when it doesn't exist
03:17 < rxr_> but once it does exist ...
03:18 < rxr_> maybe it is installed if there is not a vertain signature in it ...
03:18 < _Ragnar_> they just run a sed to insert their own package name
03:19 < rxr_> how hostile
03:19 < _Ragnar_> if we do a stage 9 the problem would disappear
03:19 < rxr_> base/gettext/gettext.conf:# flistdel="$flistdel|usr/share/locale/locale.alias
03:19 < _Ragnar_> ?
03:20 < rxr_> it is already there - commented out !
03:20 < mnemoc> who removed that?
03:20 < rxr_> no idea ...
03:20 < mnemoc> svn blame it
03:20 < _Ragnar_> but only for gettext
03:20 < rxr_> should I paste the svn cat -r 2 www.rocklinux.net/svn again ?
03:20 < mnemoc> true
03:20 < rxr_> I guess there for ages
03:20 < mnemoc> :\
03:20 -!- You're now known as rxr
03:21 < rxr> yep
03:21 < rxr> cool!
03:21 < rxr> the subversion people fixed a oddity in some recent version!
03:21 < rxr> now it works when you do this in our tree:
03:21 < rxr> t2-trunk/package/base/gettext$ svn cat -r 2 gettext.conf
03:21 < rxr> ;-)!
03:21 < rxr> this was not possible in the past - due to renames on the way!
03:21 < mnemoc> haven't that always worked?
03:22 < mnemoc> aha
03:22 < rxr> nope - not in 0.xy times when I started ..
03:22 < rxr> never tried it since then again ...
03:22 < _Ragnar_> bbl, heading home
03:22 < rxr> cu rxr
03:22 < rxr> _Ragnar_: even ...
03:22 < rxr> ;-)
03:22 < mnemoc> svn mv = svn cp + svn rm
03:22 < _Ragnar_> =)
03:22 < rxr> mnemoc: nack
03:22 < mnemoc> cu rxr <--- good
03:23 < rxr> ;-)
03:23 < mnemoc> rxr: sure?
03:23 < rxr> well - but not yet much more IIRC ..
03:23 < rxr> mnemoc: what to do about the file ?
03:23 < mnemoc> can i be generated on runitme?
03:23 < mnemoc> it*
03:24 < rxr> the alias file ?
03:24 < mnemoc> yes
03:24 < rxr> no idea
03:24 < mnemoc> like depmod and ldcache
03:24 < rxr> on a glibc build only glibc seems to install and touch it ...
03:24 < rxr> would be interesting to read the makefile or so of the packages that does touch it ...
03:25 < mnemoc> yep
03:26 < rxr> 1023 builds total, 288 completed fine, 15 with errors.
03:26 < rxr> most errors are due to missing bdb so far ...
03:26 < rxr> ^- sparc-v9-64
03:26 < mnemoc> stop and fix?
03:26 < rxr> mnemoc: it is a ugly libtool problem ...
03:26 < rxr> I'm investigating it while the build progresses ...
03:26 < mnemoc> ic
03:26 < rxr> otherwise I would loose too many cpu ticks ...
03:27 < mnemoc> i hate libtool
03:27 < rxr> haha
03:27 < mnemoc> and friends
03:27 < rxr> I have libtoo and auto*
03:27 < rxr> mnemoc: ACK!!!
03:27 < mnemoc> :D
03:27 < rxr> this f*cking libtool detects some link dependency to crt* whatever and the link then fails due to including those files twice - once my libtool manually - and once thru the toolchain
03:28 < rxr> the same on x86-64 ...
03:28 < rxr> I can not imagine who was hacking such crap ...
03:28 < mnemoc> ulrich?
03:28 < rxr> was he ... ?
03:28 < mnemoc> kidding
03:29 < rxr> ah ;-) - but a reasonable guess ...
03:29 < rxr> "and no - you never get cvs access" ,-)
03:29 < mnemoc> :)
03:29 < rxr> ("I already break enought - so others would even break more")
03:30 < mnemoc> "my bugs are intentionals, don't ever try to fix them"
03:30 < rxr> yep
03:30 < rxr> this are the culprits :
03:30 < rxr> libltdl/libtool:predep_objects="/usr/lib/gcc/sparc64-unknown-linux-gnu/3.4.3/../../../../sparc64-unknown-linux-gnu/lib/../lib/crti.o /usr/lib/gcc/sparc64-unknown-linux-gnu/3.4.3/crtbeginS.o"
03:30 < rxr> libltdl/libtool:postdep_objects="/usr/lib/gcc/sparc64-unknown-linux-gnu/3.4.3/crtendS.o /usr/lib/gcc/sparc64-unknown-linux-gnu/3.4.3/../../../../sparc64-unknown-linux-gnu/lib/../lib/crtn.o"
03:33 < mnemoc> how is ia32 ref-build going?
03:33 < rxr> well
03:33 < rxr> just pasted some minutes ago
03:33 < rxr> but I can repaste
03:33 < rxr> [5] network/xmule
03:33 < rxr> 1393 builds total, 427 completed fine, 1 with errors.
03:33 * mnemoc scrolling
03:33 < mnemoc> wx?
03:34 < rxr> yes - as in the last ref build
03:37 < mnemoc> xmule.org loose their domain
03:38 < rxr> lost, hm ?
03:40 < mnemoc> Welcome to Xmule.org
03:40 < mnemoc> This domain is for sale!
03:40 < mnemoc> xmule.ws is operative
03:42 < rxr> Linux 2.6.11-rc2
03:42 * mnemoc fears lx2.6 releases
03:43 < rxr> lol
03:43 < rxr> I fear the non availablilty of s devel tree ...
03:43 < rxr> maybe we'll never see a new major version
03:43 < rxr> 2.6.10224323
03:43 < mnemoc> that's why i fear those releases
03:43 < mnemoc> yep
03:44 < rxr> in 2014
03:58 < rxr> hacked libtool on the sparc:
03:58 < rxr> == 01/22/05 03:00:52 =[5]=> Finished building package bdb.
03:58 < rxr> ^- need to properly fix it after a sleep cycle ...
04:05 < mnemoc> CIA-12: ?
04:05 < rxr> no commit yet
04:05 < mnemoc> :)
04:06 < CIA-12> mnemoc * r5772 /trunk/package/database/postgresql/postgresql.desc: * updated postgresql (8.0.0rc4 -> 8.0.0) and maintainship taken.
04:06 < mnemoc> i should have added a note to do a proper fix to ncurses.so/bash patching aclocal.m4
04:07 < mnemoc> will you commit theses .cache files?
04:07 < mnemoc> these*
04:09 < rxr> yeah - the cache files ...
04:10 < rxr> since you all want them so much I'll commit them
04:10 < rxr> == 01/22/05 03:09:16 =[5]=> Finished building package emacs.
04:10 < rxr> ^- u30
04:10 < mnemoc> i prefer updated .cache files instead of arcanic .cache files
04:10 < rxr> the deps are correct
04:10 < rxr> the rest is just noise ...
04:11 < mnemoc> how sure are you about the deps?
04:11 < rxr> quite sure
04:12 < rxr> I also sed'ed package renaming in them
04:12 < rxr> less noise ;-)
04:17 < mnemoc> there may be new deps :) and deprecated deps too ;)
04:27 < CIA-12> mnemoc * r5773 /trunk/package/archiver/zip/ (zip.conf zip.desc): * changed zip .conf to use $root and to not use $archdir
04:30 < rxr> == 03:31:59 =[3]=> Building filesystem/dosfstools [2.9 2.1.0-beta3].
04:30 < rxr> ^- sparc64
04:30 < rxr> [5] network/xmule [5] editors/xemacs
04:30 < rxr> [5] editors/xemacs-packages [5] network/net-snmp
04:30 < rxr> [5] multimedia/libcdio [5] mail/mew
04:30 < rxr> 1393 builds total, 464 completed fine, 6 with errors.
04:30 < rxr> looks like some more NOPARALLEL candidates ...
04:30 < rxr> == 01/22/05 03:32:52 =[3]=> Finished building package dosfstools.
04:30 < rxr> but not kno
04:30 < rxr> +
04:31 < rxr> w
04:31 < rxr> I try to sleep 2 1/2 hours ...
04:31 < rxr> hm - or should i not ..
04:31 < rxr> damn - I most probably will not wake up ..
04:33 < rxr> .oO
04:33 < rxr> iBook:
04:33 < rxr> == 01:10:25 =[9]=> Building base/linux26 [2.6.10 2.1.0-beta3].
04:33 < rxr> == 01/22/05 03:33:34 =[9]=> Finished building package linux26.
04:34 < rxr> n8 all
04:38 < mnemoc> n8 rene
04:38 < _Ragnar_> gn8
04:45 < CIA-12> mnemoc * r5774 /trunk/package/audio/cdrdao/ (cdrdao.conf make-more.diff): * changed cdrdao patch to apply with $patchopt
04:58 < CIA-12> mnemoc * r5775 /trunk/package/audio/cdrdao/cdrdao.conf: * rewritten cdrdao .conf to not use custmain
05:03 < CIA-12> mnemoc * r5776 /trunk/package/audio/vorbis-tools/vorbis-tools.conf: * removed hardcoded reference to vorbisinfo tool on .conf
05:06 < mnemoc> gn8 all too
05:23 < _Ragnar_> gn8 mnemoc
05:43 < Postal> rxr, you around?
07:03 -!- mipe [~mipe@dsl10040.japo.fi] has joined #t2
07:20 < rxr> moin
07:20 < rxr> hi Postal
07:21 < _Ragnar_> hi
07:30 < rxr> http://anca.sourceforge.net/
07:41 < _Ragnar_> I'm beginning to get the hang of this ;) fixed bdb33, electricfence and ash
07:44 < rxr> ;-)
07:44 < rxr> congratulations!
07:47 < _Ragnar_> thx;)
07:57 < rxr> do you send the patches to the list? or some url to pull?
07:58 < rxr> if you plan to develop / use t2 over a longer time you could also get write access to the svn ...
08:01 < _Ragnar_> I'll collect them and send them when they have stabilised enough
08:07 < rxr> ok
08:18 < mipe> hmm,what provides ss_err.h?
08:21 < rxr> ohm - now my sparc will busy the next hours ...
08:21 < rxr> == 07:20:03 =[5]=> Building kde/qt [3.3.3 2.1.0-beta3].
08:21 < rxr> /var/adm/flists/e2fsprogs:e2fsprogs: usr/include/ss/ss_err.h
08:22 < mipe> In file included from std_rqs.c:2:
08:22 < mipe> ../../lib/ss/ss.h:26:23: ss/ss_err.h: No such file or directory
08:22 < mipe> when building e2fsprogs
08:25 < rxr> hm
08:25 < rxr> is it there?
08:25 < rxr> maybe the package has a bug to not install it sometimes ?
08:27 < mipe> yeah,bug,now it builds and didnt change anything
08:27 < rxr> which package? e2fsprogs?
08:27 < rxr> such stuff is nearly always parallel make ...
08:27 < mipe> e2fsprogs
08:29 < rxr> ;-) NOPARALLEL ...
08:29 < rxr> cu later
08:29 * rxr away
08:29 < mipe> oh and i never get to stage 2 without manually making symlink to ld-linux.so from ld-linux-x86_64.so.2
08:30 < mipe> ok,cu
09:18 < rxr> re
09:18 * rxr at uni ...
09:18 < rxr> good that the assembly exercise I tought was relocated to today dit not yet start
09:18 < rxr> leaves me some time to actually do it ...
09:35 < _Ragnar_> goodnight folks
10:22 -!- mipe [~mipe@dsl10040.japo.fi] has quit ["Lost terminal"]
11:35 -!- mipe [~mipe@dsl10040.japo.fi] has joined #t2
12:21 < rxr> hi all
12:21 < rxr> so - one assembly junk handed out ...
12:21 < rxr> and the 2nd finished, too ...
12:21 < rxr> let's hope te remaining time is enought to hand it out, too ...
12:22 < rxr> my reserved SPARC vanished ...
12:22 < rxr> there are two U5's without RAM (...) and one with few RAM ...
12:22 < rxr> that happens when one does not install something and powers the box up on the same day when it is reserved ..
12:22 < rxr> well - maybe I use the one with not that much RAM for now ...
12:30 < rxr> damn
12:30 < rxr> due to my sparc assmble hacking libGL missses some symbols and e.g. qt fails linking ...
12:30 < rxr> aside from that the sparc64 builds looks very good ...
12:32 < mipe> on x86_64 front,i cant get gcc to build biarch system,hence the grub build fail. but off to do some work
12:33 < rxr> well - with :HEAD and multilib enabled you should get a biarch compiler ..
12:33 < rxr> at least on sparc64 it built ...
12:34 < mipe> rxr:i dont,everything compiles fine but fails on grub,tested if 32bit stuff where build with gcc -m32 hello.c, it fails
12:36 -!- mipe is now known as mipe|afk
12:48 < rxr> mipe|afk: does gcc -m32 -c hello.c work ?
12:59 < rxr> erhm
12:59 < rxr> col
12:59 < rxr> cool even
13:00 < rxr> valentin: x.org HEAD has XAA and basic render support for sunffb !!!
13:00 < rxr> Bugzilla Bug 1891
13:00 < rxr> XAA support for SunFFB Last modified: 2005-01-14 13:27
13:00 < rxr> https://bugs.freedesktop.org/show_bug.cgi?id=1891
13:00 < rxr> to bad I still have no idea how to assemble the .s files ..
13:03 < rxr> maybe they just use older binutils that do not coplain ...
13:03 < rxr> complain even ...
13:22 < rxr> daned IE
14:20 -!- mipe|afk is now known as mipe
14:56 < rxr> re
15:07 < mipe> Created file outside basedir: /usr/lib64/libopcodes.la
15:07 < mipe> error :)
15:07 < rxr> .oO
15:07 < rxr> binutils ?
15:07 < mipe> yeah
15:09 < mipe> eh,i have --libdir=/usr/lib64 defined,just for testing
15:09 < rxr> oh - that is the problem ...
15:09 < mipe> how to define to allow it to create file on that dir?
15:10 < rxr> you are at the additional configure options setting ?
15:10 < rxr> --libdir=\$root/usr/lib64
15:10 < rxr> should do - or simialr quoting variants ...
15:10 < rxr> but this would not work for all packages - e.g. not the ones without configure
15:10 < rxr> I do not recomment this ...
15:11 < mipe> oh,ok
15:11 < rxr> if you really want this modify Build-Pkg to set libdir there ...
15:12 < rxr> this it is also passed to most Makefiles
15:12 < rxr> but I bet a significant number of Makefiles just ignore it - read have hardcoded dirs ...
15:12 < mipe> gonna start fresh build with multilib enabled and see if it works. after stage 1 completes there should be 32 and 64 compiler?
15:13 < rxr> yes - the compiler should be able to handle both
15:13 < rxr> but there will not be a 32bit libc ...
15:14 < rxr> but afaics grub should built
16:24 < rxr> wow
16:24 < rxr> we have telnet on the bootdisk - but not the needed libstdc++ ...
16:25 < rxr> I vote to replace the libstdc++ depending and quite useles anyway telnet with openssh ...
16:26 < mipe> just tested 32/64 compile after they are built on stage 1,and they work.
16:27 < rxr> ;-)
16:27 < mipe> did you test grub build on your machine?
16:28 < rxr> I have no x86-64 here ...
16:28 < rxr> from what I see on my ultra sparc it should work in x86-64, too
16:29 * rxr taring up the current build state ... for extraction on a university machine ...
16:29 < mipe> ok,have to wait until its on that stage
16:41 < mipe> oh and theres no ld-linux.so.2,have to link it manually to get it to stage 2 build
16:49 < rxr> how is it named over there ?
16:50 < rxr> I need some sleep - only slept 2 1/2 hours in the morning ...
16:50 < mipe> ls ld*
16:50 < mipe> ld-2.3.4.so ld-linux-x86-64.so.2
16:50 < rxr> ic ..
16:50 < rxr> ok - I patch glibc accordingly ...
16:50 < rxr> later ..
16:51 < mipe> yeah,me off too
17:58 < valentin> http://exactcode.de/valentin/dont-worry-your-build-just-failed-allright.tiff
17:58 < valentin> this apple xcode is a quite cool prg...
19:35 < _Ragnar_> hi
19:48 < mipe> hmm,from grub error log i see line:INSERT = '-Os -pipe -march=athlon64' , how can i remove the --march been setted in build. with --march=athlon64 grub doesnt build on x86_64
19:55 < _Ragnar_> yuuuuuukkk
19:55 < _Ragnar_> man ftw <-- omg, bloatware
20:03 < _Ragnar_> but attr depends on it .... *sighs*
20:03 -!- Netsplit saberhagen.freenode.net <-> irc.freenode.net quits: mipe, CIA-12
20:05 -!- Netsplit over, joins: CIA-12, mipe
20:08 < _Ragnar_> mnemoc: how about we build what we can with uclibc, and build the rest with dietlibc?
21:30 -!- mipe [~mipe@dsl10040.japo.fi] has quit ["leaving"]
22:17 < rxr> ~re
22:18 < _Ragnar_> hi
--- Log closed Sun Jan 23 00:00:13 2005