T2 IRC Log: 2005-01-03

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 Mon Jan 03 00:00:07 2005
--- Day changed Mon Jan 03 2005
00:00 < jonhbruce> would someone please pass on a message to mnemoc. johnbruce has resolved the problem of getting binutils to compile by installing Gentoo. They don't seem tp mess everything up quite as much as others. Cheers and All thebest for the New Year. Scotsman with a sore head!!
00:00 -!- jonhbruce [~chatzilla@host81-157-242-89.range81-157.btcentralplus.com] has quit ["ChatZilla 0.9.61 [Mozilla rv:1.7.5/20050102]"]
00:14 < CIA-9> jsaw * r5365 /trunk/package/games/torcs/ (depend-fix.patch torcs.conf): * fix make dep generation in torcs
02:41 < CIA-9> jsaw * r5366 /trunk/package/games/squaroid/ (multiline-strings.patch squaroid.conf):
02:41 < CIA-9> * fix squaroid:
02:41 < CIA-9> - add a multiline string patch
02:41 < CIA-9> - fix .conf to accept multiple patches
02:49 -!- rxr_ [~rene@p213.54.198.163.tisdip.tiscali.de] has joined #t2
02:49 -!- Topic for #t2: T2 | 2.1.0-beta2 RELEASED | The next generation of System Development Enviroments (SDE) | http://www.exactcode.de/t2
02:49 -!- Topic set by mnemoc [] [Tue Dec 28 14:55:44 2004]
02:49 [Users #t2]
02:49 [ _Ragnar_] [ daja77] [ N0V4K] [ rxr_ ] [ valentin]
02:49 [ CIA-9 ] [ jsaw ] [ rxr ] [ sparc-kly]
02:49 -!- Irssi: #t2: Total of 9 nicks [0 ops, 0 halfops, 0 voices, 9 normal]
02:49 -!- Channel #t2 created Sun Aug 8 21:15:33 2004
02:49 -!- Irssi: Join to #t2 was synced in 10 secs
02:55 -!- rxr [~rene@p213.54.198.201.tisdip.tiscali.de] has quit [Read error: 60 (Operation timed out)]
03:34 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has quit [Remote closed the connection]
03:41 < CIA-9> jsaw * r5368 /trunk/package/develop/clisp/ (. bindings-glibc-fix.patch clisp.conf clisp.desc): * add clisp (2.33.2)
03:48 < CIA-9> jsaw * r5369 /trunk/package/scientific/maxima/maxima.conf: * gcl 2.6.5 is buggy, use clisp to compile maxima
03:52 -!- sparc-kly [~sparc@64.237.130.154] has quit [Read error: 54 (Connection reset by peer)]
04:05 < CIA-9> jsaw * r5370 /trunk/package/scientific/wxmaxima/ (. wxmaxima.desc): * add wxmaxima (0.5.3), a wxWidgets GUI interface to maxima
04:07 < CIA-9> jsaw * r5371 /trunk/package/scientific/emacs-matlab-mode/emacs-matlab-mode.desc: * add forgotten cksum
04:49 < CIA-9> jsaw * r5372 /trunk/package/games/pingus/ (pingus.conf pingus.desc):
04:49 < CIA-9> * update pingus (0.6.0 -> 2005-01-01), necessary due to clanlib update
04:49 < CIA-9> (but did not build before also... so...)
05:43 < CIA-9> jsaw * r5373 /trunk/package/scientific/emacs-maxima-mode/ (. emacs-maxima-mode.conf emacs-maxima-mode.desc):
05:43 < CIA-9> * add emacs maxima mode (0.9)
05:43 < CIA-9> (xemacs to be done)
07:31 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has joined #t2
07:31 -!- N0V4K_ [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has joined #t2
07:31 -!- N0V4K_ [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has quit [Remote closed the connection]
07:37 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has quit [Remote closed the connection]
07:39 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has joined #t2
08:53 < rxr_> moin
08:53 -!- You're now known as rxr
10:02 < rxr> remoin
11:18 < rxr> 1019 builds total, 975 completed fine, 44 with errors.
12:49 < rxr> re
12:49 * rxr lunching ...
13:02 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has quit ["FreeBSD N0V4K.Zero 5.3-RELEASE FreeBSD 5.3-RELEASE Chao"]
13:29 < rxr> http://slashdot.org/pollBooth.pl?qid=1233§ion=mainpage&aid=8
13:29 < rxr> ^- higher tea consume than I expected ...
13:39 -!- mnemoc [~amery@200.75.27.1] has joined #t2
13:42 < mnemoc> moin!
13:55 < CIA-9> mnemoc * r5374 /trunk/scripts/Download: * Introduced download/Translations to alter original download location using sed
13:56 < rxr> moin mnemoc
13:56 < mnemoc> moin rxr
13:57 < CIA-9> mnemoc * r5375 /trunk/misc/archive/sample_download_translations.sed: * Adds sample download/Translations on misc/archive including sample translations for ftp.gnu.org and dl.sf.net
13:57 < mnemoc> i you have a minute try to see that and comment please
13:58 < rxr> yes - just diffed it ..
13:58 < rxr> please use a default for the sourcefoge mess ...
13:58 < rxr> sourceforge.net even ..
13:58 < mnemoc> sample includes it
13:58 < mnemoc> do you want me to auto copy sample into download/ ?
13:59 < rxr> no not auto copy ...
13:59 < rxr> please move the sample to:
14:00 < rxr> misc/data/DownloadTranslations
14:00 < rxr> and use it as default
14:00 < mnemoc> ah, ok
14:00 < rxr> I later move the stuff^W data used from Docuemntation/Developers/ into misc/data, too
14:00 < rxr> an no
14:01 < rxr> ah - no even
14:01 < rxr> name it misc/share
14:01 < rxr> that is what we agreed on weeks ago ...
14:01 < mnemoc> ok
14:01 < mnemoc> yes?
14:02 < rxr> yes ...
14:02 < rxr> t2-2004-12-08
14:02 < rxr> t2-2004-12-09
14:02 < rxr> ;-=
14:02 < rxr> ;-) even
14:02 < rxr> oh my typing ..
14:03 < mnemoc> :D
14:03 < rxr> cool - raning over here ...
14:03 < rxr> rain ...
14:03 < rxr> uhm
14:03 * rxr coffee ...
14:04 < CIA-9> mnemoc * r5376 /trunk/misc/ (3 files in 2 dirs): * moved sample download/translations into misc/share/ as rene requested
14:05 < rxr> requested ? ;-)
14:05 < rxr> mnemoc: btw. t2 ML aka. /dev/null is awaiting comments ;-)
14:05 < mnemoc> :D
14:06 < mnemoc> let me change Download and i comment
14:09 < rxr> yeap
14:10 < rxr> in the meantime I remove the .*.dl.sf.net stuff from the .desc files
14:10 < rxr> mnemoc: take your time with the comments - I was just joking cause you called the list /dev/null before ,-)
14:10 < CIA-9> mnemoc * r5377 /trunk/scripts/Download: * improved Translations handling using misc/share/DownloadTranslations if download/Translations is not found
14:11 < mnemoc> rxr: let me count..... i have 5 unreplied messages :)
14:11 < mnemoc> rxr: i have dl.sf.net fixed here if you want me to commit it
14:11 < rxr> well - haven't we all of them cleared here?
14:11 < rxr> you already fixed all?
14:11 < rxr> sure commit them ...
14:12 < rxr> well - amazing how long a rm -rf ccache can take ...
14:12 < mnemoc> rxr: sure, they are almost all cleared here, but not on the list :)
14:12 < rxr> quite many files to rm ... - although it is the bleeding fast reiserfs ...
14:12 < rxr> but maybe it is due to my slow laptop HD ...
14:13 < CIA-9> mnemoc * r5378 /trunk/package/filesystem/parted/parted.desc: * updated parted (1.6.16 -> 1.6.20)
14:14 < mnemoc> mi p4/256 runs far slower than my p3/512... i guess just because swapping on this damn slow HD
14:15 < CIA-9> mnemoc * r5379 /trunk/package/ (19 files in 19 dirs): * fixed download locations back to use dl.sourceforge.net because now we can translate them
14:26 < mnemoc> rxr: may i have problems if i mine -i gcc and binutil built for glibc23-xmas on a machine with glibc2.3.2?
14:27 < mnemoc> rxr: may i have problems if i mine -u glibc in that case?
14:29 < rxr> slowly slowly ...
14:29 < rxr> well - mine -u glibc will give you an unusable system in the case all is linked against glibc ...
14:30 < rxr> when building glibc the first build will give you a segfault most probably in make due to the wrapper beeing dynamically linked against the old one and it is still hold in memory due to the wrapper ...
14:30 < rxr> aside from that it should work ...
14:30 < mnemoc> i have glibc already built, on a .gem
14:31 < mnemoc> for the same config the machine was built
14:31 < mnemoc> but gcc and glibc were different versions
14:31 < mnemoc> 2.0.1 ---> 2.1.0-beta3
14:32 < CIA-9> rene * r5380 /trunk/package/develop/ccache/ (config-560.hlp config-560.in parse-config):
14:32 < CIA-9> * added cache size option to ccache, and cleaned one line in the
14:32 < CIA-9> cache.conf as well as rewrote the help text
14:32 < mnemoc> i only want to distcc good code sharing my laptop's
14:32 < rxr> hm
14:32 < mnemoc> cpu power
14:33 < mnemoc> wont `mine` crash during the update?
14:33 < rxr> the glibc should stay compatibly with 3.{2,3,4} - AFAIR no C ABI change was mde
14:33 < rxr> mine -i glibc-binary?
14:34 < rxr> it shoudl not - it it does it is a bug in mine ...
14:34 < mnemoc> -u
14:34 < rxr> well - after -u you have no system anymore ...
14:34 < mnemoc> oh
14:34 < rxr> I mean you can not start any dynamically linked binary anymore
14:34 < rxr> but - ok - you have a statically linked mine ..
14:34 < mnemoc> do i?
14:35 < mnemoc> mine.static :D
14:35 < rxr> test it before you try this route ...
14:36 < mnemoc> mine.static runs, but how can i test a core update like this without actually doing it?
14:36 < rxr> I meant if it is really statically linked ;-)
14:37 < mnemoc> glibc statics are always nss dependant
14:38 < rxr> well - quite
14:38 < rxr> don't they work if not nss thing is around?
14:38 < rxr> I mean - they only seg-fault if you mix the major version nss.so's ...
14:39 < jsaw> re
14:39 < mnemoc> they run without nss
14:40 < rxr> hi jsaw
14:40 < mnemoc> hi jsaw
14:40 < mnemoc> i'll update this: bin86-0.16.16.gem binutils-2.15.94.0.1.gem gcc-3.4.3.gem glibc-2.3.4-2004-12-24.gem libtool-1.5.10.gem linux-header-2.6.8.1.gem and prey
14:40 < jsaw> mnemoc: you could have a bootable mini-partition to check such core updates
14:40 < jsaw> hi rxr, mnemoc
14:41 < mnemoc> jsaw: qemu....
14:41 < mnemoc> jsaw: i would love to have a good rescue disk to do 'make world' safely
14:41 < mnemoc> make system*
14:43 < jsaw> btw bootdisk
14:43 < jsaw> Creating small 2nd stage filesystem:
14:43 < jsaw> -> Copy the most important programs ...
14:43 < jsaw> !> `- Program not found: mine
14:43 < jsaw> -> Copy the required libraries ...
14:43 < jsaw> Creating small 2nd stage filesystem:
14:43 < jsaw> -> Copy the most important programs ...
14:43 < jsaw> !> `- Program not found: mine
14:43 < jsaw> -> Copy the required libraries ...
14:43 < mnemoc> o_O
14:43 < jsaw> ups, sorry
14:43 < jsaw> -> Forgotten packages summary:
14:43 < jsaw> !> `- bdb
14:43 < jsaw> !> `- glibc
14:43 < jsaw> !> `- grub
14:43 < jsaw> !> `- joe
14:43 < jsaw> !> `- kbd
14:43 < mnemoc> no mine on host?
14:43 < jsaw> !> `- libol
14:43 < jsaw> !> `- syslinux
14:43 < mnemoc> that's a summary i did
14:43 < jsaw> this is very strange.
14:44 < mnemoc> those packages are built but not mapped
14:44 < jsaw> pkgs/mine.tar... is there
14:44 < jsaw> glibc is there, but still build scripts complains about it missing in pkg map
14:44 < jsaw> etc. etc.
14:44 < mnemoc> there is a map with -$pkg to not include on stage2 and +$pkg to include on stage2
14:45 < mnemoc> glibc22 and glibc23 are on the map :p
14:45 < jsaw> +glibc, -mine
14:45 < mnemoc> oops, wrong dir
14:46 < mnemoc> i'll do my updates.... cu in my next life
15:05 < rxr> ouch - I have to do some even more boring excercises regarding file i/o and pipes for my system programming lecture ...
15:05 < rxr> oh my is this boring ...
15:06 < rxr> every single excersise is a new challenge to not quit this time wasting study ...
15:13 < rxr> we again have an higher revision than rock ,)
15:24 < mnemoc> uhm.... glibc removal removed mine.static and bin/* and etc/*
15:24 < mnemoc> kiss on stage1 is damn useless
15:24 < mnemoc> it can't mount
15:24 < mnemoc> it doesn't have readline
15:24 < jsaw> what?
15:24 < mnemoc> sucks
15:24 < jsaw> remove bin/* and etc/*
15:24 < mnemoc> yes
15:25 < jsaw> mine does recursive directory removal...???
15:25 < mnemoc> i guess i had a broken mine
15:25 < mnemoc> i should had updated mine before playing
15:26 < mnemoc> now i'm forced to install t2 on my laptop... but it panics after loading both stage2 images
15:26 < mnemoc> killing init :\
15:27 < jsaw> why's init exiting?
15:27 < jsaw> I mean linuxrc
15:28 < jsaw> ?
15:28 < mnemoc> can't mount real-root
15:28 < mnemoc> btw, that massive insmod is a bit ugly
15:36 < jsaw> goota go, cu later
15:36 < jsaw> gotta *g*
15:37 < rxr> mnemoc: massive insmod ?
15:37 < mnemoc> after loading stage2 a /massive/ insmod starts
15:37 < mnemoc> lot of failings and some successes
15:43 < rxr> uhm
15:44 < mnemoc> it would be nice to have bootdisk on usb-storage :)
15:56 < CIA-9> jsaw * r5381 /trunk/target/bootdisk/build_stage2.sh:
15:56 < CIA-9> * fix stage2 build in bootdisk:
15:56 < CIA-9> - add +grub to the package map
15:56 < CIA-9> - add +joe,+libol (please review) to package map
15:56 < CIA-9> - handle tabs in package_map
15:56 < CIA-9> - 'sed s,,,p' does not work here -> workaround is
15:57 < CIA-9> inverted logic of matching $y versus "+"
15:58 < rxr> jsaw: please use * for all enumerations in commit messages ;-)
15:59 < mnemoc> jsaw: hehe
15:59 < jsaw> no sub-items?
15:59 < mnemoc> indentation
16:00 < rxr> yep 2 spaces per sub item
16:01 < rxr> http://svn.exactcode.de/ChangeLog-t2
16:01 < rxr> see *) 2005-01-01
16:01 < mnemoc> jsaw: what was the problem with "$y" == "+" ?
16:02 < jsaw> the sed s,...,p doesn't stop after matching the first entry
16:02 < jsaw> so mine yields y="+-"
16:02 < mnemoc> oh
16:03 < mnemoc> can i move ISO stuff to iso/ preserving hardlinked files to be able to export without mounting?
16:04 < jsaw> rxr: just had a look at 2005-01-01 entry. On first glance this does not look like sub-items... anyway
16:05 < mnemoc> ack @ jsaw ... /me hides
16:05 < jsaw> look at r5162, this is much more obvious
16:06 < jsaw> but probably too complicated commit rules will do a worse job...
16:13 < rxr> I talked about r5349
16:13 < rxr> but two
16:13 < rxr> but a tab per indention level would also be ok ...
16:13 < jsaw> yep, I know, I looked at r5349
16:14 < rxr> but I would perfer one enumeration indicator - like the already used * at top-level
16:14 < jsaw> ok.
16:14 < rxr> hm - just got a new editor offering ...
16:15 < jsaw> (hope I do not forget, I'm so much used to * - . items/sub/subsub in text....)
16:16 < rxr> top - 16:14:40 up 11 days, 3:43, 3 users, load average: 11.99, 6.21, 3.97
16:16 < rxr> thanks to kdebindings on my iBook
16:16 < rxr> damn package ..
16:16 < rxr> killed it ...
16:17 < mnemoc> yes, damn package
16:17 < rxr> seems to be libsmokekde_la.all_cpp.cpp
16:17 < rxr> ah
16:17 < rxr> can it be that it is due to --enable-final
16:17 * rxr testing
16:19 < rxr> mnemoc: have you tested the imagemagick update?
16:19 < rxr> btw:
16:19 < rxr> -> Compiler Cache Hits while pkg build: 99.79% (483 hits, 1 misses)
16:19 < mnemoc> it builded
16:19 < rxr> ^- imagemagick rebuild
16:19 < rxr> mnemoc: no shared files like tla-s.th.a ?
16:19 < mnemoc> but nothing yet on runtime
16:20 < mnemoc> uhm? no i didn't got shares
16:20 < rxr> ok - I removed flist - maybe it was due to my unfinished update
16:31 < rxr> should I register t2-sde.org then?
16:31 < rxr> or wait for more votes and/or idea?
16:31 < rxr> +s
16:32 < mnemoc> it would be nice to hear some users.... but i guess they will not answer
16:35 < rxr> I could phone two or three and ask for preferences ...
16:36 < mnemoc> aren't phone-surveys prohibited?
16:39 < rxr> I mean - those well^Wnot paying customers that run and know T2 already anyways ....
16:39 < mnemoc> :)
16:39 < mnemoc> phone them :)
16:39 < rxr> not suem random call - we have this cool OS you waited 20 years for oder it now (only 1499EUR) and get get special 15% savings ...
16:39 < rxr> s/sume/some/
16:39 < rxr> s/oder/order/
16:39 < rxr> damn
16:39 * rxr new coffee ...
16:43 < rxr> .oO
16:44 < rxr> when I chroot into the desktop or install build dir
16:44 < rxr> all stuff executed in the resulting shell will never return to the shell
16:45 < rxr> 28666 pts/11 R 0:21 /bin/bash -i
16:45 < rxr> 28667 pts/11 Z+ 0:00 [bash]
16:45 < rxr> 28670 pts/3 R 0:14 /bin/bash -i
16:45 < rxr> 28671 pts/3 Z+ 0:00 [ls]
16:45 < rxr> 28672 pts/4 R 0:05 /bin/bash -i
16:45 < rxr> 28673 pts/4 Z+ 0:00 [ls]
16:45 < mnemoc> oh
16:46 < rxr> linux-gate.so.1 => (0xffffe000)
16:47 < rxr> ?
16:47 < rxr> # ldd bin/bash
16:47 < rxr> linux-gate.so.1 => (0xffffe000)
16:47 < jsaw> syscall
16:47 < rxr> I imagined
16:48 < rxr> it may not be glibc's fault - the bash binary dos also behave that way when run with the in-system glibc.2.3.3 ...
16:48 < rxr> kiss and ash work
16:48 < jsaw> I don't have this.
16:48 < rxr> you don't have this updated glibc, I guess?
16:49 < jsaw> nope.
16:49 < rxr> zsh works, too
16:49 < jsaw> and also not that bash behavious.
16:49 < jsaw> (in chroot)
16:49 < rxr> I never had so far, too
16:50 < rxr> the bash is stuck in
16:50 < rxr> 0x400b8c68 in re_set_registers () from /lib/libc.so.6
16:50 < rxr> 0x400b8c68 in re_set_registers () from /lib/libc.so.6
16:50 < rxr> (gdb) bt
16:50 < rxr> #0 0x400b8c68 in re_set_registers () from /lib/libc.so.6
16:50 < rxr> #1 0x080f7b88 in ?? ()
16:50 < rxr> #2 0x080771b3 in waitchld (wpid=28673, block=1) at jobs.c:2487
16:50 < rxr> #3 0x0807653f in wait_for (pid=28673) at jobs.c:1898
16:50 < rxr> #4 0x080697af in execute_command_internal (command=0x80f3040, asynchronous=0,
16:50 < rxr> pipe_in=-1, pipe_out=-1, fds_to_close=0x80f3b00) at execute_cmd.c:700
16:50 < rxr> #5 0x0806921a in execute_command (command=0x80f3040) at execute_cmd.c:347
16:50 < rxr> #6 0x0805e31c in reader_loop () at eval.c:146
16:50 < rxr> #7 0x0805c7e8 in main (argc=2, argv=0xbffff1d4, env=0xbffff1e0) at shell.c:704
16:50 < rxr> hm - no - it is still running
16:51 < rxr> ouehm
16:51 < rxr> what the fuck
16:52 < rxr> 28606 root 25 0 2332 1312 1512 R 16.6 0.3 10:04.98 bash
16:52 < rxr> 28633 root 25 0 2340 1332 1512 R 16.6 0.3 6:28.90 bash
16:52 < rxr> 28666 root 25 0 1704 1032 1512 R 16.6 0.2 1:46.33 bash
16:52 < rxr> 28670 root 25 0 1704 1036 1512 R 16.6 0.2 1:38.70 bash
16:52 < rxr> 28682 root 25 0 2220 1492 2016 R 16.3 0.3 0:59.03 bash
16:52 < rxr> 28672 root 25 0 2404 1548 2208 R 16.3 0.3 1:20.80 bash
16:52 < rxr> my athlon is busy exacuting those affected bash's?
16:52 < rxr> .oO
16:52 < jsaw> o_O
16:52 < rxr> I do not like this ...
16:52 < rxr> neither bash nor glibc ...
16:53 < rxr> xmas 2005 whishlist:
16:53 < rxr> - glibc less system
16:53 < rxr> - bash less system
16:53 < rxr> ^A0^A1^A0^A2-> Compiler Cache Hits while pkg build: 68.13% (573 hits, 268 misses
16:54 < rxr> == 01/03/05 16:47:59 =[9]=> Finished building package imagemagick.
16:54 < mnemoc> - uclibc+busybox bootdisk :)
16:54 < mnemoc> 68%???
16:54 < rxr> yeah - no idea why
16:54 < rxr> ah - i Konw why ...
16:54 < rxr> imagemagick most probably completely cached
16:54 < rxr> but the counters are not per process
16:54 < rxr> they are ccache global - I built another package the same time
16:55 < rxr> so the bulid output is not that meaning full ...
16:55 < rxr> I would rather go dietlibc only bootdisk ...
16:55 < rxr> there is a dynamic version support in dietlibc - at least for x86
16:55 < rxr> (and arm and maybe already more)
16:56 < rxr> and on my fun todo is dyndiet on powerpc ...
16:56 < mnemoc> who is lars?
16:56 < rxr> a t2 users with write access ;-)
16:57 < rxr> living in berlin since october 2004 - now in south germany ... :-(
16:57 < mnemoc> and email address at your 'company'
16:57 < rxr> exactcode is exaping ;-)
16:57 < jsaw> dyndiet on powerpc? it is not working yet?
16:58 < rxr> the last time I looked (0.22 or so ) it was x86 and arm only ...
16:58 < rxr> I can quickly recheck if you are interested ...
16:58 < jsaw> cd ppp; grep dyn *
16:58 < jsaw> I mean cd ppc
16:58 < jsaw> start.S: bl _dyn_start
17:02 < rxr> yep - dynamic linking support seems to have been greatly extended in the meantime ...
17:06 < rxr> ok - I just called martin
17:07 < rxr> he was about to reply to the mail anyways ...
17:07 < rxr> on the phone he favoured t2-project ..
17:07 < rxr> what to do ? regisering t2-sde now ?
17:08 < mnemoc> your money :)
17:08 < rxr> yeah - I mean do we want it ?
17:08 < rxr> I do not want to register it and then get a but xyz would have been nice in tomorrow mornign ...
17:09 < mnemoc> before martin's it was unaanimous
17:11 < rxr> my brother favours -project ...
17:12 < mnemoc> has anyone voted _against_ t2-sde?
17:12 < mnemoc> (<0)
17:13 < jsaw> Re: altering some builtin variables segfaults bash
17:14 < mnemoc> tell us!
17:14 < jsaw> was on bash-bug mailing list... I'm reading it...
17:14 < mnemoc> oh, i thought it was a reply to you :)
17:14 < jsaw> but I don't know if it is already fixed with the patches
17:20 < jsaw> it seems to be indeed this thread we have problems with.
17:20 < jsaw> http://lists.gnu.org/archive/html/bug-bash/2004-11/msg00366.html
17:20 < jsaw> Thanks, this has been fixed.
17:20 < jsaw> is Chet's answer
17:21 < jsaw> but WHERE *f...*
17:23 < mnemoc> brb
17:24 < rxr> jsaw: obvioiusly not with the official patchset:
17:24 < rxr> $ sh bashcrash.sh
17:24 < rxr> Segmentation fault (core dumped)
17:24 < rxr> PPC t2:HEAD
17:25 < jsaw> yes. also here.
17:25 < jsaw> let's bug chet
17:25 < rxr> I google a bit more
17:26 < rxr> but bug him it would be wise to add the patch to the official set
17:27 < jsaw> done.
17:44 < jsaw> cu later!
17:44 < rxr> cu jsaw
17:52 < rxr> btw. I fix our bash package to use downloaded bash patches
18:05 < CIA-9> rene * r5382 /trunk/package/shells/bash/ (18 files):
18:05 < CIA-9> * converted the bash package to download the official patches in favour
18:05 < CIA-9> to importing them
18:05 < CIA-9> * added another "official" URL
18:46 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: _Ragnar_
18:47 -!- Netsplit over, joins: _Ragnar_
18:48 < CIA-9> rene * r5383 /trunk/package/shells/bash/bash.desc: * added forgotten bash checksums
18:48 -!- Netsplit kornbluth.freenode.net <-> irc.freenode.net quits: _Ragnar_
18:49 -!- Netsplit over, joins: _Ragnar_
19:01 < CIA-9> rene * r5384 /trunk/package/shells/bash/3.0-jobs.patch:
19:01 < CIA-9> * imported some bash-3.0 job-control hacking. Fixes infinte waiting
19:01 < CIA-9> for executed jobs with recentl glibc versions.
19:03 < mnemoc> re
19:04 < rxr> hi mnemoc
19:05 < mnemoc> hi rxr
19:05 < mnemoc> i see hard to defeat t2-sde.org
19:05 < mnemoc> and nobody has voted < 0 for it
19:08 < rxr> so I grab it now?
19:08 < mnemoc> i think yes
19:08 < mnemoc> but don't feel forced :)
19:13 < rxr> hum ...
19:20 < mnemoc> *g*
19:22 < rxr> I querried my DNS manager (quite amusingly the ROCK stable release maintainer) - when he awakes and I got one more discussoin regarding this on phone I'll most probably initiate it around 20:xx maybe around 21:00 ...
19:23 < mnemoc> *G*
19:32 < rxr> susan does like t2-project.org more, too ...
19:32 < rxr> damn - this does make it quite hard ...
19:32 < rxr> every normal use or occatinoal committer I ask favours t2-project. due to liking words more than some akkronym
19:33 < rxr> what to do?
19:36 < CIA-9> jsaw * r5385 /trunk/package/x11/wxwidgets/wxwidgets.conf:
19:36 < CIA-9> * as reported on ROCK-Linux ML, wxwidgets is linked to gtk-1
19:36 < CIA-9> -> so "--enable-gtk2" and a few other options (e.g. opengl)
19:36 < mnemoc> both? :p
19:36 < CIA-9> mnemoc * r5386 /trunk/scripts/Create-ISO: * improved error messages on Create-ISO
19:36 * mnemoc back to the office
19:38 < rxr> jsaw: alive?
19:38 < jsaw> (partially)
19:38 < rxr> jsaw: the bash huy replied to my mail - wihtout mentioning the real fix
19:39 < rxr> has he wrote you? - with more info then to me?
19:39 < rxr> cu mnemoc
19:39 < jsaw> nothing yet
19:42 < rxr> hm
19:43 < rxr> I find this a bit strange
19:43 < rxr> he does not make snapshot or CVS available and does not answer seg-fault bugs that detailed ...
19:43 < rxr> not the best habbit ...
19:54 < jsaw> let's use dash
19:55 < jsaw> <80k stripped
19:56 < rxr> URL ?
19:56 < mnemoc> re
19:56 < rxr> welcome mnemoc
19:57 < mnemoc> thx rxr
19:57 < mnemoc> jsaw: ash for what?
19:58 < rxr> is there something like dash? or is busybox ash just call so when linked with a diet like libc ?
19:59 < mnemoc> afaik we have ash as tarball on download/
20:00 < mnemoc> NetBSD's ash
20:00 -!- lars [~lars@port-212-202-170-162.dynamic.qsc.de] has joined #t2
20:01 < mnemoc> hi lars
20:01 < rxr> hi lars !
20:01 < lars> hi is rene there
20:01 < rxr> greetings!
20:01 < lars> hi pamela told me, you called me by phone
20:01 < rxr> lars: are you at home?
20:01 < rxr> lars: yep
20:02 < rxr> lars: I wanted to call again when you are at home and ada is not complaining about her bed dress anymore ...
20:02 < lars> I am still at work but I am home in about 45 min
20:02 < rxr> hm - qsc IP? you have qsc at hoem ?
20:02 < lars> so you can call at about 8:45?
20:03 < lars> no I am on grum
20:03 < rxr> oh - this is our server
20:03 < rxr> yep - just noticed ... ;-)
20:03 < rxr> lars: I wanted to call anyway to with a happy new year and talk a bit ...
20:03 < rxr> so I think I'll call later if you have some minutes ...
20:03 < lars> perhaps you already noticed that I am there every day for imap tunneling reasons
20:04 < rxr> yeah - sure
20:04 < rxr> but if you are already here you might like to add some lines to the domain discussion ...
20:04 < rxr> as I noted on the list in my last mail real-world people seem to prefer t2-project.org ... in favour of the akkronym one ...
20:05 < rxr> any showstopper argument before we register it (hopefully soon)?
20:05 < lars> I think t2-project sounds good -- perhaps sde is to cryptic
20:06 < rxr> that is most users argument ...
20:07 < lars> ok, I think one more time on the bike on it. I have to go home now since I left home
20:07 < lars> at 8 pm
20:08 < lars> by, hasta luego
20:08 -!- lars [~lars@port-212-202-170-162.dynamic.qsc.de] has quit ["leaving"]
20:12 < mnemoc> rxr: url must be user-friendly :\
20:12 < rxr> mnemoc: yep - so we wanna go -project.org, hm?
20:13 < mnemoc> hm
20:13 < mnemoc> reply doing a counting and argueing that
20:17 < rxr> mail sent to linus and marcelo ...
20:17 < mnemoc> linux-headers?
20:18 < mnemoc> rxr: valentin and you where who voted negative to -project.o what does he think now?
20:21 < rxr> no - driver removal request ...
20:21 < rxr> ;-)
20:21 < rxr> he is on the way home - so maybe he reappears soon
20:22 < rxr> I give -roject +5 now ... ;-)
20:22 -!- sparc-kly [~sparc@64.237.130.154] has joined #t2
20:22 < sparc-kly> t2-project.org MMmMMmM
20:22 < sparc-kly> +5
20:22 < sparc-kly> LOL
20:23 < CIA-9> mnemoc * r5387 /trunk/scripts/Create-ISO: * improved Create-ISO to support directories on output name and iso/ as default.
20:23 < mnemoc> sparc-kly: lol @ what?
20:24 * sparc-kly compiling 2.4.28 in sparc
20:24 < sparc-kly> fail 4 times
20:25 < sparc-kly> damn it
20:25 < mnemoc> o_O
20:25 < valentin> rehi
20:25 < valentin> i still would vote for sde, because we have to burn the term SDE into peoples mind
20:26 < valentin> there are thousands of projects
20:26 < mnemoc> good point
20:26 < valentin> but only 1 sde
20:27 < valentin> so joe-users will have to remember the "sde" thing somehow
20:27 < rxr> arguing this way leads to sde.org ...
20:27 < mnemoc> uhm
20:27 < valentin> hmmm
20:27 < mnemoc> .oO( kernel.org? )o
20:28 < CIA-9> rene * r5388 /trunk/package/kde/kdebindings/kdebindings.conf:
20:28 < CIA-9> * fixed kdebindings extreme memory requirements during compilation by
20:28 < CIA-9> disabling the KDE global --enable-fiinal in the package
20:28 < rxr> Registrant Name:Sanford Dole Ensemble
20:28 < rxr> Registrant Organization:Sanford Dole Ensemble
20:28 < rxr> ^- sde.org is taken ...
20:28 < mnemoc> :)
20:28 < valentin> feared that
20:28 < rxr> Sanford Dole
20:28 < rxr> Musician - Artist
20:29 < rxr> errhm ;-)
20:29 < rxr> mnemoc: I tested the kdebindings fix on my iBook - it is not an untested guess ..
20:30 < mnemoc> rxr: ccache cache size, expert or noexpert?
20:30 < rxr> both user
20:30 < mnemoc> ok
20:31 < mnemoc> http://cxx.uclibc.org/ uhm
20:46 < CIA-9> mnemoc * r5389 /trunk/ (misc/output/config-100.in scripts/config.in): * added a block for noexpert configs
20:47 < CIA-9> mnemoc * r5390 /trunk/package/develop/ (8 files in 2 dirs): * changed ccache and distcc config options to user zone
20:49 < mnemoc> rxr: do you know what happened to the guy playing with uclibc?
20:58 < mnemoc> rxr: the war for your pieces on rock has began
21:36 < jsaw> re
21:47 < mnemoc> re jsaw
21:47 < jsaw> hi mnemoc
21:47 < jsaw> what about a temporary workaround for the bash segfault?
21:49 < rxr> yeah - would love a work around ... if you have one ...
21:49 * rxr away ...
21:49 < rxr> cu
21:50 < rxr> ouhm
21:50 < rxr> got a private mail:
21:50 < rxr> >
21:50 < rxr> > I spent some time debugging it already and need this annoying bug to be fixed ... :-(
21:50 < rxr> Why is it that annoying? BASH_SOURCE, BASH_LINENO, and FUNCNAME are not
21:50 < rxr> intended to be modified by the user. What requires you to do so?
21:50 < rxr> > A verbose explanation would be welcome - a real patch even more. Many thanks!
21:50 < rxr> Two parts: make sure the variables aren't created with the `invisible'
21:50 < rxr> attribute set, and make sure the `non-unsettable' flag is set for
21:50 < rxr> BASH_ARGC, BASH_ARGV, BASH_SOURCE, and BASH_LINENO. The `no-assignment' flag should already be set.
21:50 < rxr> ^- bash authro
21:50 < rxr> author
21:51 < mnemoc> does he sign as "bash author"??
21:51 < CIA-9> jsaw * r5391 /trunk/scripts/functions:
21:51 < CIA-9> * temporary (or permanent?) workaround for bash segfaults
21:51 < CIA-9> with readonly variables (?)
21:51 < rxr> somehow I do not get into such deep conversation with this guy and dislike peoople that are unable to release patches, snapshots, version controlled source or patches
21:51 < rxr> mnemoc: yes
21:51 < rxr> he is the (only one) bash author
21:51 < mnemoc> good title
21:51 < rxr> there does not seem to be a CVS or equivalent source and so on ...
21:51 < rxr> so - me in a hury cu
21:51 < jsaw> cu rxr
21:53 < jsaw> so, now let's see if we can create an exploit for bash 3.0...
21:55 < rxr> ok
21:55 < rxr> still here
21:56 < rxr> lars' wife also votes for -project
21:57 < jsaw> "der Kunde ist König"
21:59 < jsaw> (customer's the king - or similiar)
22:02 < mnemoc> yes
22:24 < valentin> then we have to know who our customers are :/
22:25 < jsaw> well
22:25 < jsaw> hi valentin
22:30 < valentin> hi
22:32 < mnemoc> 34C!!
22:32 < mnemoc> outside, on the "shadows"
22:38 < jsaw> great!
22:38 < CIA-9> jsaw * r5392 /trunk/scripts/functions: * add some more readonly variables for bash-segfault workaroun
22:41 < mnemoc> jsaw: why not applying a .sed file after dumping?
22:45 < jsaw> yep, maybe I should do that.
22:48 < jsaw> otoh it is a workaround...
22:49 < rxr> according to the bash guy it might be longer until we see the fix ...
22:50 < jsaw> see above "(permanent ?)"...
23:13 < CIA-9> mnemoc * r5393 /trunk/package/base/uclibc/patches/ (6 files): * removed zero sized uclibc's patches
23:32 < rxr> jsaw: sorry I do not see a (permanent ?) - but maybe I'm too tired / blind ...
23:32 < jsaw> no problem. It was the CIA-9 message for r5391.
23:34 < rxr> ah - there ;-)
23:38 < mnemoc> jsaw: i still get SSP stuff linking linuxrc when it comes from actually building something
23:39 < jsaw> oh my.
23:39 < mnemoc> jsaw: if i re-run Build-Target it pass
23:44 < jsaw> check if you system has a libgcc with SSP symbols
23:44 < jsaw> first bash patch looks promising
23:45 < jsaw> I patched all dynamic variables to have att_function|att_readonly..., seems to work.
23:45 < rxr> jsaw: what patch?
23:45 < jsaw> my patch
23:46 < jsaw> not yet committed
23:47 < rxr> do you plan to revert the dump_env change then again ?
23:47 < jsaw> yes
23:47 < rxr> do you do as the bash author Chet Ramey suggested?
23:47 < rxr> (see T2 list ...)
23:48 < rxr> btw. did you got a reply ?
23:48 < jsaw> just read your post on irc
23:49 < rxr> ok
23:49 < jsaw> no, I don't do what he suggested.
23:49 * rxr waiting then
23:50 < rxr> was about to prettify your dumping ...
23:50 < rxr> but a rework would be even nicer
23:50 < rxr> damn I'm so hungry
23:50 < jsaw> The variables I patch are dynamic variables. I set the attributes to att_function|att_readonly just like bind_function would do.
23:50 < mnemoc> the only really prettify is a sed patch after dumping :p
23:51 < rxr> do I go out and buy food - ro do I enter the sleep cycle ...
23:51 < rxr> yep
23:51 < rxr> or a declare xyz that does noly yield non-internal varts
23:52 < mnemoc> that sounds even better
23:53 < mnemoc> but's a rework and and not a prettify :p
23:53 < rxr> okook
23:53 < rxr> I do not now if there is some argument to specify this ..
23:53 < rxr> I just man ...
23:54 < jsaw> there's no "internal" attribute... so
23:54 < mnemoc> hehe
23:54 < rxr> I just wait for your patch ...
23:54 < jsaw> but at least declare -p shouldn't dump dynamic variables...
23:55 < rxr> yeah ...
23:55 < rxr> the bash hacker would say "why do you need this" ... ;-)
23:55 < jsaw> however if these are not marked with "att_function" then... :(
23:55 < mnemoc> OT: where has demangler gone on bintuils??
23:55 < rxr> mnemoc: it is in gcc ...
23:55 < rxr> the one in binutils always was obsolete ..
23:55 < mnemoc> rxr: thanks
23:55 < rxr> and redundant
23:55 < mnemoc> sure
23:56 < rxr> and - uhm - out of date - not matching the g++'s used conventions
23:56 < mnemoc> good reasons to be removed
23:59 < mnemoc> can i insert a new line after a match with sed?
23:59 < rxr> afer every match?
23:59 < rxr> a/match/insert ... IIRC
23:59 < mnemoc> can i use \1,\2,... on that?
23:59 < rxr> or s/\(match\)/\n/g
--- Log closed Tue Jan 04 00:00:04 2005