--- Log opened Sat Mar 10 00:00:35 2007 00:16 -!- emte_ [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 00:16 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 110 (Connection timed out)] 02:15 -!- emte_ [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 110 (Connection timed out)] 02:25 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 04:02 -!- dj [n=dj@c219104.adsl.hansenet.de] has joined #t2 04:54 -!- Netsplit calvino.freenode.net <-> irc.freenode.net quits: TobiX 04:55 -!- Netsplit over, joins: TobiX 05:07 -!- Netsplit calvino.freenode.net <-> irc.freenode.net quits: TobiX 05:08 -!- Netsplit over, joins: TobiX 05:20 -!- dj [n=dj@c219104.adsl.hansenet.de] has quit [Remote closed the connection] 06:53 -!- emte_ [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 06:55 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 110 (Connection timed out)] 07:19 -!- emte__ [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 07:19 -!- emte_ [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 110 (Connection timed out)] 09:16 -!- AStorm [n=astralst@chello084010114027.chello.pl] has joined #t2 10:24 -!- sepp [n=sepp@p85.212.29.199.tisdip.tiscali.de] has joined #t2 10:24 < sepp> moin :) 10:42 -!- emte__ [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 145 (Connection timed out)] 10:44 < AStorm> moin 10:44 < AStorm> I've a short question: what exactly is installed in the "glibc headers" step of T2? 10:45 < sepp> "glibc headers" ? 10:45 < sepp> maybe linux header? 10:46 < AStorm> No no. 10:46 < AStorm> The glibc step in stage0 10:46 < AStorm> It installs just the headers, right? 10:47 < sepp> yes 10:47 < AStorm> I'll have to know what's the command, or is that just a smart cp? :> 10:48 < sepp> it does a 'make install-headers' 10:48 < sepp> you can just run that 10:48 < sepp> no need to mess with it 10:48 < sepp> :) 10:49 < sepp> run build-target 10:49 < AStorm> Ah yes. 10:49 < AStorm> No no, I'm building from the command line, myself 10:49 < AStorm> Some 2.5 snapshot from today. 10:50 < AStorm> (the last one I tried was compile-time botched) 10:51 < sepp> why would you want to build everything yourself? 10:51 < sepp> might be boring to type make 500 times 10:51 < sepp> :) 10:52 < AStorm> 500? :P 10:52 < AStorm> You do underestimate my build system. 10:52 -!- emte__ [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 10:52 < AStorm> That'd be more like "install-world" 10:52 < AStorm> But I need a bootstrap toolchain first. 10:55 < Stelz> hi sepp, AStorm 10:55 < Stelz> rxr: ping 10:56 < sepp> hi Stelz 11:22 -!- emte__ [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 110 (Connection timed out)] 11:35 < rxr> re 11:35 < Stelz> rxr: 11:35 < Stelz> moin 11:35 < Stelz> and 11:35 < rxr> moin 11:35 < Stelz> 23:50'21 | Stelz> rxr: ! we need to edit a link for a download location @ DistroWatch. http://www.t2-project.org/download.html -> http://www.t2-project.org/download/ 11:36 < rxr> oh 11:36 < Stelz> :-) 11:38 < Stelz> rxr: and also contact: http://www.t2-project.org/contact.html -> http://www.t2-project.org/contact/ 11:40 < rxr> yeah - I see 11:40 < rxr> I'll send them an email 11:41 < AStorm> There will be a new gcc release soon... 4.2.0 11:41 < AStorm> rc1 is upcoming 11:41 < Stelz> uhh.. 11:42 < AStorm> With less regressions than 4.1.2. Funny, eh? 11:42 < rxr> mail sent 11:43 < rxr> GCC is a complex beast 11:43 < rxr> I would not like to do their job ... 11:44 < AStorm> Complex is a gross understatement. 11:44 < AStorm> It requires a genius to hack a compiler that advanced. 11:47 < rxr> the only thing I really regret is that the beast becomes slower every release. 11:47 < rxr> especially annoying as it requires a new multi core box every year for our regression testing ... 11:48 < AStorm> 4.2.0 is faster than 4.1.2 :D 11:48 < AStorm> which was faster than 4.0.x 11:48 < CIA-8> rene * r22893 /trunk/package/base/upstart/upstart.desc: * updated upstart (0.3.5 -> 0.3.7) 11:49 < AStorm> So, the trend is getting better. 11:49 < rxr> hm? not for me :-)( 11:49 < rxr> not faster that is ... 11:50 < AStorm> hm? 11:50 < AStorm> Well, my benches are getting better with each 4.x release 11:50 < AStorm> though these are still slower a bit (4.2.0) than 3.3.x 11:50 < AStorm> (though not than 3.4.x) 11:51 < AStorm> They're the fastest with -O0 and -O1 though :> 11:51 < sepp> new version, new gb ram needed to compile a c++ template? :P 11:52 < AStorm> Nope. 11:52 < AStorm> :-) 11:52 < AStorm> Less than 4.1 and 4.0 fortunately 11:53 < AStorm> (thanks to new and better data structures) 11:55 < AStorm> 4.3 will be even lighter - offloads some work to GMP and MPFR - and now requires them 11:56 < CIA-8> rene * r22894 /trunk/package/develop/git/git.desc: * updated git (1.5.0.2 -> 1.5.0.3) 11:56 < sepp> sure? 11:56 < CIA-8> rene * r22895 /trunk/package/develop/uncrustify/uncrustify.desc: * updated uncrustify (0.32 -> 0.33) 11:56 < AStorm> sepp, sure sure. 11:56 < AStorm> They've dumped some obsolete optimisation stages too :> 11:57 < AStorm> (with a new replacement) 11:57 < AStorm> Probably, 4.3 will be to 4.2 like 4.1 was to 4.0 11:57 < AStorm> straight better :P 11:58 < AStorm> Oh, major changes even :P 11:58 < AStorm> "gcj now uses the Eclipse Java compiler for its Java parsing needs. This enables the use of all 1.5 language features, and fixes most existing front end bugs." 11:58 < sepp> well, so far it seems to become slower with every release and consume much more mem ... 11:58 < AStorm> Hmm? 11:58 < AStorm> With every major release, you say? :P 11:58 < AStorm> 3.x vs 4.x, it's true 11:59 < AStorm> 4.0.x vs 4.1.x, not 11:59 < CIA-8> rene * r22896 /trunk/package/filesystem/udev/udev.desc: * updated udev (105 -> 106) 12:00 < AStorm> Well... 4.3 will rock if they fix this: 12:00 < AStorm> http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30735 12:01 < AStorm> 40 PRs not in older versions currently 12:06 < AStorm> 4.2 has like... 10. 12:06 < AStorm> And mostly weird. 12:10 < AStorm> Heh, 4.2 has some perf. regressions 12:10 < AStorm> (in generated code) 12:14 < AStorm> Also, they're uncovering layers and layers of latent bugs :P 12:14 < AStorm> (since 4.x) 12:44 < rxr> so - I have to buy some new screwdrivers ... back in some minutes ... 13:24 -!- sparc-kly [n=mubex@adsl-64-237-249-210.prtc.net] has joined #t2 13:33 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 13:58 -!- sepp_ [n=sepp@p85.212.74.17.tisdip.tiscali.de] has joined #t2 14:03 < AStorm> Ok, now I've caught the cause of the failure 14:03 < AStorm> Some specs have to be mutated in gcc. 14:04 < AStorm> I find it funny that CLFS does with a patch, what can be done with a simple sed :P 14:05 < AStorm> find gcc/config -name \*.h | xargs sed 's:\(DYNAMIC_LINKER\(\|32\|64\) "\)/lib:\1/tools/lib:' -i 14:06 < AStorm> (older GCCs require one more sed) 14:11 < AStorm> GCC 4.2.0 also defines some more specs properly, not hardcoded. 14:12 < AStorm> And they don't use the -i parameter to sed :P 14:15 -!- sepp [n=sepp@p85.212.29.199.tisdip.tiscali.de] has quit [Read error: 110 (Connection timed out)] 14:20 -!- idealm [n=ideal@CPE-203-45-99-98.nsw.bigpond.net.au] has quit [Remote closed the connection] 14:36 -!- dj [n=dj@c158216.adsl.hansenet.de] has joined #t2 14:41 -!- d-marc [n=d-marc@HSI-KBW-091-089-001-157.hsi2.kabelbw.de] has joined #t2 14:41 < d-marc> hi 14:42 < rxr> hi d-marc 14:42 < rxr> we do not yet have scratchbox a-like yet 14:43 < rxr> we rather want to cross compile as much as possible 14:43 < rxr> but if someone would integrate support that would be fine asw ell 14:43 < d-marc> even better 14:44 < d-marc> scratchbox sucks really bad 15:48 < TobiX> A friend of mine uses Scratchbox for a project... He needs to run it in a Debian chroot/qemu because it does not run on Gentoo... 15:52 < AStorm> Drat. 15:52 < AStorm> Again... 15:52 < AStorm> Probably I can't build the 32-bit glibc, because I do not have 32-bit tools? :P 15:52 < AStorm> Maybe I have to build some :P 15:57 < AStorm> Well, it should build anyway... 15:59 < AStorm> yet, it doesn't. 15:59 < AStorm> I wonder why... 15:59 < AStorm> Hm. 15:59 < AStorm> Maybe it's just because I tried using 64-bit system to build 32-bit tools, and the guide is flawed :P 16:01 < AStorm> If you will: 16:01 < AStorm> http://cross-lfs.org/view/svn/x86_64/cross-tools/glibc.html 16:02 < AStorm> Should work, but doesn't :/ 16:02 < AStorm> 1. errors like: 16:04 < AStorm> error: "TLS support is required" 16:04 < AStorm> When I "fixed" that with a forced define, then 16:04 < AStorm> ../sysdeps/unix/sysv/linux/i386/sysdep.S: Assembler messages: 16:04 < AStorm> ../sysdeps/unix/sysv/linux/i386/sysdep.S:35: Error: invalid character '_' in mnemonic 16:06 < AStorm> I have no idea at all. The system is to be a multilib x86_64 16:06 < AStorm> (plain x86_64 built fine, BTW) 16:07 < Capey> hmm, whats this? you trying to build 32bit glibc with multilib disabled? 16:09 < AStorm> With multilib enabled. 16:09 < AStorm> But I'm baffled. 16:09 < AStorm> Well... now 64-bit build also fails with the exact same message (?!?) 16:10 < AStorm> While it went fine with T2... 16:10 < d-marc> TobiX: I couldn't install that tgz version on debian either, only the debs 16:13 < AStorm> Hel. 16:13 < AStorm> *hell 16:13 < AStorm> The thing should _just_ _work_ 16:15 < d-marc> AStorm: the first or the second thing? 16:16 < AStorm> Both :P 16:16 < AStorm> T2 should build, yes. 16:16 < AStorm> And the CLFS too (it does everything properly it seems) 16:24 < AStorm> Binutils bug, or something totally different? 16:25 < AStorm> Hehe, now I've got it :P 16:25 < AStorm> Stupid thing builds binutils w/o threads support. 16:26 < AStorm> I wonder why would the glibc build then ever. 16:27 < AStorm> I meant gcc :> 16:32 < AStorm> So I have to install glibc headers, then build binutils with threads support, and CLFS is broken. 16:32 < AStorm> T2 is too, probably :P 16:43 < CIA-8> susan * r22897 /trunk/package/x11/glew/glew.desc: * updated glew (1.3.5 -> 1.3.6) 16:43 < CIA-8> susan * r22898 /trunk/package/gnome2/gtkmm/gtkmm.desc: * updated gtkmm (2.10.7 -> 2.10.8) 16:43 < CIA-8> susan * r22899 /trunk/package/gnome2/glib/glib.desc: * updated glib (2.12.9 -> 2.12.11) 16:44 < CIA-8> susan * r22900 /trunk/package/gnome2/glibmm/glibmm.desc: * updated glibmm (2.12.6 -> 2.12.7) 16:45 < AStorm> Bonus points for glibc - forgot to install include/gnu/stubs.h :> 16:47 < AStorm> Damn, the assembler still hates me. 16:47 < AStorm> :| 16:49 * Capey curses busybox 16:49 < Capey> it just doesnt want to cross build 16:52 < AStorm> I wonder _why_ does the assembler bitch. 16:59 < AStorm> Ok, I'm done :P 17:00 < AStorm> Latest 2.17 branch of binutils is... broken. 18:06 -!- misl [n=chatzill@84-104-172-187.cable.quicknet.nl] has joined #t2 18:06 < misl> hi folks 18:50 < Capey> hi misl 18:51 < misl> hi Capey :-) 18:51 < Capey> damm, i bought wrong type of A/V card... looks like pinnacle is not well or at all supported in kernel 18:54 < Capey> rxr: what would you think if i start upgrading uclibc to latest svn rev 18:54 < Capey> rxr: i think back porting those needed features to current one awful lot of work 18:55 < Capey> rxr: well, upgrading needs work too since quick try most of the current patches in trunk does not apply 19:23 -!- sepp [n=sepp@p85.212.16.14.tisdip.tiscali.de] has joined #t2 19:23 -!- sepp_ [n=sepp@p85.212.74.17.tisdip.tiscali.de] has quit [Read error: 104 (Connection reset by peer)] 19:36 -!- sparc-kly [n=mubex@adsl-64-237-249-210.prtc.net] has quit [Read error: 104 (Connection reset by peer)] 20:15 -!- dj [n=dj@c158216.adsl.hansenet.de] has quit [Remote closed the connection] 20:23 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 110 (Connection timed out)] 20:23 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 20:27 < AStorm> Huh. 20:27 < AStorm> No, 2.17 isn't broken :D 20:27 < AStorm> It was just that glibc 2.5-branch cannot be built with binutils 2.16.1 :P 20:31 < AStorm> So, you just can't build a cross. 20:32 < AStorm> First, you have to update the binutils :P 20:34 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 145 (Connection timed out)] 20:35 < AStorm> That doesn't change the fact that Gentoo toolchain is plain broken :> 20:44 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 21:10 < AStorm> Ok, Gentoo devs lost 1:0 again :P 21:17 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has quit [Read error: 110 (Connection timed out)] 21:18 -!- emte [n=emte@d64-180-45-14.bchsia.telus.net] has joined #t2 21:26 -!- d-marc [n=d-marc@HSI-KBW-091-089-001-157.hsi2.kabelbw.de] has left #t2 ["Time makes no sense"] 21:30 -!- sparc-kly [n=mubex@adsl-64-237-249-210.prtc.net] has joined #t2 21:40 < rxr> rehi 21:40 < rxr> Capey: no uclibc trunk has many new regressions 21:40 < rxr> we downdated already once 21:40 < rxr> our trunk uclibc is pretty fine - what does make you think there's a problem ? 21:41 < rxr> and I think mtr also does use it on MIPS 21:41 < rxr> valentin: I managed to repair the Psion 5MX display 21:42 < rxr> I hope the tweaked flex plastic wire ... 21:42 < rxr> survives the rebuilding again 23:20 -!- Stelz [i=stelz@unaffiliated/stelz] has quit ["I wanna live my life like thunder!"] 23:21 -!- Stelz [i=stelz@unaffiliated/stelz] has joined #t2 --- Log closed Sun Mar 11 00:00:37 2007