--- Log opened Wed Dec 29 00:00:46 2004 00:00 < rxr> that I'm too unrelaxed to track bash? 00:10 < rxr> #0 hash_search (string=0x1009ceb0 "PS2", table=0x100ce818, flags=0) 00:10 < rxr> at hashlib.c:182 00:10 < rxr> 182 hashlib.c: No such file or directory. 00:10 < rxr> in hashlib.c 00:10 < jsaw> 23:58 < rxr> but the problem needs a fix 00:35 < mnemoc> rxr: :) 00:40 < rxr> ? 00:41 * rxr is to slow to follow this today ... 00:41 < mnemoc> my answer to your comment about berlin's t2 devs 00:44 < rxr> ah ;-) 00:45 < rxr> this works around the segfault in most if not all cases: 00:45 < rxr> - export PROMPT_COMMAND='. debug.buildenv; cd .; unset PROMPT_COMMAND' 00:45 < rxr> - cd $builddir; exec bash 200>> build.pid 00:45 < rxr> + cd $builddir; exec bash --rcfile debug.buildenv 200>> build.pid 00:45 < rxr> I think I apply it for now and try to produce some testcase for the bash developers ... 00:46 < rxr> this environment injections via PROMT_COMMAND seems to be rather hostile ... 00:47 < rxr> nevertheless it should not segfault but might be too bash hacky to track/fix 00:48 < mnemoc> uhm 00:50 < mnemoc> why that evil method was used?? 00:52 < rxr> no idea 00:52 < rxr> too many drugs in austria while coding .... 00:52 < rxr> as usual comment missing ... 00:52 < rxr> I construced a segfault with the cleaner method, too 00:53 < mnemoc> exec bash -x --rcfile debug.buildenv 200>> build.pid ? 00:57 < mnemoc> some other evilness might be segfaulting it 00:57 < rxr> -x does not help 00:57 < rxr> always segfaulting in the internal variable lookup hash function 00:58 < rxr> memory corruptions most probably ... 00:59 < rxr> #0 hash_search (string=0x1009ceac "PS1", table=0x100ce818, flags=0) 00:59 < rxr> at hashlib.c:182 00:59 < mnemoc> $? on PS1? 01:00 < rxr> PS1 - but I guess it can be any variable that cause the crash during lookup 01:09 < rxr> http://www.fsl.cs.sunysb.edu/project-unionfs.html 01:18 < rbarnich> I've installed t2-desktop and I'm having 2 problems. 1st, 2.6.9 hangs during boot and 2nd, after re-running 'stone', 2.4.25 no longer will accept my root password. I've reconfigured several time and no luck. Any ideas? 01:19 < rxr> I saw this passwd in stone failing also on a box ... 01:19 < rxr> just chroot /mnt/target 01:19 < rxr> and run passwd manually 01:19 < rxr> the next release will include a fix 01:20 < rxr> if you have some linux experience you could also boot with init=/bin/bash and then mount / -o remount,rw and run passwd 01:21 < rxr> hm - the (imperfect) PowerPC valgrind does not detect accessing out of bounds memory 01:21 < mnemoc> rxr: bash/bash30-009 <-- with these ugly names and not [D]s? 01:21 < rbarnich> Tnx, I'll try that. 01:21 < rxr> rbarnich: feel free to reask if you need more details ... 01:21 < rbarnich> ok 01:21 < rxr> valgrind does only catch this: 01:21 < rxr> ==4769== 01:21 < rxr> ==4769== Invalid read of size 4 01:21 < rxr> ==4769== at 0x10046DA4: hash_search (hashlib.c:180) 01:21 < rxr> ==4769== by 0x1002D498: hash_lookup (variables.c:1449) 01:21 < rxr> ==4769== by 0x1002D4EC: var_lookup (variables.c:1463) 01:21 < rxr> ==4769== by 0x1002D59C: find_variable_internal (variables.c:1497) 01:21 < rxr> ==4769== Address 0x76692D72 is not stack'd, malloc'd or (recently) free'd 01:21 < rxr> --4769-- adding signal 11 to pending set 01:22 < rxr> this is where i already was 01:24 < rxr> cool to see that the ppc valgrind al least yields some usful results ... 01:24 < mnemoc> wasn't it segfaulting for everything? 01:24 -!- rbarnich [~rbarnich@pcp04150125pcs.sanarb01.mi.comcast.net] has quit ["using sirc version 2.211+KSIRC/1.3.11"] 01:25 < CIA-10> jsaw * r5293 /trunk/package/scientific/root/ (asimage-reg-clobbered.patch root.conf root.desc): 01:25 < CIA-10> * update root (4.00.02 -> 4.02.00) 01:25 < CIA-10> * add in-build patch patch for register clobbered in asimage 01:26 < rxr> mnemoc: you want to [D] the bash patches? 01:27 < CIA-10> jsaw * r5294 /trunk/package/develop/libowfat/libowfat.desc: * update libowfat (0.19.2 -> 0.20) 01:28 < mnemoc> rxr: i want? :p .... well.... adding lot of downloadable patches is a /bad precedent/ and a bit ugly 01:28 < rxr> yes - I just wanted to push this from one box to another for a test ... 01:29 < mnemoc> rxr: i think we can automate this using [D]s and an a list of indexes on an [O] 01:29 < rxr> I 'll convert them to [D] when I patched bash to no longer segfault on the debug.sh use-case .. 01:29 < rxr> no idexes .. 01:29 < rxr> the patch are p0 so we can not use apply_patchfiles .... 01:29 < rxr> but for x in `match_source_file bash30-???` ; do 01:29 < rxr> ... 01:29 < rxr> done 01:29 < rxr> should do ... 01:30 < rxr> it does not need hardcodong ... 01:30 < mnemoc> yes 01:30 < mnemoc> one point to match_source_file :) 01:31 < mnemoc> match_source_file -p bash30-??? 01:36 < mnemoc> jsaw: /TOOLCHAIN/src.groff.1104280175.4757.7f0100/groff-1.19.1: 01:36 < mnemoc> - gcc conftest.c 01:36 < mnemoc> + gcc -Os -pipe -march=pentium4 --mpreferred-stack-boundary=128 conftest.c -s 01:36 < mnemoc> uhm 01:36 < mnemoc> -- 01:36 -!- rbarnich [~rbarnich@pcp04150125pcs.sanarb01.mi.comcast.net] has joined #t2 01:39 < mnemoc> jsaw: conftest.c:1: error: -mpreferred-stack-boundary=128 is not between 2 and 12 <--- the recomendation is flawed :) 01:39 < rbarnich> - yr init suggestion worked, I'm online now! Thanks. 01:39 < rxr> try 7 01:40 < rxr> rbarnich: cool - have fun ;-) 01:40 < rxr> mnemoc: try 7 since 2^7 == 128 ... 01:40 < rxr> maybe they meant that ? 01:40 < mnemoc> rxr: good guess 01:42 < mnemoc> == 12/28/04 21:42:33 =[3]=> Finished building package groff. 01:42 < mnemoc> -mpreferred-stack-boundary=128 added only for groff/pentium4 01:43 -!- rbarnich [~rbarnich@pcp04150125pcs.sanarb01.mi.comcast.net] has quit ["using sirc version 2.211+KSIRC/1.3.11"] 01:43 < rxr> no 01:43 -!- rbarnich [~rbarnich@pcp04150125pcs.sanarb01.mi.comcast.net] has joined #t2 01:43 < mnemoc> -mpreferred-stack-boundary=7 i mean 01:43 < rxr> better add it globally in the pentium4 options in architecture / x86 / gcc-options 01:43 < rxr> s.th. 01:43 < mnemoc> ok 01:49 < mnemoc> == 12/28/04 21:49:40 =[3]=> Finished building package groff. 01:52 < rxr> hm - my handcompiled bash binaries do not segfault including the same files .... 01:52 < rxr> damn 01:52 < mnemoc> something else in the env? 01:53 < CIA-10> mnemoc * r5295 /trunk/ (architecture/x86/gcc-options package/base/groff/groff.conf): * added proper workaround for gcc bug #15617 (-mpreferred-stack-boundary=7 for pentium4) 01:58 < rxr> ;-) 01:58 < mnemoc> descriptive message :) 02:06 < mnemoc> rxr: did you got answer from sourceforge about dl.sf.net? 02:07 < rxr> no 02:07 < rxr> seems bash only segfault so far when build with --without-gnu-malloc as we do 02:07 < rxr> my builds without any config options do not crash 02:07 < mnemoc> bash-malloc is kind of broken? 02:08 < rxr> bash-malloc does seem to hide the bug ... 02:10 < mnemoc> what sf mirror are you using to fix this? i used unc the last time 02:12 < rxr> then let's use unc until we have a solution 02:14 < rxr> the hash table corruptions 02:14 < rxr> corrupts even 02:14 < mnemoc> i like adding a workaround on Download more that patching every [D] we update 02:15 < rxr> hum hm ... 02:15 < rxr> well both are not that nice ... 02:15 < mnemoc> we do very intensive usage of bash's variables hash table 02:15 < mnemoc> first is definitive an easier to revert 02:15 < mnemoc> and* 02:17 < mnemoc> and having http://foo.dl.. on [D]s interfere with [CV-URL] generator for sourceforge :) 02:18 < mnemoc> but i'll not try to convise you while you hunt hash problem :) 02:19 < CIA-10> mnemoc * r5296 /trunk/package/python/pysqlite/pysqlite.desc: * fixed pysqlite download location due to changes on dl.sf.net 02:23 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has left #t2 ["Leaving"] 02:28 < rxr> this damn gd 02:28 < rxr> it now claims again that s simple symbol is not present in the program when it clearly ist ... 02:28 < rxr> so this starts to happen for simple C code, too - not only my big C++ files ... 02:29 < rxr> annoying this is 02:35 < rxr> hm 02:35 < rxr> $12 = {next = 0x100d5d08, key = 0x100ba2f8 "ROCKCFG_CREATE_TARBZ2", 02:35 < rxr> data = 0x100d76c8, khash = 1973740136, times_found = 0} 02:35 < rxr> (gdb) p * (table->bucket_array[bucket]->next->next->next->next->next->next->next) 02:35 < rxr> $13 = {next = 0x100bec50, key = 0x0, data = 0x1, khash = 0, 02:35 < rxr> times_found = 269216856} 02:35 < rxr> (gdb) p * (table->bucket_array[bucket]->next->next->next->next->next->next->next->next) 02:35 < rxr> $14 = {next = 0x32383038, key = 0x19
, 02:35 < rxr> data = 0x100bce80, khash = 1, times_found = 269587904} 02:36 < mnemoc> # ./debug.sh 02:36 < mnemoc> free(): invalid pointer 0x8101270! 02:36 < mnemoc> debug-gsmlib:[src.gsmlib.1104283943.15997.0]# 02:36 < rxr> mnemoc: where id this free() from ? 02:37 < mnemoc> no idea yet 02:43 -!- rxr_ [~rene@p213.54.199.47.tisdip.tiscali.de] has joined #t2 02:43 -!- Topic for #t2: T2 | 2.1.0-beta2 RELEASED | The next generation of System Development Enviroments (SDE) | http://www.exactcode.de/t2 02:43 -!- Topic set by mnemoc [] [Tue Dec 28 14:55:44 2004] 02:43 [Users #t2] 02:43 [ _Ragnar_] [ daja77] [ jsaw ] [ rbarnich] [ rxr_ ] [ valentin] 02:43 [ CIA-10 ] [ ideal ] [ mnemoc] [ rxr ] [ sparc-kly] 02:43 -!- Irssi: #t2: Total of 11 nicks [0 ops, 0 halfops, 0 voices, 11 normal] 02:43 -!- Channel #t2 created Sun Aug 8 21:15:33 2004 02:43 -!- [freenode-info] help freenode weed out clonebots, please register your IRC nick and auto-identify: http://freenode.net/faq.shtml#nicksetup 02:43 -!- Irssi: Join to #t2 was synced in 10 secs 02:44 < mnemoc> 01:42:48 open("/etc/inputrc", O_RDONLY|O_LARGEFILE) = 5 <0.000012> 02:44 < mnemoc> 01:42:48 read(5, "# --- ROCK-COPYRIGHT-NOTE-BEGIN "..., 1261) = 1261 <0.000021> 02:44 < mnemoc> 01:42:48 close(5) = 0 <0.000009> 02:44 < mnemoc> 01:42:48 fstat64(2, {st_dev=makedev(0, 7), st_ino=4, st_mode=S_IFCHR|0620, st_nlink=1, st_uid=0, st_gid=5, st_blksize=1024, st_blocks=0, st_rdev=makedev(136, 2), st_at 02:44 < mnemoc> ime=2004/12/29-01:42:47, st_mtime=2004/12/29-01:42:47, st_ctime=2004/12/25-16:42:09}) = 0 <0.000009> 02:44 < mnemoc> 01:42:48 old_mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x40014000 <0.000014> 02:44 < mnemoc> 01:42:48 write(2, "free(): invalid pointer 0x80ff92"..., 35) = 35 <0.000331> 02:44 < mnemoc> flood :) 02:46 -!- rxr [~rene@p213.54.195.155.tisdip.tiscali.de] has quit [Read error: 60 (Operation timed out)] 02:48 < mnemoc> rxr_: ping 02:54 < mnemoc> rxr_: ping :( 03:13 < mnemoc> rxr_: dead? 03:20 < rxr_> pong 03:20 -!- You're now known as rxr 03:20 < mnemoc> =) 03:20 < CIA-10> mnemoc * r5297 /trunk/package/archiver/cpio/no_rmt.patch: * fixed share between cpio and tar (rmt) 03:21 < mnemoc> how can i improve what i queried you? 03:22 < mnemoc> only rebuilding bash with debug? 03:22 < rxr> building bash with debug symbols ... 03:23 < rxr> strange that error 03:23 < rxr> and mine, too 03:23 < rxr> $ s 03:23 < rxr> bash: s: command not found 03:23 < rxr> $ q 03:23 < rxr> bash: q: command not found 03:23 < rxr> $ w 03:23 < rxr> Segmentation fault (core dumped) 03:23 < rxr> something in this hashes corrupts ... 03:23 < mnemoc> outch 03:23 < mnemoc> ROCKCFG_DEBUG='1' 03:24 < rxr> yep 03:24 < mnemoc> # ./scripts/Build-Target -cfg full -job 3-bash 03:24 < rxr> ack 03:25 < mnemoc> Creating isolinux setup: 03:25 < mnemoc> -> Extracting isolinux boot loader. 03:25 < mnemoc> tar: -O: Not found in archive 03:25 < mnemoc> tar: Error exit delayed from previous errors 03:25 < mnemoc> i randomly get that or _ssp undefines building linuxrc 03:26 < rxr> hm 03:27 < mnemoc> nice, isn't it? 03:28 < rxr> yeah 03:28 < rxr> as is bash 03:28 < mnemoc> when do-nothing rebuild i get tar -O, when building i get _ssp 03:28 < rxr> reducing my testcase fiel I know even get the egfault during bash stgartup 03:28 < mnemoc> only $tooldir changes 03:28 < mnemoc> good 03:30 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has joined #t2 03:31 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has quit [Remote closed the connection] 03:31 < mnemoc> rxr: i get files and lines now 03:31 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has joined #t2 03:32 -!- rbarnich [~rbarnich@pcp04150125pcs.sanarb01.mi.comcast.net] has quit ["using sirc version 2.211+KSIRC/1.3.11"] 03:36 < rxr> dre 03:36 < rxr> damn w-lan here 03:36 < rxr> I know go cases (reducing my testcase) where I get the free error - or even crashes in free ... 03:37 < rxr> bash 3.0 seems to have some big problems with memory management or some random memory corruption 03:37 < mnemoc> you got the free i got? 03:38 < mnemoc> -> Extracting isolinux boot loader. 03:38 < mnemoc> +++ type -p tar 03:38 < mnemoc> /usr/src/building/t2/test/build/boot2-2.1.0-beta3-x86-pentium-mmx-32-bootdisk-expert/TOOLCHAIN/tools.cross/bin/tar 03:38 < mnemoc> +++ tar --version 03:38 < mnemoc> tar (GNU tar) 1.15.1 03:38 < mnemoc> +++ tar --use-compress-program=bzip2 -xf /usr/src/building/t2/test/download/mirror/s/syslinux-2.13.tar.bz2 syslinux-2.13/isolinux.bin -O 03:38 < mnemoc> tar: -O: Not found in archive 03:38 < mnemoc> tar: Error exit delayed from previous errors 03:40 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has quit [Remote closed the connection] 03:41 < mnemoc> what is -O doing that far? 03:41 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has joined #t2 03:45 < CIA-10> mnemoc * r5298 /trunk/target/bootdisk/x86/build.sh: * fixed bootdisk's build.sh to call tar with -O in the begining and not as last arg. 03:51 < rxr> ouhm, 03:54 < rxr> and now thunderbird crashed - what a day .. 03:57 < CIA-10> mnemoc * r5299 /trunk/package/base/lsof/lsof.conf: * fixed lsof to not use $taropt to untar .tar file 04:00 < rxr> ouhm - that might have been me .. 04:00 < rxr> :-( 04:01 < mnemoc> smile :) 04:01 < mnemoc> now it's fixed 04:05 < rxr> http://linuxfromscratch.org/pipermail/lfs-dev/2002-February/023380.html 04:06 < mnemoc> how can i tell gdb where the sources are? 04:07 < mnemoc> uhm 04:08 < rxr> no idea - I usually cd into the sourcedir ... 04:08 < mnemoc> i need bash sources to debug gsmlib's debug.sh :) 04:09 < mnemoc> directory statement 04:14 < mnemoc> why they contrast --with-bash-malloc and -without-gnu-malloc ? A and ! (!A) are the same to me 04:18 < rxr> hm? 04:19 < mnemoc> without gnu-malloc _means_ with bash-malloc 04:19 < mnemoc> at least to me 04:20 < mnemoc> but they treat them as opossite while talking 04:21 < mnemoc> i would understand --with-gnu-malloc vs. --with-bash-malloc 04:21 < mnemoc> or --with-gnu-malloc vs. --without-gnu-malloc 04:22 < mnemoc> but --with-A and --without-B where B is !A is kind of senseless.... what am i missing? 04:23 < mnemoc> OT: 63000 dead people counting on southeast asia 04:29 < mnemoc> http://linuxfromscratch.org/pipermail/lfs-dev/2002-February/023427.html <--- uff 04:30 < mnemoc> What's with the ^^^^^ ? This option DOES exist. There are THREE(!!) 04:30 < mnemoc> mallocs to choose from when compiling bash. 04:30 < mnemoc> uhm 04:31 < mnemoc> Which version of bash are you using? Maybe it's been removed in 2.05a. 04:31 < mnemoc> Anyway, it creates a bash that segfaults. 04:32 < mnemoc> what if --without-gnu-malloc on bash3.0 uses glibc-malloc instead of bash-malloc? 04:34 < mnemoc> uhm... 04:36 < mnemoc> rxr? 04:36 < mnemoc> debug-gsmlib:[TOOLCHAIN]# pwd 04:36 < mnemoc> /TOOLCHAIN/src.gsmlib.1104283943.15997.0 04:37 < mnemoc> do you understand why i get that on prompt? 04:37 < mnemoc> (using --rcfile instead of clifford thing) 04:40 < rxr> sorry 04:40 < rxr> bash and gnu malloc are the same 04:40 < mnemoc> o_O 04:41 < mnemoc> so the thing is bash vs. glibc ? 04:42 < rxr> yes - AFAICS 04:42 < rxr> but I think it is a bash bug that only shows up using glibc's malloc 04:45 < mnemoc> we are using: --without-gnu-malloc 04:45 < rxr> fecore code: 04:45 < rxr> if ! autoconf; then 04:45 < rxr> # Yuck. We're using autoconf 2.1x. 04:45 < rxr> ln -s /bin/true autoconf 04:45 < rxr> mnemoc: yes, which is the same as --without-bash-malloc 04:46 < mnemoc> ln -s /bin/true autoconf <-- wtf? 04:46 < rxr> note the Yuck ;-) 04:46 < rxr> this is from the fedora bash package 04:46 < rxr> they use: 04:46 < rxr> --with-bash-malloc=no --with-af 04:47 < mnemoc> --with-af? 04:48 < rxr> +s 04:48 < rxr> with-afs ... 04:48 < mnemoc> :) 04:48 < mnemoc> on bash3 too? 04:48 < rxr> --with-afs if you are running AFS 04:48 < rxr> --with-gnu-malloc synonym for --with-bash-malloc 04:49 < mnemoc> aha 04:49 < mnemoc> but afaik --with-bash-malloc=no is ideantical tu --without-bash-malloc 04:49 < mnemoc> to* 04:49 < CIA-10> rene * r5300 /trunk/scripts/Build-Pkg: * nicer debug.sh debug.buildenv injection 04:49 < mnemoc> debug-gsmlib:[TOOLCHAIN]# pwd 04:49 < mnemoc> /TOOLCHAIN/src.gsmlib.1104283943.15997.0 04:49 < rxr> yes 04:50 < mnemoc> ^^^ do you get that misbehaviour? 04:50 < CIA-10> rene * r5301 /trunk/package/shells/bash/bash.conf: 04:50 < CIA-10> * converted bash.conf to use --without-bash-malloc instead of the older 04:50 < CIA-10> and less verbose --without-gnu-malloc 04:51 < rxr> hm - let me take a look 04:54 < rxr> hm - I currently only get segfaults .. 04:55 < rxr> mnemoc: which misbehaviour? 04:56 < mnemoc> using clifford's debug_x.sh: debug-gsmlib:[src.gsmlib.1104283943.15997.0] 04:56 < mnemoc> using --rcfile: debug-gsmlib:[TOOLCHAIN]# 04:57 < mnemoc> for both $PWD=/TOOLCHAIN/src.gsmlib.1104283943.15997.0 04:59 < rxr> hm - interesting 04:59 < rxr> I missed to notice that ... 04:59 < rxr> need to take a look tomorrow ... 04:59 < rxr> I take sleop now.. 05:01 < mnemoc> me too 05:01 < mnemoc> gn8 rxr, see you tomorrow 05:04 < rxr> cu 07:25 -!- sparc-kly_ [~mubex@66.50.161.111] has joined #t2 07:27 -!- sparc-kly [~mubex@66.50.161.111] has quit [Read error: 104 (Connection reset by peer)] 08:05 -!- mipe [~mipe@dsl10040.japo.fi] has joined #t2 09:26 -!- mipe_ [~mipe@dsl10040.japo.fi] has joined #t2 09:35 -!- mipe [~mipe@dsl10040.japo.fi] has quit [Read error: 60 (Operation timed out)] 09:45 -!- mipe_ is now known as mipe 09:46 < mipe> anyone awake? 10:09 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has quit ["FreeBSD N0V4K.Zero 5.3-RELEASE FreeBSD 5.3-RELEASE Chao"] 10:29 -!- jonhbruce [~john@host81-157-242-89.range81-157.btcentralplus.com] has joined #t2 10:33 < jonhbruce> Seasons Greetings, have tried to use new script, but get SED error. Is College Linux likely to throw up any more errors due to elderly nature? 10:35 < jonhbruce> sed version 3.02 10:38 < mipe> segfaults? i had many sed segfaults when i used fc3 as host system 10:39 < jonhbruce> nope, just plain unrecognised version of SED 10:43 < jonhbruce> sorry, ERROR_LOG gives "invalid option -- i". 10:46 < mipe> hmm. i got same kind of errors but not from sed, didnt find any solution for it. you could ask the core devs when they wake up 10:47 < jonhbruce> I'm updating to slackware 10.1 in hope this gets around it, Thanks. 10:47 < mipe> same kind of error, meaning i get invalid option from even if it is correct option but it just didnt work under t2 building. 10:49 -!- jonhbruce [~john@host81-157-242-89.range81-157.btcentralplus.com] has quit ["Leaving"] 10:53 -!- N0V4K [~N0V4K@246.Red-80-33-149.pooles.rima-tde.net] has joined #t2 10:53 * N0V4K is away: Sobando :D 11:30 -!- mipe [~mipe@dsl10040.japo.fi] has quit ["using sirc version 2.211+KSIRC/1.3.11"] 11:55 -!- mipe [~mipe@dsl10040.japo.fi] has joined #t2 11:57 < mipe> awake? 12:04 -!- ideal [~idealm@211.100.227.107] has quit ["Leaving"] 12:26 < mnemoc> moin! 12:30 < mipe> hi mnemoc 12:31 < mnemoc> hi mipe 12:31 < mipe> got some problems with gcc and linux26 :) 12:32 < mipe> on linux26 there has to be config_x86_ and something enabled because of the sub build of rivatv,which isnt enabled by default, but i think it should be since build fails with out it 12:33 < mipe> and on gcc i get error 12:33 < mipe> -c ../../gcc/crtstuff.c -DCRT_BEGIN \ 12:33 < mipe> -o crtbegin.o 12:33 < mipe> In file included from ../../gcc/crtstuff.c:62: 12:33 < mipe> ../../gcc/tsystem.h:79:19: stdio.h: No such file or directory 12:33 < mipe> ../../gcc/tsystem.h:82:23: sys/types.h: No such file or directory 12:33 < mipe> ../../gcc/tsystem.h:85:19: errno.h: No such file or directory 12:33 < mipe> ../../gcc/tsystem.h:92:20: string.h: No such file or directory 12:33 < mipe> ../../gcc/tsystem.h:93:20: stdlib.h: No such file or directory 12:33 < mipe> ../../gcc/tsystem.h:94:20: unistd.h: No such file or directory 12:33 < mipe> ../../gcc/tsystem.h:100:18: time.h: No such file or directory 12:33 < mipe> make[1]: *** [crtbegin.o] Error 1 12:34 < mipe> when doing amd64 build, linux26 i just want to recompile/update 12:41 < mipe> linux26 error is that it needs CONFIG_X86_CMPXCHG which i have in linux.cfg (setted by Config) but it still fails to build because build doesnt turn it on 12:43 -!- mnemoc_ [~amery@200.75.27.14] has joined #t2 12:44 < mipe> crap,i guess you didnt see the posts 12:44 < mipe> :) 12:46 -!- sparc-kly_ is now known as sparc-kly 12:49 -!- mnemoc [~amery@200.75.27.20] has quit [Read error: 110 (Connection timed out)] 12:49 -!- mnemoc_ is now known as mnemoc 13:20 < mnemoc> mipe: nope, i didn't :) 13:20 < mipe> :) ok,will copy paste. then going for dinner 13:21 < mipe> got some problems with gcc and linux26 :) 13:21 < mipe> [13:32:49] on linux26 there has to be config_x86_ and something enabled because of the sub build of rivatv,which isnt enabled by default, but i think it should be since build fails with out it 13:21 < mipe> [13:33:37] and on gcc i get error 13:21 < mipe> [13:33:43] -c ../../gcc/crtstuff.c -DCRT_BEGIN \ 13:21 < mipe> [13:33:43] -o crtbegin.o 13:21 < mipe> [13:33:43] In file included from ../../gcc/crtstuff.c:62: 13:21 < mipe> [13:33:44] ../../gcc/tsystem.h:79:19: stdio.h: No such file or directory 13:21 < mipe> [13:33:44] ../../gcc/tsystem.h:82:23: sys/types.h: No such file or directory 13:21 < mipe> [13:33:45] ../../gcc/tsystem.h:85:19: errno.h: No such file or directory 13:21 < mipe> [13:33:45] ../../gcc/tsystem.h:92:20: string.h: No such file or directory 13:21 < mipe> [13:33:46] ../../gcc/tsystem.h:93:20: stdlib.h: No such file or directory 13:21 < mipe> [13:33:46] ../../gcc/tsystem.h:94:20: unistd.h: No such file or directory 13:21 < mipe> [13:33:47] ../../gcc/tsystem.h:100:18: time.h: No such file or directory 13:21 < mipe> [13:33:47] make[1]: *** [crtbegin.o] Error 1 13:21 < mipe> [13:34:14] when doing amd64 build, linux26 i just want to recompile/update 13:21 < mipe> [13:41:13] linux26 error is that it needs CONFIG_X86_CMPXCHG which i have in linux.cfg (setted by Config) but it still fails to build because build doesnt turn it on 13:21 * mipe dinner,back bit later 13:21 < mnemoc> cu mipe 13:23 < mnemoc> mipe: rivatv _is_ built as module 13:23 < mnemoc> ./lib/modules/2.6.10-dist/kernel/drivers/media/video/rivatv.ko 13:23 < mnemoc> ./lib/modules/2.6.10-dist/kernel/drivers/video/riva/rivafb.ko 13:23 < mnemoc> ./lib/modules/2.4.28-dist/kernel/drivers/media/video/rivatv.o 13:23 < mnemoc> ./lib/modules/2.4.28-dist/kernel/drivers/video/riva/rivafb.o 13:24 < mnemoc> rxr: there is a leak building 5-kdebindings here (p4) 13:29 < mnemoc> rxr: it has passed two hours after "x_8.cpp:2379: warning: unused parameter 'args'" with 700m used and 0.0% of CPU 13:30 < mnemoc> rxr: executing: /usr/libexec/gcc/i686-unknown-linux-gnu/3.4.3/cc1plus-quiet-I.-I.-I../..-I./..-I/opt/kde3/include-I/opt/qt33/include-I/usr/X11R6/include-MD.libs/libsmokekde_la.all_cpp.d-MF.deps/libsmokekde_la.all_cpp.Tpo-MP-MTlibsmokekde_la.all_cpp.lo-MQ.libs/libsmokekde_la.all_cpp.o-D_GNU_SOURCE-DHAVE_CONFIG_H-DQT_THREAD_SUPPORT-D_REENTRANT-D_XOPEN_SOURCE=500-D_BSD_SOURCE-DPIClibsmokekde_la.all_cpp.cpp-quiet-dumpbaselibsmokekde_la.all_cpp.cpp-march=pentiumpro-ansi 13:36 < mipe> [14:23:39] mipe: rivatv is built as module 13:37 < mipe> oh sorry,it actually finished sub build rivatv and the error i get is from sub build xorg 13:37 < mipe> heres the error 13:37 < mnemoc> mipe: xorg or kernel? 13:38 < mipe> when going Build-Pkg linux26, it does several sub builds,nvidia,rivatv and now erroring at xorg 13:38 < mipe> error is src.linux26.1104318207.4927.7f0100/xc/extras/drm/linux/ 13:38 < mipe> Makefile:284: *** CONFIG_X86_CMPXCHG needs to be enabled in the kernel. Stop. 13:39 * mipe afk 13:39 < mnemoc> mipe: you can enable CONFIG_X86_CMPXCHG adding the rule on Config's kernel config. 13:40 < mnemoc> but i think that may be set for default, rxr? 13:40 < mipe> yeah,was just writing a message about it. i have that line + =y but the build doesnt set it. should i change it to m? 13:40 < mnemoc> did you rebuild linux-header? 13:41 < mipe> nope :) 13:41 < mnemoc> :) 14:11 < mnemoc> rxr: cpp1plus still leaking/froken for kdebindings...... have you experiences anything similar? 14:12 < mnemoc> yahoo!! /google cc1plus kdebindings ---> dl.exactcode.de/irclogs on first match 14:13 < daja77> haha! 14:14 < mnemoc> hi daja77 14:14 < mnemoc> ./debug.sh 14:14 < mnemoc> ./debug.sh: line 3: 10794 Segmentation fault 14:14 < mnemoc> i hate this 14:14 < daja77> O_o 14:14 < daja77> never had that 14:14 < mnemoc> bash3 produce this quite regulary here 14:15 < daja77> ah that might be a reason why we delayed bash3 14:17 < mnemoc> rene was the whole day (yestarday) tracing it on hash tables and glibc-malloc 14:23 < mnemoc> http://lkml.org/lkml/2004/8/24/118 <--- i'm not the only one with problems with kdebindings 14:33 < mipe> hmm,how do i rebuild linux-header, when i do ./scripts/Build-Pkg linux-header it gives me error: 14:34 < mipe> ./scripts/Build-Pkg: line 842: /t2_src/t2/t2-trunk/build/default-2.1.0-beta3-x86-generic/var/adm/logs/9-linux-header.out: No such file or directory 14:34 < mipe> cp: cannot stat `/t2_src/t2/t2-trunk/build/default-2.1.0-beta3-x86-generic/var/adm/logs/9-linux-header.out': No such file or directory 14:34 < mipe> grep: /t2_src/t2/t2-trunk/build/default-2.1.0-beta3-x86-generic/var/adm/logs/9-linux-header.out: No such file or directory 14:34 < mipe> -> $root/var/adm/logs/9-linux-header.out -> 9-linux-header.err 14:34 < mipe> mv: cannot stat `/t2_src/t2/t2-trunk/build/default-2.1.0-beta3-x86-generic/var/adm/logs/9-linux-header.out': No such file or directory 14:36 < mnemoc> mipe: never call Build-Pkg by hand 14:37 < mnemoc> mipe: ./scripts/Build-Target -job 1-linux-header 14:37 < mnemoc> mipe: that will rebuild linux-header on stage 1 for config=default 14:46 < mipe> oh,i thought it builds the package for distro i'm building. i want to upgrade stuff. kernel and other packages 14:46 < mipe> for system in currently use 14:46 < mnemoc> Emerge-Pkg does that 14:47 < mipe> but it gives same error like Build-Pkg 14:47 < mnemoc> yes, because in that case it does the same :\ 14:49 < mnemoc> mipe: df /t2_src/t2/t2-trunk/build/ 14:50 < mipe> df? 188k total size for the dir 14:51 < mnemoc> does it has free space? 14:51 < mipe> 60G :) 14:52 < mnemoc> do you have errors _above_ that? 14:52 < mipe> nope 14:54 < mnemoc> asume your linux-header built fine then 14:55 < mipe> nope,heres the complete output 14:55 < mipe> > Reading package configuration from package directory. 14:55 < mipe> -> Preparing build in src.linux-header.1104328399.4979.7f0100 14:55 < mipe> -> Creating backup of old package data. 14:55 < mipe> -> Building. Writing output to $root/var/adm/logs/9-linux-header.out 14:55 < mipe> ./scripts/Build-Pkg: line 842: /t2_src/t2/t2-trunk/build/default-2.1.0-beta3-x86-generic-expert/var/adm/logs/9-linux-header.out: No such file or directory 14:55 < mipe> cp: cannot stat `/t2_src/t2/t2-trunk/build/default-2.1.0-beta3-x86-generic-expert/var/adm/logs/9-linux-header.out': No such file or directory 14:55 < mipe> grep: /t2_src/t2/t2-trunk/build/default-2.1.0-beta3-x86-generic-expert/var/adm/logs/9-linux-header.out: No such file or directory 14:55 < mipe> -> $root/var/adm/logs/9-linux-header.out -> 9-linux-header.err 14:55 < mipe> mv: cannot stat `/t2_src/t2/t2-trunk/build/default-2.1.0-beta3-x86-generic-expert/var/adm/logs/9-linux-header.out': No such file or directory 14:55 < mipe> == 12/29/04 15:53:25 =[9]=> Aborted building package linux-header. 14:58 < mnemoc> the weird part is that Emerge-Pkg should have used / as $root 14:58 < mipe> oh,never mind about the error. my own mistake, but if i make Emerge-Pkg linux26, i assume it upgrades the kernel? and uses the kernel.cfg made by Config to include kernel options? 14:58 < mipe> yes 14:58 < mnemoc> and the log whould be at /var/adm/logs/ 9-linux-header.out 14:58 < mnemoc> instead of /t2_src/t2/t2-trunk/.... 14:59 < mipe> yes. but do i need to rebuild linux-header if i want to rebuild kernel? and nvidia package 14:59 < mipe> upgrade kernel i mean 15:02 < rxr> moin 15:02 < mnemoc> moin rxr! 15:02 < mnemoc> mipe: afaik nvidia package will need kernel headers 15:03 < mipe> oh yeah 15:15 < mipe> and my amd64 build gives error on 0-gcc,error is: 15:15 < mipe> -a-time -fno-omit-frame-pointer -fno-asynchronous-unwind-tables \ 15:15 < mipe> -c ../../gcc/crtstuff.c -DCRT_BEGIN \ 15:15 < mipe> -o crtbegin.o 15:15 < mipe> In file included from ../../gcc/crtstuff.c:62: 15:15 < mipe> ../../gcc/tsystem.h:79:19: stdio.h: No such file or directory 15:15 < mipe> ../../gcc/tsystem.h:82:23: sys/types.h: No such file or directory 15:15 < mipe> ../../gcc/tsystem.h:85:19: errno.h: No such file or directory 15:15 < mipe> ../../gcc/tsystem.h:92:20: string.h: No such file or directory 15:15 < mipe> ../../gcc/tsystem.h:93:20: stdlib.h: No such file or directory 15:15 < mipe> ../../gcc/tsystem.h:94:20: unistd.h: No such file or directory 15:15 < mipe> ../../gcc/tsystem.h:100:18: time.h: No such file or directory 15:15 < mipe> make[1]: *** [crtbegin.o] Error 1 15:15 < mipe> make[1]: Leaving directory `/t2_src/t2/t2-trunk/src.gcc.1104329099.1507.7f0100/g 15:15 < mipe> cc-3.4.3/objdir/gcc 15:16 < mnemoc> mipe: are you using current nvidia package from trunk:HEAD? 15:16 < mipe> mnemoc:yep 15:17 < mnemoc> rxr: here? any idea about ^^^^ 15:17 < mipe> and the nvidia error is: 15:17 < mipe> Uncompressing NVIDIA Accelerated Graphics Driver for Linux-x86 1.0-6629......... 15:17 < mipe> ................................................................................ 15:17 < mipe> ................................................................................ 15:17 < mipe> ................................ 15:17 < mipe> `libXvMCNVIDIA.a' -> `/usr/X11R6/lib/libXvMCNVIDIA.a' 15:17 < mipe> `libXvMCNVIDIA.so.1.0.6629' -> `/usr/X11R6/lib/libXvMCNVIDIA.so.1.0.6629' 15:17 < mipe> cp: omitting directory `drivers' 15:17 < mipe> No 9-nvidia.log file. 15:17 < mipe> --- BUILD ERROR --- 15:18 < rxr> hm 15:18 < rxr> .oO 15:18 < rxr> I fix this 15:19 < mnemoc> past or future? 15:19 < mipe> nvidia package? 15:19 < rxr> yes - simple present 15:19 < daja77> looks like a broken download 15:19 < rxr> nope 15:19 < mnemoc> cp -r :) 15:20 < rxr> looks like I wrote an imperfect script 15:20 < daja77> :D 15:20 < mnemoc> :D 15:25 < rxr> mipe: could you retry? 15:25 < rxr> oh - mom 15:25 -!- madtux [~mike@200.91.101.102] has joined #t2 15:25 < rxr> now commit finished ,-) 15:25 < mnemoc> commit first :) 15:25 < rxr> moin madtux 15:25 < madtux> hello. 15:25 < mnemoc> hi madtux 15:25 < CIA-10> rene * r5302 /trunk/package/x11/nvidia/nvidia.conf: * fixed (?) nvidia file installation 15:25 < mnemoc> mipe: svn up :p 15:25 < madtux> mnemoc: thx for the rmt thing. 15:25 < mnemoc> madtux: :D now it has a comment 15:26 < rxr> what? 15:26 < madtux> mmm.. still fighting with nvidia 15:26 < mnemoc> rxr: what what? 15:27 < madtux> ok minimal built flawless.. minimal desktop and minimal+xorg still running 15:27 < rxr> "rmt thing" 15:27 < rxr> I can not follow ... 15:28 < madtux> currently building router target.. i think its fixed now 15:28 < mnemoc> rxr: when madtux updated cpio he deleted the no_rmt patch because it didn't applyed 15:28 < mnemoc> rxr: and we didn't know why that was there 15:28 < madtux> and there was no note of why it was needed 15:28 < mipe> rxr:nvidia works now 15:28 < mnemoc> rxr: but it was due to a share with tar, so i re patched it with a comment saying why 15:29 < rxr> ah this rmt thing ... 15:29 < rxr> saw it ... 15:29 < madtux> i should say that in the future we should add a comment to the patches that are required to solve conflicts with other packages.. so when they are updated we actually keep them .. and make them work if it is the case 15:29 < rxr> mnemoc: yes. I tried to enforce this in rock times but failed 15:29 < madtux> good day mipe 15:29 < mnemoc> any patch needs a comment telling why it exists 15:29 < madtux> ack 15:29 < rxr> when you look at most if not all new t2 patches you see a comment with author on top 15:30 < rxr> mnemoc: madtux: full ACK 15:30 < madtux> rxr: are we ever going to do anothger reference build? i believe u told me u had a better idea to replace this? 15:33 < mnemoc> i'm doing a generic build from where i can grab .cache files 15:33 < mnemoc> but i'm stoped on kdebindings 15:33 < mnemoc> which leaks 15:33 < daja77> afaik nothing depends on kdebindings, so there should be no problem with that 15:34 < madtux> hiho daja77 15:34 < mnemoc> daja77: thanks, i'll flag it as broken then 15:34 < madtux> now do we need / want reference? 15:35 < madtux> can we do this different? 15:35 < madtux> ideas? suggestions? 15:35 < mnemoc> reference is a generic which copy .cache files into package dirs 15:36 < madtux> the main reason that borke router target btw is the fact that we change repositories names and moved packages around and did not provide updated cache files 15:36 < daja77> mnemoc: it was broken due ruby, right? 15:36 < mnemoc> we need .cache files to see dependencies 15:36 < mnemoc> daja77: no, cc1plus had /problems/ compiling a .cpp 15:36 < daja77> oh 15:37 < mnemoc> http://lkml.org/lkml/2004/8/24/118 <--- this 15:37 < rxr> madtux: .cache discussion later 15:37 < rxr> lunch here now .... 15:37 < madtux> rxr: enjoy lunch 15:37 < mnemoc> mine ate 800m of memory with 0.0 cpu usage 15:37 < mnemoc> rxr: ok, discussion postponed 15:38 < madtux> mnemoc: i know we need them... but then again i wonder if there is other way to do it.. anyways lets stand by on the tpic so we wait for rxr 15:38 < mnemoc> madtux: however we fix it it can't go into 2.1 :) 15:39 < madtux> tststst!!!!!!!!! 15:39 < madtux> mnemoc: ok. 15:39 < mnemoc> :) 15:40 < madtux> so when are u going to release some mnemosyne isos? 15:40 < madtux> we need much more isos than just desktop 15:40 < madtux> :) 15:40 < madtux> as the stable maintainer u should provide different isos even if they are just for x86 15:40 < madtux> :) 15:40 < mnemoc> madtux: i have some problems building bootdisk 15:40 < mipe> so any ideas why i get this:-o crtbegin.o 15:40 < mipe> In file included from ../../gcc/crtstuff.c:62: 15:40 < mipe> ../../gcc/tsystem.h:79:19: stdio.h: No such file or directory 15:40 < mipe> ../../gcc/tsystem.h:82:23: sys/types.h: No such file or directory 15:40 < mipe> ../../gcc/tsystem.h:85:19: errno.h: No such file or directory 15:40 < mipe> ../../gcc/tsystem.h:92:20: string.h: No such file or directory 15:40 < mipe> ../../gcc/tsystem.h:93:20: stdlib.h: No such file or directory 15:40 < mipe> ../../gcc/tsystem.h:94:20: unistd.h: No such file or directory 15:41 < mipe> ../../gcc/tsystem.h:100:18: time.h: No such file or directory 15:41 < mipe> make[1]: *** [crtbegin.o] Error 1 15:41 < mipe> doesnt find the headers but why 15:41 < mnemoc> madtux: i have a mini-xfce ready :) 15:41 < mnemoc> mipe: nvidia? 15:41 < mipe> nope,0-gcc 15:41 < madtux> mipe: no glibc ? 15:41 < madtux> no linux-headers? 15:42 < mnemoc> mipe: what are you using as host? 15:42 < mipe> hmm. nope. bug in Build-Target? all i have 0-00-dirtree.log 0-binutils.log 0-gcc.err 0-linux-header.log in build logs 15:42 < mnemoc> mipe: trunk:HEAD? 15:43 < madtux> mipe: mmm... 15:43 < mipe> yep 15:43 < mipe> head 15:43 < madtux> what arch? optimization? target? pkg template? 15:43 < mnemoc> what _host_? 15:43 < madtux> mipe: what ditro do u run on the host machine 15:43 < madtux> ? 15:43 < madtux> :) 15:44 < mipe> amd64,athlon64,desktop,minimal desktop and beta2 as host 15:44 < mipe> :) 15:44 < madtux> do'h.. 15:44 < mnemoc> beta2 for amd64? 15:44 < mipe> theres one for amd64? 15:44 < mnemoc> no afaik 15:44 < mipe> these are pmm 15:44 < mipe> x 15:44 < madtux> mm.. 15:45 < mnemoc> rebuilt kernel for amd64? 15:45 < mipe> hmm,kernel doesnt rebuild 15:45 * mnemoc wants an amd64 to /test/ 15:46 < mipe> brb,going for smoke and see news president speaking about the asian tsunami 15:47 < madtux> mipe: and u are from.. 15:47 < mipe> Finland 15:47 < madtux> ah yeah :) 15:47 < mipe> my friend was there, but arrived here safely this morning 15:51 < madtux> glad to hear 15:55 < madtux> powerplay devices has not yet released poerpaly 1x :( 15:55 < madtux> power* 15:55 < madtux> powerplay even ... geez. 15:55 < mnemoc> there was a guy from filipines here 15:55 * mnemoc has a cousin on the survivors list 16:05 < madtux> good. 16:06 < mnemoc> we (chilean) were lucky, only 3 missing persons until now 16:07 < mnemoc> there were ppl from .cr? 16:08 < madtux> yes there were.. but to be honest i have not been following news at all.. i've been to busy to waste my resting time watching tv 16:09 < mnemoc> :) 16:10 < mnemoc> madtux: so i guess you haven't talked to chris :) 16:11 < madtux> mnemoc: no i haven't.. but besides he is US.. he and joce are safe :) 16:11 < mnemoc> :) 16:11 < madtux> he was SOOO fscking lucky 16:11 < madtux> :) 16:11 < mnemoc> yeah 16:13 < jsaw> re 16:13 < madtux> wb jsaw 16:13 < jsaw> hi madtux, mnemoc, mipe! 16:13 < mnemoc> wb jsaw 16:14 < mnemoc> jsaw: /usr/src/building/t2/test/target/bootdisk/linuxrc.c:31:22: sys/swap.h: No such file or directory <--- is it ok to add linux-headers to it's -Is? 16:15 < rxr> think so 16:15 < rxr> hi jsaw 16:15 < mnemoc> ignoring nonexistent directory "/usr/src/building/t2/test/build/boot2-2.1.0-beta3-x86-pentium-mmx-32-bootdisk-expert/usr/include" 16:15 < madtux> ack 16:15 < jsaw> oh, damn. I forgot that. 16:15 < mnemoc> uhm 16:15 < madtux> wb rene 16:15 < jsaw> hi rxr 16:15 < mnemoc> jsaw: you forgot what? 16:16 < jsaw> I already had this problem and forgot to add the either -isystem or -I option. 16:16 < mnemoc> jsaw: all yours :) 16:17 < jsaw> However the real fix is to add swap.h to /usr/dietlibc/include/sys ... 16:18 < rxr> oh - yes, indeed the glibc chips one ... 16:18 < rxr> ships even 16:19 < rxr> so /me away again - tweaking various linux boxes in my parents house and continue the fith 16:19 < daja77> hf 16:19 < jsaw> hi daja77 16:19 < rxr> fight with the riva tnt card and the nvidia binary only crap 16:20 < mnemoc> binary only crap we must learn to live with :\ 16:21 < madtux> i refuse to 16:22 < madtux> soon or late they WILL have to release the code 16:22 < mnemoc> why? 16:25 < madtux> because i believe and sharing and been to access the code to improve it? 16:25 < madtux> because i strongly believe the the FreeSoftware movement? 16:27 < valentin> LKA is here on the congress 16:27 < madtux> LKA ? 16:27 < valentin> LKA == state police 16:27 < madtux> oh.. 16:27 < madtux> hide quick 16:27 < madtux> :) 16:28 < valentin> hehe 16:28 < CIA-10> jsaw * r5303 /trunk/package/base/dietlibc/swap-h.patch: * add sys/swap.h header to dietlibc 16:28 < jsaw> mnemoc: please test 16:36 < mnemoc> mom 16:37 < mnemoc> madtux: you do, i do, but they don't 16:38 < madtux> mnemoc: ... test? 16:38 < mnemoc> madtux: no, believe 16:38 < madtux> yeah.. but soon or late they will believe too 16:40 < mnemoc> some month ago a guy from nvidia was interviewed and said they _can't_ release the code due to legal problems 16:40 < mnemoc> neither their specs 16:40 < mipe> hmm. does O in packages list that it's disabled? 16:41 < madtux> yes 16:41 < madtux> mnemoc: then we will have to find it the hard way 16:41 < mipe> hmm. no wonder my gcc doesnt compile. glibc is disabled in packages 16:41 < madtux> DO'H! 16:41 < madtux> mipe: what target are u building? 16:41 < mipe> desktop 16:42 < mnemoc> what options did you checked on config? 16:42 < mnemoc> "disable broken" was enabled? 16:43 < madtux> *G* 16:43 < mipe> hmm. nope its disabled 16:43 < mipe> no wait enabled 16:43 < mipe> :) 16:43 < mnemoc> :D 16:43 < mnemoc> evil feature 16:43 < mipe> but why its marked as broken? 16:43 < mnemoc> i think i should make it consistent to ABORT_ON_ERROR_AFTER 16:44 < mnemoc> if it has no .cache it's considered 'broken' 16:44 < mipe> oh wait hang on. i havent checked the box so it builds packages that are broken? 16:45 < mnemoc> safer is to create a _new_ config without changing more that arch 16:45 < madtux> :) 16:45 < mnemoc> or rm -rf config/default :) 16:45 < madtux> mnemoc: NO.. he sould at least do: 16:46 < mipe> on system and default target its not checked but default doesnt have O in package list 16:46 < madtux> arch, optimization, target, and on expert, enable xtrace, retry building and maybe cleanup src dirs 16:46 < mnemoc> yes 16:47 < mnemoc> mipe: grep glibc config/default/packages 16:47 < mipe> X 01---5---- 101.600 base glibc 2.3.4-2004-08-01 / base/library 0 16:47 < madtux> good 16:47 < mipe> O 01---5---- 101.600 base glibc 2.3.4-2004-08-01 / base/library 0 16:47 < mipe> on system target 16:47 < mipe> diff between those are arch 16:48 < mnemoc> mipe: on pkgsel add: X glibc 16:48 < mipe> but brb. sauna is warm :) 16:48 < mipe> ok 16:49 < madtux> mipe: take the laptop with u 16:49 < madtux> :) 16:49 < mipe> :) 16:49 < madtux> ok router target build @ stage 5 16:54 < mnemoc> jsaw: built :) 16:54 < mnemoc> jsaw: -> Creating isofs.txt file .. 16:54 < jsaw> *good* 16:54 < jsaw> now sending patch upstream 17:00 < jsaw> cu later 17:00 < mnemoc> cu jsaw 17:14 -!- rbarnich [~root@adsl-68-252-147-41.dsl.sfldmi.ameritech.net] has joined #t2 17:18 < rbarnich> the BUILD docs with t2-2.1.0-bets2 indicate that the distribution you are building on should be ROCK. Has anyone used knoppix\kanotix? 17:20 < mnemoc> we use to required devfs, but not anymore 17:21 < mnemoc> require* 17:21 < mnemoc> dev packages and some new versions of usual stuff are needed to 17:21 < mnemoc> no real /distribution/ restrictions 17:23 < rbarnich> tnx. I want to build min x dist w/ firebird for browser app 17:25 -!- rbarnich [~root@adsl-68-252-147-41.dsl.sfldmi.ameritech.net] has quit [Remote closed the connection] 17:30 < mipe> huoh,1-glibc fails. damn 17:30 < mipe> /t2_src/t2/t2-trunk/src.glibc.1104336715.30985.7f0100/libc/objdir/debug/segfault.os 17:30 < mipe> ../sysdeps/generic/segfault.c: In function `catch_segfault': 17:30 < mipe> ../sysdeps/generic/segfault.c:192: internal compiler error: in exp_equiv_p, at cse.c:2659 17:31 < mnemoc> mipe: try to do a 'cross-build' 17:31 < madtux> i _MUST_ get me an amd64 17:31 < madtux> writting email to amd 17:32 < mnemoc> mipe: your OS is not for amd64 so it cannot run amd64 binaries 17:33 < madtux> mnemoc: perhaps "a crossbuild" can do the trick? 17:33 < mipe> oh crap,i thought i checked it, stupid me. do i need to rebuild everything up to this point? 17:33 < mnemoc> a crossbuild might give him an amd64 base to do a native build 17:33 < mnemoc> mipe: yes 17:33 < madtux> mipe: yes i would suggest clean build 17:37 < mnemoc> mipe: generic cross build, not desktop or any other target 17:47 < mipe> hmm. package list went quite short. 55 packages. but ./scripts/Download -cfg system -list-missing reports way more than that 17:47 < mnemoc> that's only stages 0 and 1 17:47 < mnemoc> a base to dump into a machine 17:47 < mnemoc> and start a native build 17:48 < mipe> ah yeah :) 17:50 < CIA-10> mnemoc * r5304 /trunk/target/bootdisk/build_stage2.sh: * improved bootdisk's $package_map iteration 17:51 < madtux> mnemoc: bootdisk target broken? 17:51 < mnemoc> improved != fixed 17:53 < madtux> just asking 17:53 < madtux> :) 17:58 < mnemoc> talking about bootdisk.... 17:58 < mnemoc> -> Forgotten packages summary: 17:58 < mnemoc> !> `- bdb 17:58 < mnemoc> !> `- glibc 17:58 < mnemoc> !> `- grub 17:58 < mnemoc> !> `- joe 17:58 < mnemoc> !> `- kbd 17:58 < mnemoc> !> `- libol 17:58 < mnemoc> !> `- patchutils 17:58 < mnemoc> !> `- syslinux 17:59 < mnemoc> these packages are built on bootdisk but they are not mapped to install or skip on stage2 18:02 < mnemoc> i think i can skip them all.... comments? 18:03 < madtux> i think syslinux should be in 18:03 < mnemoc> yes? why? 18:04 < madtux> well don't we wactually need it to actually boot for install ? 18:05 < mnemoc> stage2 is used on runtime during installation 18:06 < madtux> nevermind :) 18:06 < mnemoc> what have you missed installing rock/t2? 18:07 < madtux> be more specific 18:07 < mnemoc> a package 18:07 < mnemoc> have you installed rock/t2 using bootdisk lately? :) 18:08 < madtux> no. but i will today 18:09 < madtux> got to test my minima* stuff 18:09 < madtux> minimal* 18:09 < mnemoc> btw, i have a mini-xfce is 18:09 < mnemoc> s/is/here/ 18:09 < mnemoc> iso* 18:09 < madtux> upload ? 18:10 < mnemoc> i'll put it in www.geeks.cl/~amery 18:10 < mnemoc> let me create md5 18:11 < madtux> very well amery 18:11 < mnemoc> Error logs from full-2.1.0-beta3-x86-pentiumpro-32-generic-expert: 18:11 < mnemoc> [5] network/gsmlib [5] e17/efsd 18:11 < mnemoc> [5] e17/entrance [5] kde/kdebindings 18:11 < mnemoc> 1348 builds total, 677 completed fine, 4 with errors. 18:12 < mnemoc> do you want to fix some of those? 18:12 < mnemoc> madtux: done 18:13 < madtux> mnemoc: ok 18:13 < mnemoc> melt my bandwidth :( 18:13 < madtux> when do u want me to do it? 18:13 < mnemoc> now? :) 18:15 < madtux> ok 18:15 < mnemoc> and you can put it on /unofficial/2.3.0-prebeta3 on your t2's iso 'mirror' :D 18:15 < madtux> _MY_ t2 iso mirror.. 18:15 < madtux> :P 18:15 < mnemoc> :D 18:21 < mipe> hmm. now 1-dietlibc fails 18:22 < madtux> whats the error? 18:22 < mipe> mkdir bin-i386 18:22 < mipe> x86_64-unknown-linux-gnu-gcc -I. -isystem include -pipe -nostdinc -Os -fomit-fra 18:22 < mipe> me-pointer -falign-functions=1 -falign-jumps=1 -falign-loops=1 -mpreferred-stack 18:22 < mipe> -boundary=2 -Wall -W -Wchar-subscripts -Wmissing-prototypes -Wmissing-declaratio 18:22 < mipe> ns -Wno-switch -Wno-unused -Wredundant-decls -c i386/start.S -o bin-i386/start.o 18:22 < mipe> i386/start.S:0: error: -mpreferred-stack-boundary=2 is not between 4 and 12 18:22 < mipe> :) 18:29 < mnemoc> re 18:30 < mnemoc> what? 18:30 < mnemoc> an stack of size 4?? :p 18:31 < mnemoc> mipe: can you check cmd_wrapper.log and see if is the wrapper which is setting that? 18:32 < mipe> sure 18:32 < mipe> yeah.it does 18:33 < mnemoc> can you query me the last block? 18:37 < rxr> re 18:38 < rxr> is this dietlibc? 18:38 < rxr> it might build for i386 and not for x86-64 18:38 < mnemoc> ? 18:40 < rxr> mnemoc: we ahve this unorthodox rock inherited thing that we that arch=x86 and this ..._ARCH_BIT=64 and so on mess 18:40 < rxr> maybe dietlibc.conf does not inject the right arch for x86 64bit builds ... 18:40 < rxr> I think we should revert this clifford wolf idea anywaz 18:40 < rxr> anyway even 18:41 < rxr> it only creates mess and on sparc it is also way more handy to have seperate sparc and sparc64 kernel configs ... 18:41 < rxr> currently it is quite uneasy to add the differences to the archtecture/* files 18:42 < mnemoc> ic 18:42 < mnemoc> i have to go now... can you try to fix that? 18:43 < mnemoc> i have never touched a 64 bits machine 18:43 < mnemoc> you have sparc64 to test and mipe amd64 18:43 < madtux> mnemoc: where are u going? 18:43 < mnemoc> house 18:44 < mnemoc> wife is calling :p 18:44 < mnemoc> i have to pick her to the mall :\ 18:44 < madtux> okis.. have fun 18:44 < madtux> drive safe 18:44 < rxr> cu mnemoc 18:45 < mnemoc> cu 18:45 < rxr> mipe: I'm currently only on a console - could you send me the full log? 18:45 < rxr> or put it on some http:// location? 18:45 < rxr> if you want to send it please use rene@exactcode.de 18:45 < mipe> i'll paste it to you 18:45 < mipe> whole error-log? 18:45 < mnemoc> and cmd_wrapper 18:46 < mnemoc> compress and mail 18:47 < rxr> error-log would be enough for me ... :-) 18:47 < mnemoc> =) 18:49 < madtux> mnemoc: GO! ur wife is waiting 18:49 < madtux> :P 18:49 < mnemoc> yes 18:49 < mnemoc> bye 18:49 < madtux> u'll get ur ass kicked if u don't hurry 19:01 < mipe> rxr:did you get the mail? 19:02 < rxr> I have to look later 19:03 < rxr> I'm on some other box investigating nvidia binary only driver bugs ... 19:03 < mipe> oh ok 19:07 < mipe> well i found the cause of it. findflags.sh returns allways the stack-size of 2 19:07 < mipe> well always for this compiler 19:31 -!- nzg [~tschmidt@p508EAE95.dip.t-dialin.net] has joined #t2 19:31 * N0V4K is away: Estoy ocupado 19:35 < rxr> mipe: findflags.sh is shipped with dieltibc? 19:43 < rxr> hm - the open source nv driver does not freeze when i disable the hwcursor ... 19:43 < rxr> hm ... 19:43 < rxr> :-( 19:43 < rxr> or at least ;-) with this damn nvidia crap chips ... 19:44 < rxr> let's hope it stays stable with this config ... 20:04 < mipe> mipe: findflags.sh is shipped with dieltibc? 20:04 < mipe> rxr:yes,its shipped 20:05 < mipe> rxr:but i think the arch is setted wrong. it starts to compile on i386 when it should go to x86_64? 20:05 * mipe is afk 20:15 < rxr> mipe: ack - as I feared 20:15 < rxr> I take a look in a second 20:22 < rxr> btw. mail arrived 20:26 < rxr> mipe: still here? 20:30 < mnemoc> rehi 20:33 < rxr> hi mnemoc 20:33 < mnemoc> hi rxr 20:34 < rxr> this Riva TNT2 cars gives me headaches ... 20:34 < rxr> card even 20:34 < rxr> no cars ... 20:34 < mnemoc> :) 20:35 < mnemoc> gstreamer-gconf <--- is it a package itself or generated by another package? 20:35 < rxr> oehm 20:35 < rxr> /var/adm/flists/gst-plugins:gst-plugins: opt/gnome28/lib/pkgconfig/gstreamer-gconf-0.8.pc 20:36 < mnemoc> ok 20:36 < mnemoc> thanks 20:36 < rxr> .oO 20:36 < rxr> collect2: ld terminated with signal 11 [Segmentation fault], core dumped 20:36 < rxr> seems I have some bug to fix... 20:37 < mnemoc> :) 20:37 < rxr> no :-( 20:38 < mnemoc> no? 20:38 < rxr> no ;-) it is sad that I have to tweak the globbing patch again ... 20:39 < mnemoc> :D 20:39 < mnemoc> i told you something wrong happened on it ;) 20:40 < rxr> yes - but I can still not see the regression you experience ... 20:42 < rxr> hm ld crashes here : 20:42 < rxr> #0 0x4008e2ff in malloc_set_state () from /lib/libc.so.6 20:42 < rxr> #1 0x4008ff79 in free () from /lib/libc.so.6 20:42 < mipe> back 20:43 < rxr> mipe can you open src.dietlibc*/debug.buildenv and tell me the value assigned to the "cpu" variable? 20:43 < rxr> btw I started a amd64 crossbuild on my home athlon, too ... 20:43 < rxr> so I can fix errors, too 20:43 < rxr> fix and test that is ... 20:43 < mnemoc> Error logs from full-2.1.0-beta3-x86-pentiumpro-32-generic-expert: 20:43 < mnemoc> [5] network/gsmlib [5] e17/efsd 20:43 < mnemoc> [5] e17/entrance [5] kde/kdebindings 20:43 < mnemoc> 1348 builds total, 709 completed fine, 4 with errors. 20:44 < mipe> rxr:declare -- cpu="x86_64" 20:44 < rxr> where 3 of them are well known ... 20:44 < rxr> mipe: hm- that looks good - so that should not be the problem ... 20:45 < mnemoc> gsmlib has an /ilegal/ use overloading operators on a template only you know how to fix :) 20:45 < rxr> ouch 20:46 < mnemoc> acording to what i read it is a problem in the order of the declarations 20:46 < mnemoc> but that (c++) syntax confuse me a lot 20:46 < rxr> ok - I'm on it right now 20:47 < mnemoc> finish real problems first ;) 20:47 < rxr> have you already looked out for updates of that package? 20:47 < rxr> rebuilding binutils with debugging symbols takes some time anyway 20:47 < mnemoc> rxr: yes, no updates 20:47 < rxr> thanks 20:47 < mipe> after i got the stack thingie "fixed" i get the these errors: 20:47 < mipe> i386/start.S: Assembler messages: 20:47 < mipe> i386/start.S:8: Error: suffix or operands invalid for `pop' 20:47 < mipe> i386/start.S:10: Error: suffix or operands invalid for `push' 20:47 < mipe> i386/start.S:22: Error: suffix or operands invalid for `push' 20:47 < mipe> i386/start.S:23: Error: suffix or operands invalid for `push' 20:47 < mipe> i386/start.S:24: Error: suffix or operands invalid for `push' 20:47 < mipe> i386/start.S:69: Error: suffix or operands invalid for `push' 20:48 < mnemoc> i386? 20:48 < mipe> ? 20:48 < rxr> I think dietlibc does mess s.th. up between i386 and x86_64 and / or when in crosscompile mode ... 20:49 < mipe> well arch is set to i386 on Makefile 20:49 < rxr> this is partly ok 20:49 < rxr> dietlibc also builds the diet binary for the building architecture 20:49 < rxr> maybe it does use the wrong compiler for it ... 20:59 < CIA-10> mnemoc * r5305 /trunk/package/gnome2/teatime/teatime.desc: * repriorized teatime after gst-plugins 20:59 < rxr> what ? 20:59 < rxr> teatime needs gst-plugins ? 21:01 < mnemoc> yep 21:01 < mnemoc> well... it needs gstreamer-gconf 21:01 < rxr> unbelievable ... 21:01 < rxr> or better amazing ;-) 21:01 < mnemoc> improvements :) 21:11 < madtux> :P 21:27 < rxr> gsmlib fixed 21:27 < rxr> commit pending 21:28 < mnemoc> i knew it was easy for you :) 21:29 < rxr> no actually not 21:29 < rxr> the code is strange 21:30 < rxr> and I just hacked it to compile ... 21:31 -!- mipe [~mipe@dsl10040.japo.fi] has quit ["using sirc version 2.211+KSIRC/1.3.11"] 21:31 < rxr> mipe was at 1-dietlibc ? 21:33 < mnemoc> imagine was strange it was for me :) 21:34 < CIA-10> rene * r5306 /trunk/package/network/gsmlib/ (compile.patch gcc33-fix.patch): * updated the gsmlib gcc compile patch and added info and credit 21:35 < mnemoc> :) 21:35 < madtux> WEEEEEEEE!!!!!!!!! Router target builds again 21:36 < mnemoc> stage-9 included? 21:36 < madtux> got other small fixes to do but that will be done tonight.. now i will commit 21:36 < madtux> mnemoc: i don't do stage 9 on router 21:37 < mnemoc> madtux: please do atomic commits 21:38 < madtux> atomic commits? 21:39 < mnemoc> one commit per fix 21:39 < mnemoc> and "fix the whole s.th" is not one fix :) 21:41 < madtux> "Updated the package list so the target builds again" 21:41 < madtux> ? 21:41 < madtux> :) 21:41 < mnemoc> nope :) 21:42 < mnemoc> each update on a commit :) 21:42 < mnemoc> something easy to revert :p 21:42 < mnemoc> target/router can be a single one :D 21:43 < mnemoc> madtux: sh misc/archive/Commit.sh package/*/foo will be your friend 21:45 < CIA-10> miguel * r5307 /trunk/target/router/init.sh: Changed the init message with T2 info 21:47 < CIA-10> miguel * r5308 /trunk/target/router/ (pkgsel.sed pkgsel.sh): * Updated the pkg selection so the target builds again 21:48 < madtux> mnemoc: the pkg list stuff is just reenabling stuff that got disabled with the repositories renaming and the pkgs move 21:48 < mnemoc> that's an atomic commit :) 21:49 < madtux> thats the way i make my commits all the time! 21:49 < madtux> :S 21:51 < madtux> now i have a question about a commit.. before i dare to do it 21:51 < madtux> as i have been looking for $foo to remove and have a cleaner trunk.. i realized that scripts/config.hlp is not been used.. 21:51 < madtux> so i think that we should actually remove it 21:51 < madtux> please correct me if i'm wrong 21:51 < rxr> gsmlib does also not ccache ... 21:52 < rxr> -> Compiler Cache Hits while pkg build: 2.06% (2 hits, 95 misses) 21:54 < rxr> madtux: if it is not used due to a bug but contains good help text we should not remove it ... 21:56 < madtux> please take a look yourself 21:56 < madtux> :) 21:57 < mnemoc> config.hlp files are magicly read by Config 21:57 < madtux> mnemoc: i'm taling about scripts/config.hlp specifically 21:58 < madtux> please see the the file then u will see why i think it should be removed 22:02 < rxr> madtux: the file is used 22:02 < rxr> maybe it needs polish - but why do you wanna remove it ? 22:03 < mnemoc> madtux: press 'help' over the first item of Config 22:03 < madtux> i did.. 22:03 < mnemoc> COMMENT_ARCHCPUOPT 22:03 < mnemoc> where is this help defined? 22:03 < mnemoc> scripts/config.in:comment_id '- Architecture, CPU and Optimization' COMMENT_ARCHCPUOPT 22:03 < mnemoc> scripts/config.hlp:COMMENT_ARCHCPUOPT 22:03 < madtux> mnemoc: NO 22:05 < mnemoc> no? 22:05 < mnemoc> grep -R COMMENT_ARCHCPUOPT scripts/ package/ target/ misc/ 22:05 < madtux> look 22:05 < madtux> wait 22:05 < madtux> maybe i'm not seen something. 22:05 < madtux> ok i run ./scripts/Config 22:06 < madtux> and look for help as the dialog displays right? 22:06 < madtux> oh scheisse.. you are right 22:07 < madtux> i guess i was quite asleep when i saw it 22:07 < madtux> sorry. 22:07 < mnemoc> :D 22:08 < madtux> ;-) 22:09 < rxr> hm - seems libtool does cuase this ccache misbehaviour 22:09 < mnemoc> it adds random noise? 22:10 < madtux> FSCK! 22:11 < madtux> i need 'perl' package to build the router target.. and its fine when i want to build and all... BUT if i run download.. i will get ALL the 'perl' repository 22:12 < madtux> why couldn't we just keep the perl5 name for the package ? :( 22:15 < rxr> use "perl perl" as quick workaround to patch only perl ... 22:15 < rxr> (repository and package) 22:15 < rxr> we renamed perl because it makes no sense to carry around a perl package named perl5 ... 22:17 < mnemoc> madtux: that was one of the first threads on t2ml 22:17 < rxr> in our pkgsel redesign this is already solved - you need perl/* to select all in a repository with it ... 22:18 < rxr> yes - I think it is due to the libtool dependency file output: 22:18 < rxr> like: .deps/gsm_util.TPlo 22:18 < madtux> rxr: so for now '/ perl perl / { p; d; }' ?! 22:18 < mnemoc> it still needs to be improved for categories selection 22:18 < rxr> yes 22:18 < madtux> whould that actally work? 22:19 < mnemoc> madtux: you can use target/$target/pkgsel instead of sed inside config.in 22:20 < madtux> mnemoc: well i was very happy with the little pksel.sh thingy... untill everything the repos were renamed, the folers moved and or renamed and no new cache files were created 22:20 < madtux> now that script is worth a $foo because it doesn't add the dependicies 22:20 < mnemoc> =) 22:21 < madtux> mnemoc: and i DON'T like the target/$target/pkgsel idea.. it will make the target look like yet another $foo pkg selection 22:21 < madtux> pkg slection template 22:21 < rxr> madtux: the cache files got mostly adapted to the renaming 22:22 < mnemoc> if your target is more that a pkgsel it will get cleaner 22:22 < madtux> ah well.. 22:22 < madtux> i still have a lot of stuff to improve on the target 22:22 < madtux> so i will see that cleanup tonight 22:22 < madtux> i initailly wanted it building again 22:23 < mnemoc> you commited everything to make it build? 22:24 < madtux> yes but not in the way u are thinking :) 22:24 < mnemoc> :( 22:24 < madtux> it is still getting depedencies from pkgsel.sh 22:25 < madtux> but i added those not included automatically. 22:27 < madtux> but yes i guess i will end up removing the pkgsel.s* stuff and having the packages static 22:27 < madtux> meaning ur suggestion of target/$target/pkgsel 22:31 < mnemoc> btw, you can call pkgsel_parser as many times as tu want, so you can generate pkgsel on the fly to fix dependencies :) 22:31 -!- sparc-kly is now known as sparc-kly|ZzZzZ| 22:36 < madtux> meditate on this i will. 22:37 < madtux> mnemoc: btw.. I'm eating ur bandwidth 22:37 < mnemoc> :) 22:40 < madtux> mnemoc: seriously. 22:41 < mnemoc> you have only 200M to eat 22:41 < mnemoc> not much 22:44 < madtux> maybe i'm evil and want to download the same file from 30 different locations at the same time just to get on ur nerves? 22:45 < madtux> *g* 22:46 < rxr> ccache fix in testing 22:49 < mnemoc> no, you are not because i have the same 128k to share no all of those locations 22:49 < mnemoc> to all* 22:49 < mnemoc> rxr: ccache or libtool? 22:49 -!- jeru [~jeru@pD9516221.dip.t-dialin.net] has joined #t2 22:49 < rxr> ccache 22:50 < jeru> hi all 22:50 < rxr> hi jeru ! 22:52 < madtux> welcome jeru 22:52 < jeru> is there any good reason for having strace on the router target? 22:52 < jeru> hi madtux 22:52 < rxr> hm - a normal rebuild already shows an improvement: 22:52 < rxr> -> Compiler Cache Hits while pkg build: 47.42% (46 hits, 51 misses) 22:53 < rxr> let's see what a rebuild does ... 22:53 < madtux> jeru: atm its there because its a dependency froma pkg to BUILD the router target, but trace is not going IN the router 22:53 < madtux> jeru: yet i'm cleaning it up and improving it 22:54 < jeru> ahhh ok ... sorry 22:54 < rxr> (it might also be needed for flist trackign) 22:54 < rxr> hey! 22:54 < rxr> -> Compiler Cache Hits while pkg build: 100.00% (97 hits, 0 misses) 22:54 < madtux> jeru: nothing to be sorry, btw nice to meet you.. we were actually wondering yesterday why we never see u here 22:55 < mnemoc> rxr: grear! 22:55 < mnemoc> rxr: great! 22:55 < rxr> for gsmlib's C++ code it also reduces the compile time quite significantly ... 22:55 < madtux> jeru: ich heisse Miguel, Wie heisse du ? 22:56 < mnemoc> Error logs from full-2.1.0-beta3-x86-pentiumpro-32-generic-expert: 22:56 < mnemoc> [5] e17/efsd [5] e17/entrance 22:56 < mnemoc> [5] kde/kdebindings 22:56 < mnemoc> 1348 builds total, 739 completed fine, 3 with errors. 22:56 < jeru> Hi Miguel, Ich heisse Christian 22:56 < madtux> :) 22:56 < jeru> till now I wasn't a big fan of IRC communication 22:57 < jeru> :-) 22:59 < jeru> I want to enhance the router target to layer2 bridging functionality 22:59 < jeru> should we do it "hard wired" or via a config option 22:59 < mnemoc> jeru: madtux is router's author 23:00 * madtux hides 23:00 < rxr> jeru: hardwireing it for the first implementation might be ok - I think 23:00 < mnemoc> aaarg... a new broken package.... 5-joystick :( 23:01 < jeru> allright 23:01 < madtux> :) 23:02 < rxr> I think the router should provide some cisco like interface to an enhanced router 23:03 -!- Sparc-kly|G4 [~sparc@66.50.161.111] has joined #t2 23:03 < madtux> we are moving forward to that 23:03 < mnemoc> improving rocknet :) 23:04 < jeru> for configuration issues I commited the fwbuilder packages some time ago 23:04 < jeru> I think it's a smart way 23:05 < jeru> at least for an enviroment with more than one firewall 23:05 < jeru> fwbuilder is also capable to create cisco PIX rules, but you'll need a commercial plugin for it 23:06 < madtux> mm... 23:06 < madtux> no commercial stuff :) 23:06 < jeru> yeah that's ok 23:06 < jeru> but all the other stuff is free 23:06 < madtux> BUT it would be great to look at the free software based posibilities that we have 23:06 < jeru> take a look at fwbuilder.org 23:07 < jeru> ;-) 23:07 < madtux> 23:07 < CIA-10> rene * r5309 /trunk/package/develop/ccache/libtool-options.patch: 23:07 < CIA-10> * improved ccache to strip some options for dependency generation 23:07 < CIA-10> that just add noise and are often used by libtool (97.94% improve- 23:07 < CIA-10> ment rebuilding gsmlib) 23:07 < madtux> hehe interesting.. first word i see when the site comes up "Gentoo" 23:08 < jeru> don't worry ;-) 23:08 < madtux> i will have to write a policy for it.. 23:09 < jeru> madtux: fow what? 23:09 < madtux> as i'm looking forward to mix router with selinux 23:09 < madtux> :) 23:09 < jeru> why selinux? 23:09 < mnemoc> because madtux loves selinux 23:10 < jeru> I don't feel good if I think about using NSA stuff ;-) 23:10 < jeru> what about grsecurity? 23:10 < madtux> well actually i'm adding selinux as a feature to t2.. and considering the advanced secrutiy that selinux can offer on the OS basis.. 23:10 < madtux> jeru: now selinux is going to be an option to enable/disable from Config 23:10 < madtux> :) 23:11 < madtux> jeru: i feel much more confrotable with selinux than grsecurity 23:11 < mnemoc> jeru: we have /problems/ currently with kernel security patches because 2.6.9 and 2.6.10 are useless as linux-headers 23:11 < jeru> madtux: why? 23:12 < mnemoc> jeru: we would have to make smarter packages to patch linux-header with one patch and linux2x with another 23:12 < mnemoc> jeru: but it is on my TODO 23:15 < rxr> is .10 as broken as .9 in regards of using it as header for user-space ? 23:15 < mnemoc> net-tools failed 23:16 < mnemoc> i didn't continue 23:16 < mnemoc> but i can re-try 23:17 < rxr> do not try too hard 23:17 < rxr> it is quite ok the way we have it right now ... and we have other stuff to do ... 23:18 -!- sparc-kly [~mubex@66.50.161.111] has joined #t2 23:19 -!- sparc-kly is now known as sparc-kly|AMD 23:25 < mnemoc> hi sparc-kly 23:25 < madtux> its _NOT_ sparc-kly its _sparc-kly|AMD_ 23:25 * madtux runs 23:26 < mnemoc> and you are tinytuxie and _NOT_ madetux 23:26 < mnemoc> madtux* 23:27 < madtux> this tinytuxie can knock you down :) 23:27 < mnemoc> you can hit harder but i'm faster :) 23:27 < jeru> whats up boys ;-) 23:28 < madtux> jeru: just playing :) 23:28 < jeru> i know ;-) 23:28 < sparc-kly|AMD> lol 23:28 < sparc-kly|AMD> hi all 23:28 < madtux> mnemoc: our damn dark humor.. 23:28 < sparc-kly|AMD> :D 23:28 < madtux> this is all u fault sparc-kly|AMD 23:28 < mnemoc> yep maidtux 23:28 < jeru> but isn't so dark 23:29 < madtux> sparc-kly|AMD: btw welcome, how are u ? 23:29 < mnemoc> it has trailing smilies 23:29 < jeru> mnemoc: I had maidtux on my "fingers" ;-) but tinytuxie seems to be very violent ,-) 23:29 < jeru> ahmmm 23:30 < mnemoc> yeah 23:30 < madtux> :P 23:30 < madtux> jeru: now getting back to u on ur grsecurity / selinux question.. 23:30 < jeru> I'm still here 23:31 < mnemoc> PAX people consider 2.6 a dont go path 23:31 < madtux> i just feel much more confortable with it, i would say i'm more into selinux's RSBAC 23:31 < sparc-kly|AMD> im fine . 23:31 < mnemoc> so grsecurity will be always limited by that 23:31 < CIA-10> jsaw * r5310 /trunk/package/gnome2/gnome-volume-manager/ (. gnome-volume-manager.conf gnome-volume-manager.desc): * add gnome-volume-manager (based on HAL) 23:31 * sparc-kly|AMD test pahnter in powerbook :P 23:31 < madtux> mnemoc: thats a big plus.. 23:32 < madtux> sparc-kly|AMD: are u familiar with FLASK? 23:32 < sparc-kly|AMD> flask ? 23:32 < mnemoc> there is PAX for 2.6 but with closed development 23:33 < madtux> sparc-kly|AMD: guess not :) 23:34 < jeru> I've checked PAX site already 23:35 < mnemoc> and? 23:35 < mnemoc> i got an 'alpha' of PAX for 2.6.9 23:35 < jeru> can't find a sign that development is closed for 2.6 kernel 23:36 < mnemoc> there is no repository for ir 23:36 < mnemoc> only sporadic releases from pipacs 23:36 < jeru> ok 23:36 < mnemoc> and the topic is taboo on #pax 23:37 < jeru> mnemox: do you know the reason why? 23:37 < mnemoc> http://www.grsecurity.net/~paxguy1/ for 2.6.10 test versions 23:37 < mnemoc> jeru: they dislike 2.6 23:38 < mnemoc> they don't consider it stable enough to do serius work 23:38 < mnemoc> they seem to have go faster with 2.6.10 than with 2.6.9 23:39 < mnemoc> test7 is open 23:39 < jeru> yep 23:39 < mnemoc> 2.6.9-test5 i had to get it via mail to pipacs 23:44 < CIA-10> jsaw * r5311 /trunk/package/develop/pwlib/sasl2-path-fix.patch: * fix where pwlib expects sasl2 headers 23:44 < jeru> menomc: can I help you with the new kernel packages? 23:45 < mnemoc> kernel is updated 23:45 -!- madtux [~mike@200.91.101.102] has quit ["leaving"] 23:45 < mnemoc> what do you mean? 23:45 < jeru> regarding the security patches 23:47 < mnemoc> i have, kind of abandoned due to lack of time, vserver and rsbac 23:47 < mnemoc> and i want to add at least pax 23:48 < mnemoc> they need to have patches for 2.6.8.1 and 2.6.10 23:48 < mnemoc> and apply consistently 23:48 < mnemoc> also 2.4 23:49 < jeru> why 2.6.8.1? 23:50 < mnemoc> linux-header 23:50 < jeru> ahh..ok 23:50 < mnemoc> tools use linux-header to get the stuff 23:52 < mnemoc> gcc -v -id ./images -fd ./fonts taillights.edc -o taillights.eet <--- do you understand what this suppose to do? images and fonts are dirs, taillights.edc is on . and taillights.eet (output?) doesn't exist 23:54 < mnemoc> i see no reference to -id and -fd on man 23:55 < jsaw> there is no... this is small. 23:55 < jsaw> shouldn't be gcc but, eh, well ... lemma dig... 23:56 < mnemoc> -o was introduced by me to test... 23:56 < mnemoc> MAkefile.am use ${CC} 23:56 < jsaw> This is e17, or? 23:56 < mnemoc> yes, entrance 23:57 < jsaw> It should be the "small" compiler (the lang itself is similiar to C). 23:57 < mnemoc> they have their own c-like compiler? 23:58 < jsaw> embryo 23:58 < jsaw> http://enlightenment.org/pages/embryo.html 23:58 < mnemoc> ic 23:58 < mnemoc> *click* --- Log closed Thu Dec 30 00:00:46 2004