--- Log opened Sat Feb 26 00:00:29 2005 00:08 < jsaw> re 00:08 < mnemoc> wb jsaw 00:09 < jsaw> hola mnemoc 00:09 < mnemoc> hola jorge 00:09 < mnemoc> rxr: distcc kills 1-zlib here 00:09 < jsaw> :) 00:09 < mnemoc> :p 00:11 < jsaw> seems like I should get the distcd target going... only that I have to wait for the multiple targets gcc infrastructure... 00:11 < mnemoc> jsaw: rene doesn't want that on 2.1 00:11 < mnemoc> sadly 00:11 < rxr> jsaw: just use multiple targets to build the multiple cross compilers 00:12 < jsaw> matches with my personal situation... 00:12 < rxr> mnemoc: we can never make that in time nor cleanly 00:12 < rxr> we should not follow this rock attidute to a) make things ugly and b) endless add random stuff 00:12 < jsaw> rxr, I know. and mnemoc knows it too. We always said we want to have a _clean_ solution... 00:13 < rxr> we should get our 2.1. tree stable for a really solid base and to et people / a community 00:13 < mnemoc> yep 00:13 < mnemoc> rxr: look: 00:13 < mnemoc> distcc[5298] (dcc_cleanup_tempfiles) deleted 1 temporary files 00:13 < mnemoc> libz.a adler32.o compress.o crc32.o gzio.o uncompr.o deflate.o trees.o zutil.o inflate.o infback.o inftrees.o inffast.o 00:13 < mnemoc> make: libz.a: Command not found 00:13 < jsaw> btw, rxr, our repos got transfered to public in ROCK 00:13 < mnemoc> make: *** [libz.a] Error 127 00:13 < mnemoc> it ate $(AR) 00:14 < jsaw> mnemoc: have I been rude with my mail? 00:14 < mnemoc> jsaw: nope, i even saw a 'sad jsaw' on that mail 00:15 < jsaw> ok... I was only wondering that only praenti replied. 00:15 < jsaw> which I much apreciate 00:16 < jsaw> rxr: you look good with short hair, btw!!!! 00:16 < mnemoc> people expect cliff to answer those 00:16 < jsaw> ic 00:16 < jsaw> <- only had that long hair for about 3years 00:17 < mnemoc> i feel unconfortable with more than 1cm of hear :p 00:17 < jsaw> then switched to no hair, now again about 4cm and in a few days 0mm again... 00:17 < mnemoc> :) 00:17 < mnemoc> i use it around 5mm.... never 0mm 00:18 < jsaw> I had it all... 00:18 < jsaw> but I like 0 most. You never have to care about it whenever you stand up... :) 00:18 < rxr> mnemoc: wich stage without distcc did you complain about? 00:18 < rxr> jsaw: thanks ;-) 00:18 < jsaw> :) 00:19 < mnemoc> from 0 to (1-zlib)-1 it built on localhost only 00:19 < mnemoc> and on 1-zlib it gave that error which only happens if distcc is enabled 00:19 < rxr> 0 is currently not distributed 00:19 < rxr> but 1- should I think 00:19 < rxr> I see soon 00:20 < rxr> for really functional and perfect distcc support we would need to switch to use the arch-triplet-gcc on all stages 00:24 < mnemoc> 2.1.0 doesn't need perfect distcc support, just distcc support :) 00:26 < rxr> 0-* will not yet be distribute 00:26 < rxr> +d 00:26 < rxr> too bad my rs6k is that slow 00:26 < rxr> good that I have a video here I have not yet watched: Matrix Revolution ... 00:26 < mnemoc> :) 00:27 < rxr> and thanks to the forced dual head setup my desk looks like in matrix ... 00:27 < rxr> the movie on the iBook display - the workspace on teh external TFT ;-) 00:27 < jsaw> :) 00:27 < jsaw> my boss always makes jokes when he sees my dual head workplace (2 EIZO 21" monitors)... 00:28 < jsaw> esp. when the laptop is in place too... 00:28 < rxr> oh nice ;-) 00:28 < rxr> what graphic card(s) do you use to drive it ? 00:28 < jsaw> yeah, they are old but still better then the LCDs everybody else has... 00:28 < jsaw> matrox g450 00:28 < rxr> ,-) 00:29 < jsaw> now that everybody seems to get LCDs, maybe I should try to add another gfx card, and add two more monitors... 00:29 < jsaw> :) 00:30 < jsaw> .oO(then add a joystick and play flightgear...) 00:30 < rxr> lol 00:31 < mnemoc> :D 00:32 * mnemoc moving home, bbl 00:32 < jsaw> cu mnemoc 00:34 < rxr> yeah - cu 00:34 < rxr> damn - still 0-binutils 00:34 < rxr> ... 00:34 < rxr> guess matrix ends before the slow ppc box enters 1- ... 00:37 < jsaw> I loved the first Matrix... hope I find some time till the end of the year to watch it... 00:41 < rxr> 0-binutils: 25m ... 00:41 < jsaw> o oh 00:41 < jsaw> okay, have fun watching it another 5 times this night... 00:48 < rxr> nope 00:48 < rxr> I look at the distcc built result in te morning *g* 01:25 < rxr> ok - n8 all 01:26 < jsaw> cu rxr! 01:26 < jsaw> <- still waiting for Charlotte to have her "I'm awake" hour per night... :) 01:33 < mnemoc> re 01:35 < jsaw> wv 01:35 < jsaw> wb 01:35 < jsaw> ... 01:39 < mnemoc> thx jsaw 01:40 < mnemoc> distcc[21073] (dcc_cleanup_tempfiles) deleted 1 temporary files 01:40 < mnemoc> rm gcc.pod 01:40 < mnemoc> make[1]: Leaving directory `/usr/src/building/t2-trunk/src.gcc.1109374085.5139.a8c01505/gcc-3.4.3/objdir/gcc' 01:40 < mnemoc> make: *** [all-gcc] Error 2 01:40 < mnemoc> 0-gcc :'( 01:40 < jsaw> :( 01:44 < mnemoc> errr 01:44 < mnemoc> svn: PROPFIND request failed on '/t2/trunk' 01:44 < mnemoc> svn: PROPFIND of '/t2/trunk': Could not read status line: connection was closed by server. (http://svn.exactcode.de) 01:45 < mnemoc> second pass worked 01:48 < jsaw> o_O 01:48 < jsaw> oh, I guess dsl reconnect 01:49 < mnemoc> i'm quite lucky today :\ 01:49 < jsaw> :) 01:51 -!- jeru [~jeru@p54BFAF01.dip.t-dialin.net] has quit [Read error: 110 (Connection timed out)] 01:51 < mnemoc> jsaw: http://wiki.t2-project.org/index.cgi?TailFormat misc/tail/parse-config and package/*/runit/packages/*.tail 01:51 < mnemoc> comments? 01:51 < jsaw> *click* 01:52 < jsaw> (including firing up firefox on a 64Mb/200MHz machine....) 01:52 < mnemoc> uff 01:52 < mnemoc> links -g :p 01:55 < jsaw> how would "stop" look like then? 01:58 < mnemoc> i was thiniking is a pid line 01:58 < mnemoc> pointing to pid file 01:58 < jsaw> okay, but what about programs needing a special shutdown sequence? 01:59 < mnemoc> like? 01:59 < mnemoc> do you mean 'post-shutdown' steps? 02:00 < jsaw> not all programs have a pid file 02:00 < mnemoc> finish rm -f D_localstatedir/foo.pid 02:00 < jsaw> or apache uses apachectl to start/stop 02:01 < mnemoc> stop and poststop maybe 02:01 < jsaw> also alsa 02:01 < mnemoc> if no stop, try pid, if no pid try guessing pid 02:02 < jsaw> pppd may have multiple instances 02:02 < mnemoc> but it has a pid file 02:03 < jsaw> but you would always have to add checks how many 02:03 < jsaw> oh no. 02:03 < mnemoc> the compiler, tail2initstyle may get complex, but i'm convised that a simple set of options is posible 02:04 < jsaw> you would only have to check for wildcards in "option pid" 02:04 < jsaw> maybe adding an explicit "start", "stop" and so on option would help 02:04 < mnemoc> foo -nodetach &; pid=$_ 02:05 < jsaw> you don't have that variable 02:05 < mnemoc> pushing daemons to not self-demonize 02:05 < mnemoc> what variable? 02:05 < jsaw> "pid=$_" will not survive from "start" to "stop" 02:06 < mnemoc> i can record it on a pid file 02:06 < jsaw> hmm... 02:06 < mnemoc> /var/run/$service.pid 02:06 < jsaw> which may interfere with the programs pid file itself 02:07 < mnemoc> those files may have a pid option telling the compiler to not guess it 02:09 < jsaw> adding another guess, services with no pid file.. 02:09 < mnemoc> that's the third case 02:09 < mnemoc> i) hasoption("stop") exec(readoption("stop")) 02:10 < mnemoc> ii) hasoption("stop") exec(terminatepid(service,readoption("pid")) 02:11 < mnemoc> iii) exec(terminapid(service,recordedpid)) 02:11 < mnemoc> iv) exec(terminatepid(service,guesspid(service))) 02:11 < mnemoc> but the path shall be decided on compile time 02:12 < mnemoc> tail2initstyle compilation 02:12 < mnemoc> on supervised styles stop doesn't even exist 02:12 < mnemoc> i mean, it's ignored 02:13 < mnemoc> on non-supervised styles, log is ignored 02:13 < mnemoc> i'll start moving all runit's to .tail 02:15 < jsaw> bias 02:15 < mnemoc> then adding them to $confdir of packages which are easy to tail-ize and using an option to let install_init silently use .tail instead of the proposed 02:15 < mnemoc> .init 02:15 < mnemoc> ok 02:20 < jsaw> where I hestitate is the language construct itself actually 02:20 < jsaw> I'd prefer something like a makefile syntax 02:22 < mnemoc> i choosed the dummies parser i could imagine :p 02:22 < mnemoc> due to the lack of time 02:23 < jsaw> I understand, only I miss something that would add more structure 02:23 < jsaw> like 02:23 < jsaw> if you need a pid file: 02:23 < jsaw> PID=... 02:23 < jsaw> for start,stop events one would use 02:23 < jsaw> start: 02:23 < jsaw> stop: 02:23 < jsaw> next line with indent 02:24 < jsaw> or if you don't need a specialized start/stop 02:24 < jsaw> only 02:24 < mnemoc> the main difference is that i mainly see superviced services where start is a single line 02:24 < jsaw> EXEC=... 02:24 < mnemoc> and you focus on sysvinit's 02:24 < mnemoc> but we must find a simple interim point 02:25 < jsaw> why? 02:25 < mnemoc> why what? 02:25 < jsaw> (I mean the "and you focus on sysvinit's") 02:26 < mnemoc> start: 02:26 < mnemoc> stop: 02:26 < mnemoc> both multiline 02:26 < jsaw> not necesarrily, but possible 02:26 < mnemoc> i see prestart and poststop, and atomic start 02:26 < jsaw> what's about samba without multiline? 02:27 < mnemoc> i have single-line smb inits 02:27 < jsaw> okay, but still starting three programs... 02:27 < mnemoc> http://svn.exactcode.de/t2/trunk/package/base/runit/packages/samba-nmbd_run.txt 02:28 < mnemoc> those are different inits 02:28 < mnemoc> one for smbd 02:28 < mnemoc> one for nmbd 02:28 < mnemoc> and one for winbind 02:28 < jsaw> in case I have to restart samba,... 02:29 < jsaw> what then? 02:29 < jsaw> first stopping three services... 02:29 < mnemoc> sysvinit's can have dependencies 02:29 < jsaw> the starting three services... 02:29 < mnemoc> but yes 02:29 < jsaw> which would mean forward and backward dependencies...? 02:29 < mnemoc> the clean way is to start and stop three services 02:30 < mnemoc> backward 02:30 < jsaw> ok, clean, but not very handy 02:30 < mnemoc> rc script can be able to identify services of the same group 02:31 < jsaw> does sysvinit's "rc" do this already? 02:31 < mnemoc> ours? no 02:32 < jsaw> that adds complexity then, doesn't it? 02:32 < mnemoc> you want 'handy', 'handy' adds complexity 02:32 < mnemoc> talking about samba 02:32 < mnemoc> smbd,nmbd,winbind 02:32 < mnemoc> i can want to restart only one of them 02:33 < mnemoc> i can want to stop only one of them 02:33 < jsaw> handy for me means, I do not have to hack sysvinit, but only having a metalanguage able to do the job for all, sysvinit, minit, runit...etc 02:33 < mnemoc> if you group them in a single service script you can't 02:33 < mnemoc> rc is not part of sysvinit, it's an invention of each distribution 02:34 < jsaw> ic 02:34 < jsaw> well, I'm use to /sbin/init.d/-service- restart ... 02:34 < jsaw> used 02:34 < jsaw> I'm not using "rc" that often 02:35 < mnemoc> init.d/samba calling init.d/{smbd,nmbd,winbind} ? 02:35 < jsaw> so, what I mean, it would add complexity to the service script itself to identify the dependency 02:35 < mnemoc> backward or forward? 02:35 < mnemoc> backward is a good thing 02:35 < jsaw> please, don't let me have to type that whenever one wild XP server (having a firewall but SMB service activated) is going wild 02:35 < mnemoc> DEPEND=serv1 serv4 02:36 < jsaw> with smb, the order which service is started matters 02:36 < mnemoc> not really 02:36 < jsaw> believe me, it does... 02:36 < jsaw> if you have vmware running it matters. 02:36 < mnemoc> i have them supervised and start 'randomly' on 3 companies 02:37 < mnemoc> .oO 02:37 < jsaw> yes, .oO 02:38 < mnemoc> from atomic-services-declarations i'm able to compile a joined-service-script 02:38 < mnemoc> but not on the other way 02:38 < jsaw> btw, what happens with the "system" service with your tail impl? 02:38 < mnemoc> system is not a service 02:38 < jsaw> ? 02:39 < jsaw> cleanup, preparation service,isn't it? 02:39 < mnemoc> system configures your machine 02:39 < mnemoc> configurator vs. service 02:39 < mnemoc> services stay running 02:39 < mnemoc> configurator run once only 02:39 < jsaw> but this must fit into tail too, or? 02:40 < mnemoc> raw script including a common /etc/functions with helpers? 02:41 < mnemoc> compiled scripts may source /etc/functions too and share behaviors and helpers 02:41 < mnemoc> to me alsa is not a 'service' neither 02:41 < jsaw> no, I meant how does this fit into the language? 02:42 < mnemoc> it doesn't fit 02:42 < mnemoc> or maybe a raw option, but i don't like it 02:42 < mnemoc> i prefer raw script sourcing a helper library 02:42 < jsaw> how does your "system" thing look like with runit? 02:43 < mnemoc> runit has 3 stages 02:43 < mnemoc> 1: configure 02:43 < mnemoc> 2: services 02:43 < mnemoc> 3: de-configure 02:43 < mnemoc> system is called explicitly on stage1 script 02:43 < mnemoc> same does alsa 02:43 < mnemoc> and system is called explicitly on stage3 02:44 < mnemoc> services are only supervisable stuff 02:44 < jsaw> so, you mean, tail would apply to "stage2" only? 02:44 < mnemoc> as i see it, yes 02:45 < jsaw> hmm.... 02:45 < mnemoc> i don't think configuration scripts and services are mixable 02:45 < jsaw> I have to bug again about the makefile syntax thing 02:45 < mnemoc> makefile syntax is good for system 02:45 < mnemoc> but overhead for services 02:46 < jsaw> no 02:46 < jsaw> service could look like this (one line): 02:46 < jsaw> SERVICE=/path/to/some/daemon 02:47 < jsaw> and stage1/3 would use the "start:" "stop:" thing 02:47 < mnemoc> another complex side is $log_to_stderr argument of services 02:48 < mnemoc> .oO 02:48 < mnemoc> how does pre-start, and post-stop statements fit on your Makefile syntax? 02:48 < jsaw> not really, I that case one could always add "|tee /dev/tty10" or so... 02:49 < mnemoc> i mean syslog vs. stderr 02:49 < jsaw> first I have to ask for a case where you need pre-start 02:49 < jsaw> oh 02:49 < jsaw> hmmm 02:50 < mnemoc> pre-start may be seting vars to be used on 'exec' line 02:50 < mnemoc> like sysklogd 02:51 < mnemoc> on kernel 2.4 nobody can read /proc/kmsg 02:51 < mnemoc> on 2.6 only root can 02:51 < mnemoc> or detecting `hostname` on a dhcpc service 02:51 < jsaw> Charlotte... back in some seconds I hope 02:51 < mnemoc> :) 02:58 < jsaw> quite again... let's see 02:59 < mnemoc> pre-start: activities to realize before starting _the_ service 03:00 < mnemoc> post-stop: activities to do after stoping _the_ service 03:00 < jsaw> hmmm.... crying again.... 03:00 < mnemoc> :p 03:00 < mnemoc> plug her =) 03:01 -!- _Ragnar__ [loki@66-146-166-62.skyriver.net] has joined #t2 03:01 < mnemoc> hi _Ragnar__ 03:01 -!- _Ragnar_ [loki@66-146-166-62.skyriver.net] has quit [Read error: 101 (Network is unreachable)] 03:03 < mnemoc> pre-start activities may prepare the enviroment to the run, check for dependencies, set envvars, run configuration rutines or cleanup stuff of a previous bad-shutdown 03:03 < mnemoc> post-start activities are mainly cleanup stuff 03:03 < mnemoc> post-stop i mean 03:07 < jsaw> <- can't type with only one arm without light on my kbd... 03:07 < mnemoc> :p 03:07 < jsaw> <- needs voip 03:07 < mnemoc> btw, this tail v0 is a quick and dirty approach to clean and generalize .init files 03:08 < mnemoc> having the cleaned and limitations recognized a nicer C++ parser can be written 03:09 < mnemoc> and tail v1 get born 03:09 < jsaw> yeah, of course, but one does not want tge syntax to change again,,, 03:10 < mnemoc> can you do Makefile-like syntax parsing with few lines of code? 03:10 < mnemoc> maybe awk? 03:10 < mnemoc> i don't know awk :\ 03:11 < jsaw> bash should work also 03:11 < mnemoc> http://svn.exactcode.de/t2/trunk/package/base/runit/packages/djbdns-dnscache.tail <--- look 03:13 < CIA-9> amery * r6819 /trunk/package/base/runit/packages/ (djbdns-dnscache.tail djbdns-tinydns.tail): * fixed runit/tail djbdns declarations to use unique log id-s 03:19 < mnemoc> jsaw: i wonder how to mix decently: unique-oneliners, multiple-oneliners and unique-multiliners 03:20 < mnemoc> jsaw: jsaw that's why i choosed 'tab' as single delimited 03:20 < mnemoc> jsaw: djb-dnscache.tail doens't look ugly 03:20 < mnemoc> jsaw: imho 03:33 < mnemoc> jsaw: OT: do you agree with me about the need of /etc/functions or alike? 03:46 < jsaw> ooh, i liked the standalone character of the init scripts, otoh, there's no need for it... 03:47 < jsaw> so functions should make life way easier 03:47 < jsaw> .tail file looks nice actually... 03:48 < jsaw> ...getting a beer than proabably (or should I say hopefully) gonna try to sleep myself... 03:56 < mnemoc> :) 03:57 < jsaw> well, I skip the beer and go to sleep immediately... 03:58 < mnemoc> are you 'allowed' to fall sleep? 03:59 < jsaw> seems like her "one-hour-awake" phase has passed... 03:59 < jsaw> so, yes 03:59 < mnemoc> good :) 03:59 < jsaw> cu mnemoc! 04:00 < jsaw> P.S.: looking at djbdns-dnscache.tail, seems like the right way to go, finally 04:00 < jsaw> gn8... 04:01 < mnemoc> gn8 jsaw 05:56 -!- rxr_ [~rene@p213.54.193.177.tisdip.tiscali.de] has joined #t2 05:56 -!- Topic for #t2: T2 | 2.1.0-beta3 RELEASED | The next generation of System Development Enviroments (SDE) | http://www.t2-project.org/ | CIA, don't ever leave us again! 05:56 -!- Topic set by mnemoc [] [Sat Feb 19 18:03:55 2005] 05:56 [Users #t2] 05:56 [ _Ragnar__] [ jsaw ] [ mtr_] [ rxr ] [ SerWou ] 05:56 [ CIA-9 ] [ mnemoc] [ nzg ] [ rxr_] [ valentin] 05:56 -!- Irssi: #t2: Total of 10 nicks [0 ops, 0 halfops, 0 voices, 10 normal] 05:56 -!- Channel #t2 created Sun Aug 8 21:15:33 2004 05:56 -!- [freenode-info] help freenode weed out clonebots, please register your IRC nick and auto-identify: http://freenode.net/faq.shtml#nicksetup 05:56 -!- Irssi: Join to #t2 was synced in 11 secs 06:00 -!- mnemoc_ [~amery@200.75.27.59] has joined #t2 06:05 -!- rxr [~rene@p213.54.200.158.tisdip.tiscali.de] has quit [Read error: 60 (Operation timed out)] 06:12 -!- mnemoc [~amery@200.75.27.12] has quit [Read error: 110 (Connection timed out)] 08:45 -!- mipe [~mipe@dsl10040.japo.fi] has joined #t2 09:59 -!- mipe [~mipe@dsl10040.japo.fi] has quit ["leaving"] 10:24 -!- jeru [~jeru@p54BFAFDE.dip.t-dialin.net] has joined #t2 11:18 -!- mtr_ is now known as mtr 11:53 -!- You're now known as rxr 11:54 -!- jeru [~jeru@p54BFAFDE.dip.t-dialin.net] has quit ["using sirc version 2.211+KSIRC/1.3.11"] 12:14 -!- mtr_ [~michael@H9b07.h.pppool.de] has joined #t2 12:14 < rxr> mnemoc_: on stage 1-* distcc works here ... 12:15 < rxr> rs6k master: 12:15 < rxr> load average: 3.99, 3.46, 2.13 12:15 < rxr> Cpu(s): 49.8% user, 40.4% system, 0.0% nice, 0.0% idle, 9.8% IO-wait 12:15 < rxr> athlon slave compiling 12:15 < rxr> Cpu(s): 49.8% user, 40.4% system, 0.0% nice, 0.0% idle, 9.8% IO-wait 12:15 < rxr> grummel 12:16 < rxr> load average: 0.51, 0.31, 0.12 12:17 < rxr> and distcc as well as powerpc-.... processes running 12:17 < rxr> Feb 26 12:17:15 haudraufinix distccd[14075]: (dcc_check_client) connection from 192.168.2.1:48386 12:17 < rxr> Feb 26 12:17:15 haudraufinix distccd[14075]: compile from iovdprintf.c to iovdprintf.os 12:17 < rxr> Feb 26 12:17:16 haudraufinix distccd[14038]: (dcc_r_file_timed) 270657 bytes received in 0.032219s, rate 8204kB/s 12:17 < rxr> Feb 26 12:17:16 haudraufinix distccd[14038]: (dcc_collect_child) cc times: user 0.093985s, system 0.015997s, 2373 minflt, 0 majflt 12:17 < rxr> Feb 26 12:17:16 haudraufinix distccd[14038]: powerpc-t2-linux-gnu-gcc freopen64.c on localhost completed ok 12:17 < rxr> Feb 26 12:17:16 haudraufinix distccd[14038]: job complete 12:25 -!- mipe [~mipe@dsl10040.japo.fi] has joined #t2 12:30 < mipe> hmm,i get cant extract 2nd_stage*.tar.gz during boot, no /mnt_root/linuxrc 12:35 -!- mtr [~michael@H9058.h.pppool.de] has quit [Read error: 110 (Connection timed out)] 12:47 -!- mipe [~mipe@dsl10040.japo.fi] has quit ["leaving"] 12:58 < rxr> http://check.sourceforge.net/ 12:58 < rxr> Check: A unit test framework for C 13:11 -!- mipe [~mipe@dsl10040.japo.fi] has joined #t2 13:17 < rxr> hi mipe 13:18 < mipe> hi rxr 13:20 < mipe> the 2nd_stage error i got is because of tar not working on that stage 13:20 < rxr> ? 13:21 < rxr> http://matt.ucc.asn.au/dropbear/dropbear.html 13:21 < rxr> ^- cool 13:21 < mipe> on bootdisk, in first stage(1?) on menu i'm loading 2nd stage tar to ram, but it fails because of tar 13:22 < mipe> on first stage theres no libs(?) so everything should be statically compiled? 13:25 < rxr> you tried booting your 64bit build ? 13:26 < rxr> A small memory footprint suitable for memory-constrained environments - Dropbear can compile to a 110kB statically linked binary with uClibc on x86 (only minimal options selected) 13:29 < mipe> rxr:yep 64bit build 13:31 < rxr> you disabled dietlibc for the build or so, right ? 13:31 < rxr> mipe: yes 1st_stage is statially linked dietlibc 13:32 < mipe> :( trying without dietlibc,everything else should be working,just figuring out how to get tar working 13:35 < mipe> well actually only *.static works plus kiss,ip 13:36 < CIA-9> rene * r6820 /trunk/ (23 files in 9 dirs): 13:36 < CIA-9> * added libsvg-cairo (0.1.5) 13:36 < CIA-9> * updated waimea (0.4.0 -> 0.5.0-2005-02-24) and moved it into x11-ng 13:36 < CIA-9> * added libpixman (0.1.3) 13:36 < CIA-9> * added libsvg (0.1.3) 13:36 < CIA-9> * added glitz (0.1.3) 13:36 < CIA-9> * added cairo (0.3.0) 13:36 < rxr> you could add the glibc and other libs needed by tar to the 1st stage 13:37 < rxr> but best would be to resolve why dietlibc hinders your build in so strange ways ... 13:45 < rxr> # ./dbclient ec-outpost.dyndns.org 13:45 < rxr> Host 'ec-outpost.dyndns.org' is not in the trusted hosts file. 13:45 < rxr> (fingerprint md5 d7:74:08:2b:56:5d:8b:34:53:80:e3:32:7b:65:11:3a) 13:45 < rxr> Do you want to continue connecting? (y/n) 13:46 < rxr> cool - /me packaging this dropbear and scheduling for psion-pda target inclusion ... 13:53 -!- mipe [~mipe@dsl10040.japo.fi] has quit ["sleep"] 14:06 < CIA-9> rene * r6821 /trunk/package/security/dropbear/ (. dropbear.desc): * added dropbear (0.44) 14:26 < rxr> Sources: IBM ditching Itanium altogether 14:26 < rxr> IBM is not only discontinuing support for Intel's Itanium processor in a new generation of server technology, it's going a step further, dropping Itanium servers from the product line altogether, 14:43 -!- maddin [~martin@pD9E73B29.dip.t-dialin.net] has joined #t2 14:43 < mnemoc_> rxr: on my second try 0-gcc failed with distcc 14:43 < mnemoc_> moin! 14:44 < maddin> moin moin 14:44 < mnemoc_> moin maddin 14:44 < maddin> hi mnemoc_ 14:45 < rxr> hi mnemoc_ maddin 14:46 < maddin> hi rxr 14:50 < maddin> My new iBook will come next week. I hope, it's true 14:51 < rxr> ;-) 14:51 < mnemoc_> rxr: x11-ng? 14:54 < rxr> yeah 15:17 -!- Postal [~Postal@ool-4356292a.dyn.optonline.net] has joined #t2 15:17 < Postal> hey all 15:20 < rxr> hey Postal 15:20 < Postal> rxr, any news on gcc error yet? 15:21 < rxr> mipe did a lot f x86-64 building 15:21 < rxr> but I think he has not yet published his work ... 15:21 < Postal> ok 15:21 < rxr> as far as I know he has a quite complte generic target and install disk 15:21 < rxr> the later not yet fuly working ... 15:22 < Postal> ic 15:22 < Postal> this deban base really pisses me off, every time i compile something , my system slows down to a halt... 15:24 < rxr> oehm ... ? 15:25 < Postal> the make seems to use 100% of my cpu.. as in my mouse flickers and everything... 15:26 -!- mipe [~mipe@dsl10040.japo.fi] has joined #t2 15:27 < mipe> so running t2 amd64 15:27 < mipe> with few problems 15:27 < rxr> congrats ! 15:27 < mnemoc_> :D 15:28 < mnemoc_> mipe: will you commit your fixes? 15:28 < mipe> grub segfaults while running embed and cracklib didnt have /var/cache/cracklib/* files 15:28 < mipe> http://koti.japo.fi/~pmika/amd64-t2/ 15:28 < mipe> but i doubt rxr will like the dir style i have made 15:28 < mipe> and changes might include something that breaks other builds,so lots of testing still to do 15:30 < mipe> but this is a start :) 15:30 < mnemoc_> why you didn't `diet` your md5 stuff? 15:31 < mipe> changes doesnt include such as,dietlibc off for amd64 builds 15:31 < mnemoc_> oh 15:31 < mipe> it breaks gcc somehow,so have to look into it 15:32 < mnemoc_> mipe: losetup -d /dev/loop\$1 > /dev/null 2>&1 <---- no devfs? 15:33 < mipe> oh crap,that was for ubuntu :( 15:33 < mipe> had to change it since ubuntu didnt have it 15:34 < rxr> hm - there are still losetup occurances ? 15:34 < rxr> I thought I already removed them after -beta3 ... 15:35 < rxr> using mount -o loop ... 15:35 < rxr> ah! 15:35 < rxr> -LD=gcc 15:35 < rxr> +LD=ld 15:35 < rxr> interesting ... ,-) 15:36 < rxr> this might break this dropbear on some of my configs ... 15:36 < rxr> ! 15:37 < rxr> when you ever encounter s.th. like this: 15:37 < rxr> ld: warning: cannot find entry symbol _start; defaulting to 100015b4 15:37 < rxr> libtommath/libtommath.a(bn_mp_div.o)(.text+0x27c): In function `mp_div': 15:37 < rxr> bn_mp_div.c: undefined reference to `__udivdi3' 15:37 < rxr> it is ld used for linking instead of gcc 15:37 < mnemoc_> uhm... i loose my 'third-button emulation' 15:37 < rxr> mnemoc_: ? 15:38 < mnemoc_> no idea, it just doesn't paste any more 15:41 < rxr> hm 15:41 < rxr> what 3rd butto emulation do you sue ? 15:41 < rxr> use even 15:51 < CIA-9> rene * r6822 /trunk/ (2 files in 2 dirs): 15:51 < CIA-9> * fixed parse-config to set LD to gcc (not ld) - fixes dropbear 15:51 < CIA-9> compilation on powerpc as well as most probably many of the 15:51 < CIA-9> undefined reference to `__udivdi3' and friends during cross 15:51 < CIA-9> compilation we know about since ancient ROCK times 15:51 < CIA-9> * added dropbear.cache 15:54 < rxr> wow - even this works with the dropbear client just fine: 15:55 < rxr> dbclient rene@some-box.tld -L 25:localhost:25 -L 143:localhost:143 15:56 < rxr> 29288 Feb 26 15:56 scp 15:56 < rxr> ! 16:28 < CIA-9> rene * r6823 /trunk/package/security/dropbear/dropbear.conf: * install manual pages for dropbear 16:29 < mnemoc_> rxr: re 16:29 < rxr> mnemoc_: yes ? 16:33 < mnemoc_> Option "Emulate3Buttons" "on" <--- on mouse InputDevice 16:33 < mnemoc_> i restarted X and problem solved 16:43 < rxr> hm 16:43 < mnemoc_> aaarg! /usr/lib/gcc/i686-t2-linux-gnu/3.4.3/../../../../i686-t2-linux-gnu/bin/ld: cannot find -lgcc_s 16:44 < mnemoc_> this week will as as crapy as it started.. dammit 16:44 < mnemoc_> will end as* 16:45 < rxr> ? 16:46 < rxr> http://exactcode.de/rene/text2175.png 16:46 < rxr> my current logo working base ... 16:46 < rxr> of course only for high res prints 16:46 < rxr> not suitable for online usgae ... 16:46 < mnemoc_> nice logo 16:46 < mnemoc_> without the SDE it can be used online 16:47 < mnemoc_> did T2 turned into T^2 officialy? 16:50 < rxr> nope 16:55 < valentin> rxr: nice - if you make it a bit more colorfull and modern it will be cool 16:56 < mnemoc_> color means 'life', and T2 is an alive project :) 16:58 < valentin> btw: sde is a common abbreviation for 'state department of education' :) 16:58 < valentin> inside the us 16:59 < rxr> hehe 16:59 < rxr> I can scp he .svg so you can play with it ... 17:00 < valentin> ok 17:01 < valentin> i have some ideas 17:02 < rxr> t2-logo-2005.svg 17:02 < rxr> in the same dir 17:03 < valentin> i see you had almost the same idea than me - creating a pseudo 3D orbit for the round text 17:04 < valentin> sth like saturn rings... 17:07 < valentin> how can i select the text in inkscape ? 17:08 < valentin> ok - found out 17:25 -!- maddin [~martin@pD9E73B29.dip.t-dialin.net] has quit ["Verlassend"] 17:35 < mipe> oh theres lot of crap in my diff files,so dont commit them,i'll remove those crap on this build(syne) 17:36 < mnemoc_> :) 17:38 < mnemoc_> rxr: you s/ld/gcc/ change seems to make troubles, my new test build stoped on 1-glibc (with distcc disabled because it produced trobules last time) 17:38 < mnemoc_> make[2]: *** [/usr/src/building/t2-trunk/src.glibc.1109434695.18567.0/glibc-2.3.4/objdir/elf/rtld-libc.a] Error 2 17:45 < rxr> anything else ? 17:46 < mnemoc_> do you want the ld^Wgcc line? 17:47 < mnemoc_> and the one from cmd_wrapper? 17:47 < mnemoc_> -o /usr/src/building/t2-trunk/src.glibc.1109434695.18567.0/glibc-2.3.4/objdir/io/rtld-lxstat.os btw 17:52 < rxr> http://exactcode.de/rene/text2176.png 17:52 < rxr> not fully symetric - since I had to pratice a bit ... 17:53 < mipe> nice logo :) 17:54 < mnemoc_> nice :) 17:59 < jsaw> colors+t^2 is almost perfect, only the text I dislike 17:59 < mnemoc_> circular text, or SDE on the right? 17:59 < jsaw> both 18:00 * mnemoc_ wondering about that colorfull circle looking like a colorfull CD 18:01 < jsaw> yep... only a bit of more a 3d view, not only a stretched circle... 18:10 < valentin> jsaw: yes 18:11 < valentin> with a bit more 3d look that would be cool 18:11 < valentin> and the SDE could be a bit greater 18:12 < valentin> i dragged it a bit around in inkskape but i found no good looking transformation 18:13 < jsaw> I rather take "System\nDevelopment\nEnvironement" ragged right5D5D 18:13 < jsaw> flushed right I mean 18:13 < jsaw> lower right corner, half text size of the t 18:13 < valentin> instead of the 'SDE' ? 18:13 < jsaw> yep 18:13 < valentin> and the text around the cd ? 18:14 < valentin> discard it ? 18:14 < jsaw> no text around the CD... 18:14 < jsaw> text on paths seldomly look great imo 18:14 < jsaw> I prefer text you can read without turning the head 18:15 < mnemoc_> what about the text (tiny) inside the 'center' of the cd? 18:15 < valentin> moment i'll do it 18:16 < valentin> mnemoc: what text ? 18:16 < mnemoc_> Sys.. Dev... Enviro... 18:16 < mnemoc_> like the 'brand' or real CDs 18:17 < valentin> jsaw: flushed right === ? 18:17 < valentin> system 18:17 < valentin> development 18:17 < jsaw> right aligned 18:17 < valentin> ie 18:17 < valentin> abc 18:17 < valentin> defghi 18:17 < valentin> damn... 18:17 < mnemoc_> a 18:17 < mnemoc_> bc 18:17 < mnemoc_> err 18:17 < jsaw> *grin* 18:17 < jsaw> a 18:17 < jsaw> bc 18:17 < jsaw> def 18:17 < mnemoc_> :) 18:18 < valentin> jsaw: that does not look goot because strlen("system") << strlen("development") 18:19 < mnemoc_> strlen("system") = strlen("development")? 18:19 < valentin> just a moment. i'll upload 18:19 < valentin> << means 'much smaller than' 18:20 < mnemoc_> i now 18:20 < mnemoc_> i meaned alter "system" height 18:20 < mnemoc_> weight? 18:20 < mnemoc_> len :p 18:20 < valentin> http://exactcode.de/valentin/t2logoValentin.png 18:21 < valentin> ^- what i said 18:21 < valentin> a bit small *g* 18:21 < jsaw> almost what I meant 18:21 < mnemoc_> looks like a 'powered by' :) 18:21 < jsaw> if the "sys..." text stays as it is, make t2+ring double size 18:22 < valentin> hm 18:22 < valentin> ok 18:22 < jsaw> (was what I meant) 18:23 < valentin> just reload 18:23 < valentin> http://exactcode.de/valentin/t2logoValentin.png 18:24 < valentin> how about 'The System\nDevelopment.....' ? 18:24 < jsaw> yes! and bottom aligned, text shifte a bit under the ring... 18:24 < jsaw> *damn* I don't have gfx prog here... 18:25 < jsaw> but this looks like what I had in mind after seeing rxr's ring 18:25 < valentin> ./scrips/EmergePkg inkscape ? 18:25 < mnemoc_> \nhttp://www.t2-project.org/" 18:25 < mnemoc_> \n\n 18:25 < jsaw> 3.2G 2.8G 324M 90% / 18:25 < jsaw> ... 18:27 * mnemoc_ moving home to lunch 18:27 < mnemoc_> bbl 18:28 < valentin> how can i set the text color in inkscape ? 18:31 < valentin> jsaw: reload ? 18:32 < rxr> re 18:32 < valentin> looks a bit unbalanced with the url 18:32 < valentin> hi rxr, we are messing up your logo ,) 18:33 < valentin> http://exactcode.de/valentin/t2logoValentin.png 18:34 < valentin> oh - i should s/exactcode/t2-project/ .oO 18:34 < valentin> forgot that 18:35 < rxr> text color is the normal style color dialog ... 18:35 < valentin> reload once again 18:36 < valentin> rxr: i thought so - but where is the color dialog ? 18:36 < rxr> this style thing in thr toolbat 18:37 < rxr> this rectangle button left of the text-style one .. 18:37 * rxr away 18:37 < rxr> much fun styling ... 18:37 < valentin> ok, cu 18:39 < valentin> reload 18:39 * valentin away now, too 18:52 < jsaw> looks good 20:02 < mipe> ~/quit 20:02 < mipe> ups 20:02 -!- mipe [~mipe@dsl10040.japo.fi] has quit ["Leaving"] 20:28 < mnemoc_> re 20:50 < mnemoc_> http://ktoon.toonka.com/ 21:00 < rxr> re 21:01 < rxr> valentin: saw you mother driving past while walking home from the greek resteraunt here in Laueunau ... ;-) 21:09 < mnemoc_> Laueunau? 21:44 < rxr> lauenau of course ;-) 21:45 < mnemoc_> .oO( at least _that_ produces answers on google )o 21:46 < mnemoc_> what are you doing there? 21:47 < rxr> my parents life here ... 21:48 < mnemoc_> multimap.com is cool :D 21:48 < mnemoc_> you went to school there? 21:49 < mnemoc_> multimap.com is amazing 21:49 < rxr> in lauenau only for the 1 to 4h class .. 21:49 < rxr> later in other cities around ... 21:50 < mnemoc_> what a beatiful town^Wcity 21:51 < rxr> cu later 21:51 < rxr> yeah - and nowadays we hav DSL here ;-) *g* 21:51 < mnemoc_> :D 21:52 < mnemoc_> i would love to live in a place like that 22:06 < rxr> re 22:06 < rxr> really ? 22:07 < mnemoc_> yep 22:07 < rxr> btw. my current location is here: 22:07 < rxr> http://multimap.com/map/browse.cgi?client=public&X=1043000.61198844&Y=6816000.79259163&width=500&height=300&gride=1042876.61198844&gridn=6815952.79259163&srec=0&coordsys=mercator&db=DE&addr1=&addr2=&addr3=lauenau&pc=&advanced=&local=&localinfosel=&kw=&inmap=&table=&ovtype=&zm=0&scale=10000&multimap.x=381&multimap.y=0 22:08 < rxr> why wouldyou love to live in such a small "town" ? 22:12 < mnemoc_> i have cities 22:12 < mnemoc_> hate* 22:13 < mnemoc_> your town is beatiful, looks calmed, has conectivity :p and it's close to bigger and noisy places where you could get what you can't find there 22:13 < rxr> ,-) when do you plan to visit germany ? 22:13 < rxr> (to take a look in real life ...) 22:14 < CIA-9> rene * r6824 /trunk/target/psion-pda/arlo.cfg: * adapted arlo.cfg for minit 22:14 < mnemoc_> i guess i'll never have the money to go there :\ 22:14 < rxr> don't say never ... 22:15 < mnemoc_> :) 22:15 < rxr> or better never say never ;-) 22:16 < mnemoc_> hehe 22:17 * rxr starting the first uclibc build ever ... 22:18 < rxr> sed: can't read Rules.mak: No such file or directory 22:18 < rxr> .oO 22:19 < mnemoc_> uhm? 22:19 < rxr> uclibc 22:20 < rxr> ar -chf - include/* | tar -xf - -C /usr/src/t2-arm/build/pda-2.1.0-beta4-arm-ar 22:20 < rxr> m710t-cross-psion-pda/usr/; 22:20 < rxr> tar: This does not look like a tar archive 22:20 < rxr> tar: Skipping to next header 22:20 < rxr> tar: Read 9728 bytes from - 22:20 < rxr> tar: Error exit delayed from previous errors 22:20 < rxr> make: *** [install_dev] Error 2 22:20 < mnemoc_> uhm? 22:20 < rxr> yeeah - um 22:20 < rxr> why is tar used for the copy ? 22:21 < mnemoc_> i did? 22:21 < rxr> no idea? 22:21 < rxr> I just asked to the round here .... 22:21 < mnemoc_> preserving something? 22:21 < rxr> symlinks ? 22:22 < mnemoc_> yes, dates, permissions.... 22:22 < rxr> dates and permissions for header files ? 22:22 < rxr> youh 22:22 < rxr> svn blame on the uclibc.conf still has some clifford lines .. 22:23 < mnemoc_> uclibc cames from rock 22:23 < rxr> 2783 clifford if [ $ROCKCFG_CROSSBUILD = 1 ]; then 22:23 < rxr> 5539 mnemoc $SED -i "s,^CROSS=.*,CROSS=$bindir/${arch_target}-,g" Rules.mak 22:23 < mnemoc_> everything there belongs to clifford 22:23 < mnemoc_> rev 2783 22:23 < mnemoc_> that should be mine 22:24 < rxr> all your sources are mine *g* 22:24 < mnemoc_> :p 22:24 < rxr> http://exactcode.de/rene/sco-we-own-all-your-code-pay-us-all-your-money.png 22:24 < rxr> that line as no effect .. 22:25 < rxr> it does run sed on a file that does not exist at that time .. 22:25 < mnemoc_> :) 22:25 < mnemoc_> weird 22:25 < rxr> why - the h**l does the first uclibc build scream on me ... 22:25 < mnemoc_> or just 'old' 22:25 < mnemoc_> weird 22:25 < mnemoc_> it never screamed here 22:25 < rxr> I adapt to compile for me ... 22:25 < mnemoc_> ./scripts/Build-Target -cfg uboot 22:25 < rxr> what did you do to your lgibc build I broke ? 22:26 < mnemoc_> == 18:26:17 =[0]=> Building base/uclibc [0.9.27 2.1.0-beta4]. 22:26 < mnemoc_> == 02/26/05 18:26:34 =[0]=> Finished building package uclibc. 22:26 < mnemoc_> rxr: on second pass it built :\ 22:27 < mnemoc_> if [ $ROCKCFG_CROSSBUILD = 1 ]; ?? you are crossbuilding 22:27 < mnemoc_> it had never been tested crosscompiling 22:27 < mnemoc_> always native 22:27 < mnemoc_> that's why the old code didn't scream here 22:29 < rxr> oh yes 22:29 < rxr> I cross build ... 22:29 < rxr> I want to see the sizes for my ARM PDA ... 22:29 < rxr> I need something quite polished mostly for the CeBIT 22:29 < rxr> but also for me .. 22:29 < rxr> especially now with a tiny ssh client ... 22:30 < mnemoc_> are you using udhcp? 22:31 < mnemoc_> my t2-default.script needs to be polished :p 22:31 < mnemoc_> uhm 22:31 < mnemoc_> 0-gcc, gcc-300-force-visibility.patch ... 2 out of 2 hunks FAILED 22:32 < mnemoc_> .oO 22:33 < rxr> hm 22:33 < mnemoc_> it is applied _Twice_ 22:34 < mnemoc_> /usr/src/building/t2-trunk/package/base/uclibc/patches/gcc-300-force-visibility.patch 22:34 < mnemoc_> :) 22:34 < rxr> remove the uclibc one ... 22:34 < mnemoc_> sure 22:34 < mnemoc_> they are identical :p 22:35 < CIA-9> amery * r6825 /trunk/package/base/uclibc/patches/gcc-300-force-visibility.patch: * removed gcc/visibility patch from uclibc because now it is on gcc itself 22:37 < rxr> looking at uclibc.conf already makes me uncomfortable to switch to it ... 22:37 < mnemoc_> :( 22:37 < mnemoc_> is it too ugly? 22:37 < rxr> no comment ,-) 22:37 < mnemoc_> :'( 22:37 < mnemoc_> look at uclibc/parse-config :p 22:38 < rxr> does it make me happier? 22:38 < mnemoc_> look at it :) 22:38 < rxr> not witout an answer 22:39 < mnemoc_> it has three cute helpers on top 22:39 < mnemoc_> and a nice case/esac below 22:44 < rxr> == 02/26/05 22:43:55 =[0]=> Finished building package uclibc. 22:51 < rxr> kde-3.4-rc1 out 22:52 < mnemoc_> :) 22:59 < rxr> !> cc1: error: invalid option `align-loops=0' 22:59 < rxr> !> make[2]: *** [i386/resolve.o] Error 1 22:59 < rxr> == 02/26/05 22:57:57 =[1]=> Aborted building package uclibc. 22:59 < rxr> oh my ... 23:01 < rxr> does uclibc support powerpc64, sparc64 or x86-.64 ? 23:01 < mnemoc_> i686-t2-linux-uclibc-gcc -z now -Bsymbolic -shared --warn-common --export-dynamic --sort-common -z combreloc --discard-locals --discard-all 23:01 < mnemoc_> --no-undefined -s -e _dl_boot -soname=ld-uClibc.so.0 \ 23:01 < mnemoc_> -o ld-uClibc-0.9.27.so i386/resolve.o ldso.o /usr/src/building/t2-trunk/build/uboot-2.1.0-beta4-x86-pentium4-uboot-expert/TOOLCHAIN/tools.cross/crosscc/../lib/gcc/i686-t2-linux-uclibc/3.4.3/libgcc.a; 23:01 < mnemoc_> i686-t2-linux-uclibc-gcc: unrecognized option `-soname=ld-uClibc.so.0' 23:01 < mnemoc_> your s/ld/gcc/ thing 23:01 < mnemoc_> /usr/src/building/t2-trunk/build/uboot-2.1.0-beta4-x86-pentium4-uboot-expert/TOOLCHAIN/tools.cross/crosscc/../lib/gcc/i686-t2-linux-uclibc/3.4.3/../../../../i686-t2-linux-uclibc/bin/ld: crti.o: No such file: No such file or directory 23:01 < mnemoc_> collect2: ld returned 1 exit status 23:01 < mnemoc_> make[2]: *** [lib] Error 1 23:01 < rxr> my LD change ? 23:01 < mnemoc_> it built yestarday 23:02 < mnemoc_> .oO( well, not yesterday, 4 or 5 days ago :p )o 23:02 < rxr> can you try with my LD change reverted ? 23:02 < mnemoc_> ok 23:03 < mnemoc_> # svn merge -r6822:6821 . 23:03 < mnemoc_> svn: error while loading shared libraries: libaprutil-0.so.0: cannot open shared object file: No such file or directory 23:03 < mnemoc_> wtf? 23:03 < rxr> .oO 23:04 < mnemoc_> my fault 23:04 < mnemoc_> i mine -r apache on one window and i had an _old_ build output of apache in another 23:05 < mnemoc_> so i thought i had rebuilt it 23:05 < CIA-9> rene * r6826 /trunk/package/base/uclibc/uclibc.conf: 23:05 < CIA-9> * fixed uclibc for cross builds: not run sed on a non existing files 23:05 < CIA-9> even before source extraction - but pass the options to make 23:07 < rxr> uclibc even has a custmain ... 23:07 * mnemoc_ hides 23:07 < rxr> the config does not seem to work mostly ... 23:07 < rxr> the arch setting does not make it into it ... 23:08 < rxr> are you sure this X bla bla stuff does s.th. at all = 23:08 < rxr> ? even 23:08 < mnemoc_> it suppose to work 23:09 < mnemoc_> but i guess uclibc.conf doesn't know your arch 23:09 < rxr> it does 23:09 < rxr> # TARGET_ARCH 23:09 < rxr> cpu="`echo "$arch_machine" | sed -e s/i.86/i386/ `" 23:09 < rxr> echo "X TARGET_$cpu" 23:09 < rxr> echo "X TARGET_ARCH \"$cpu\"" 23:09 < rxr> and some more crap code ... 23:09 < rxr> and in the .config i386 is set ... 23:10 < rxr> where is this uclibc.cfg suppost to be in our sources ? 23:11 < mnemoc_> that's like linux.cfg, config dependant 23:11 < mnemoc_> is your $arch_machine a valid TARGET_ARCH on uclibc's .config ? 23:11 < rxr> yes 23:12 < mnemoc_> send me your error log please, i need to read that 23:13 < rxr> you do not need it ... 23:13 < rxr> the compiles does not want to take i386 code and options ... 23:13 < rxr> and the uclibc is configured for i386 ... 23:14 < mnemoc_> i want to read the config loop 23:14 < mnemoc_> btw, you should have .config.$i for each loop 23:14 < rxr> > 8. i386 (TARGET_i386) (NEW) 23:14 < mnemoc_> on the first .config ?? 23:15 < rxr> on any 23:16 < mnemoc_> what the hell is happening here today? my rebuild apache doesn't have any .so on it, only .a 23:16 < rxr> shared disaabled ? 23:16 < mnemoc_> no, never 23:17 < rxr> hm the /tmp/$$.config files has this: 23:17 < rxr> X TARGET_arm 23:17 < rxr> X TARGET_ARCH "arm" 23:17 < rxr> X ARCH_LITTLE_ENDIAN 23:17 < rxr> so I guess this strane stuff never worked that well - if at all ... 23:17 * rxr investigating 23:18 < mnemoc_> what about the sed file? 23:18 < rxr> btw. why the h*ll do I know have so many /tmp/*.confg files when they should be in the build dir ... 23:18 < rxr> what sed file ? 23:18 < rxr> the file not found by sed ? 23:18 < mnemoc_> for debuging 23:18 < mnemoc_> that code has two years 23:19 < rxr> it should be in src.uclibc.../.config or so .. 23:19 < rxr> now I have 6 files in /tmp and 6 src 23:19 < mnemoc_> that pkgsel-like file is used to create a sed file to alter default .config 23:19 < rxr> .uclibc ... 23:19 < rxr> I do not have the pkgsel like file 23:20 < mnemoc_> /tmp/$$.config 23:20 < rxr> the content is partly posted above 23:20 < mnemoc_> the one with the Xs 23:20 < rxr> arm is in there ... 23:20 < mnemoc_> that is a pkgsel-like file 23:21 < rxr> i tho9ught you meant the $confdir one I asked for ... 23:21 < rxr> next you want me to decustmain and clean uclibc, right ? 23:21 < mnemoc_> no, that is the .cfg to replace the default 23:21 < mnemoc_> i need to have svn running again here before 'thinking' anything 23:22 * mnemoc_ rebuilding apache with s/ld/gcc/ reverted 23:23 < mnemoc_> rxr: that pkgsel like think was 'designed' for kernel config two years ago, and i never ever got a reply about 23:23 < rxr> ic 23:23 < mnemoc_> rxr: i tested it on uclibc and busybox and worked 23:23 < mnemoc_> that was a bad time on rock 23:24 < _Ragnar__> *g* 23:24 -!- _Ragnar__ is now known as _Ragnar_ 23:24 < rxr> mnemoc_: what time was it? 23:24 < rxr> hi _Ragnar_ 23:24 < mnemoc_> 1.6 deprecation, etc.. 23:24 < rxr> .oO(was there a good time ...) 23:25 < mnemoc_> rxr: there are monthes on rock where you can get answers 23:25 < mnemoc_> rxr: those are good monthes 23:27 * mnemoc_ puzzled between heimdal/ldap ldap/heimdal/ldap uclibc s/ld/gcc/ Build-Package -update and a broken svn 23:27 < rxr> never change too much at a time ;-) 23:27 < mnemoc_> and a boss asking for his demo of my GNUe-based system 23:28 < CIA-9> rene * r6827 /trunk/package/base/uclibc/uclibc.conf: 23:28 < CIA-9> * removed the ugly /tmp/$$.config from uclibc and placed the file 23:28 < CIA-9> in $builddir/.config instead 23:28 < rxr> mnemoc_: what is your state of your GNUe based system? 23:28 < mnemoc_> rxr: your s/ld/gcc/ leave apache without .so files, they are back now 23:28 < mnemoc_> rxr: 60% 23:28 < rxr> ok - I revert 23:29 < rxr> ok - 60% is a good start ;-) 23:29 < mnemoc_> :) 23:30 < mnemoc_> does ./scripts/Emerge-Pkg -force apache suppose to honor -update mode? 23:30 < mnemoc_> -> Creating backup of old package data. <-- it seem so 23:32 < CIA-9> rene * r6828 /trunk/ (package/security/dropbear/dropbear.conf scripts/parse-config): 23:32 < CIA-9> * reverted global LD change (caused regressions) and instead placed 23:32 < CIA-9> it in the dropbear.conf 23:37 < rxr> ouhm - my last uclibc change does not work out that well 23:38 < mnemoc_> CROSS= or .config ? 23:39 < rxr> .config 23:39 < mnemoc_> why? it looks harmless 23:40 < rxr> builddir does not exist at that time 23:40 < rxr> and that is why the orig author most probalby choose /tmp ... 23:40 < rxr> I convert more to the path of custmain less ... 23:40 < rxr> to fix this 23:41 < mnemoc_> i was that author :p 23:41 < mnemoc_> move that code to a preconf hook :p 23:41 < rxr> and did I correctly guess your thoughts ? 23:41 < rxr> I already did 23:41 < mnemoc_> yep 23:42 < rxr> hm - my "just look how uclibc performs on my ARM" get's nto a uclibc support cleanup ... 23:43 < CIA-9> rene * r6829 /trunk/package/base/uclibc/uclibc.conf: 23:43 < CIA-9> * move immediately executed code in uclibc.conf that makes life hard 23:43 < CIA-9> into preconf 23:43 < mnemoc_> and looking at the code i guess it was written on pre-hooks age 23:43 < mnemoc_> pre-hook_add i mean 23:44 < mnemoc_> locale is very recient on uclibc, and that's the only hooked thing 23:44 < _Ragnar_> recient? 23:44 < rxr> recent 23:45 < mnemoc_> :p 23:45 < _Ragnar_> ah 23:45 < mnemoc_> sorry, too hot in here 23:45 < _Ragnar_> I thought svn up merges differences? 23:47 < mnemoc_> it applies a diff3 23:47 < mnemoc_> afaik 23:47 < mnemoc_> starmerge for tla and svk merges lots better 23:47 < _Ragnar_> it reverted my changes to parse-config-9 completely ;p 23:48 < mnemoc_> oh 23:48 < _Ragnar_> I should've commited that stuff earlier :( 23:49 < mnemoc_> i svn diff > backup.$$.diff before anything like a merge :p 23:51 < rxr> svn should not looe conflicting changes ... 23:51 < rxr> are you sure you have not parse-config-9.mine .left .right or so files ? 23:58 -!- lars__ [~lars@port-212-202-42-223.dynamic.qsc.de] has joined #t2 23:59 < rxr> hi lars__ !! 23:59 < lars__> hi, as a remember right there was a possibility to create patches automatically in ./debug.sh environment? 23:59 < lars__> hi, rene 23:59 < rxr> when you use fixfile --- Log closed Sun Feb 27 00:00:02 2005