[06:16] <fabbione> morning
[08:16] <lamont> i386 90.2
[08:16] <lamont> powerpc 89.6
[08:16] <lamont> amd64 89.1
[08:16] <lamont> ia64 88.7
[08:16] <lamont> sparc 74.1
[08:16] <lamont> hppa 64.8
[08:16] <fabbione> ehhe
[08:16] <fabbione> lamont: weren't you supposed to be sleeping?
[08:17] <lamont> busted,
[08:17] <fabbione> lamont: btw.. the info i gave you yesterday.. are they enough or you need more?
[08:17] <lamont> btw, gcc-{3.3,3.4,4.0} are in the archive for hppa
[08:17] <lamont> or will be tonight.. 4.0 is uploading
[08:17] <fabbione> ok
[08:17] <lamont> I think it's enough
[08:17] <fabbione> do you still want hppa on 3.4 or 3.3?
[08:18] <lamont> hppa will stay with the pack
[08:18] <fabbione> (btw.. he is testing a "clean" linux kernel.. we don't patch for ipv6)
[08:18] <lamont> so if you move everyone, move hppa.  otherwise, it does just fine with 3.4
[08:18] <fabbione> ok
[08:18] <fabbione> well i am more for keeping all in sync
[08:18] <fabbione> so we have only 2 exceptions.. ppc64 needs 3.4 and ia64 needs 3.3
[08:18] <lamont> and even the ia64 issue is quite likely not gcc-3.4, but rather something else...
[08:18] <fabbione> i am waiting one guy to test sparc
[08:19] <lamont> it would help my laziness if a daily d-i build picked up a 3.3-based kernel.  But that's just me being lazy...
[08:19] <lamont> I can netboot to test it quite easily as well...
[08:19] <fabbione> lamont: don't you have debian installed there?
[08:20] <lamont> yep
[08:20] <fabbione> well you can install our kernel there...
[08:20] <lamont> it has a debian partition, and an ubuntu-hoary partition
[08:20] <fabbione> at least it should be possible
[08:20] <fabbione> so why do you need to netboot?
[08:20] <lamont> because it's less steps to netboot than to boot, copy, reboot, curse, reboot, iterate
[08:21] <fabbione> eheh
[08:21] <lamont> but not many less, since that's also the build machine for me to do test builds on...
[08:21] <fabbione> if i get the time i will build the kernel for you
[08:21] <lamont> so not much diff
[08:21] <fabbione> halley is pretty fast
[08:21] <lamont> iz not a big deal - I just need to actually make time to start the build - it'll build faster on my machine than in the DC anyway, so ...
[08:22] <fabbione> ah
[08:22] <fabbione> ok :)
[08:22] <lamont> but for now, back to bed.  binutils is chunking along, and I don't want to wait for it's build to finish just to see how dbus did
[08:22] <fabbione> hehe
[08:22] <fabbione> good night
[08:27] <doko> lamont: the next gcc-defaults update to unstable can be synced, and will make gcc-4.0 the default again
[08:29] <fabbione> doko: any news about the PPC ICE?
[08:30] <doko> no, not yet.
[08:30] <doko> btw, do you really need/want gcc-3.3 biarch? it looks doable, but it costs some time ...
[08:30] <fabbione> doko: no.. don't worry
[08:30] <fabbione> i will manage it in another way
[08:30] <fabbione> if it will become an issue i will tell you per time
[08:31] <fabbione> oh new gcc4 is almost done on sparc :)
[08:31] <fabbione> doko: promise no more gcc4 upload for the week, ok? ;)
[08:32] <doko> fabbione: next one will be needed, when seb128 uploads cairo 0.6
[08:33] <fabbione> doko: yeah.. just not in the next 2 hours :)
[09:47] <doko> fabbione: the ppc problem was 32bit?
[09:49] <fabbione> both
[09:49] <fabbione> 32 and 64
[09:50] <doko> hmm, ok. I did prepare a new gcc-snapshot, but this one is not yet biarch. I'll try to reproduce it on davis, when it's built.
[10:06] <fabbione> doko: if i have a dynamically linked binary.. what command should i use to make it static?
[10:11] <doko> fabbione: link using gcc -static
[10:12] <fabbione> gcc -static ld
[10:12] <fabbione> ../sysdeps/i386/elf/start.S:115: undefined reference to `main'
[10:12] <fabbione> collect2: ld returned 1 exit status
[10:12] <fabbione>  gcc -static ld
[10:12] <fabbione>  /usr/lib/gcc/i486-linux-gnu/4.0.1/../../../crt1.o: In function `_start':
[10:12] <fabbione> i mean.. ld is already an elf binary
[10:13] <fabbione> bah i hate to change build stuff
[10:17] <doko> ehh, no you can't change an existing executable, AFAIK
[10:18] <fabbione> so i guess i need to do 2 builds to get one shared and one static...
[10:19] <doko> or just find the line, and call make explicitely to link ld statically
[10:20] <fabbione> hmm yeah.. that would work too, but it needs to be multi-archaware i guess
[03:31] <lamont> doko: re 4.0 defaults, I believe the current source already does that too.
[03:32] <fabbione> hey lamont
[03:32] <fabbione> i just uploaded 4.4
[03:32] <fabbione> so you can give it a shot on ia64 + gcc-3.3
[03:32] <fabbione> it builds fine on hppa.. so you don't need to babysit it
[06:39] <infinity> doko : gcc-snapshot is FTBFS on amd64, whining about lib32 issues.
[08:16] <desrt> fabbione; everything is working perfectly.  thank you very much for your efforts.
[08:21] <doko> infinity: yes, seen it. the default is now to build biarch, so yo have to disable it explicitely
[08:27] <doko> lamont: what's wrong with 4.0?
[08:38] <lamont> ??
[08:38] <lamont> doko: ECONTEXT
[08:54] <doko> lamont: you were complaining about gcc-4.0, but people seem to consider this good style, so don't care ;-)
[09:15] <lamont> doko: no.  I was complaining about glibc
[09:15] <lamont> and how it's b0rked
[09:43] <doko> infinity: I did close the "poor GCC install" report again
[10:17] <doko> fabbione: elmo did install a biarch gcc-snapshot on davis/breezy-ppc64, please let me have some cpu time ...