[12:16] <Mithrandir> fabbione: do you have any xen packaging done or should I start from scratch on it?
[02:12] <jbailey> zul: Angie just came from Ottawa, said it was brutal.
[04:55] <zul> it is brutal..
[04:56] <zul> 26 degrees 79% relative humitidy 79%
[04:57] <zul> tomorrow its going to feel like 39
[05:05] <fabbione> morning
[05:05] <zul> hey fabbione 
[05:05] <fabbione> hey zul
[05:05] <zul> how goes it?
[05:05] <fabbione> just woke up
[05:05] <zul> nifty
[05:06] <fabbione> zul: is there anything i need to merge from your tree?
[05:06] <zul> i dont think so..
[05:07] <zul> nothing that cant wait
[05:09] <fabbione> nah it's an easy merge
[05:10] <zul> right im off to bed..
[05:11] <zul> i might be back later if i cant sleep in this freaking heat..
[05:13] <fabbione> night zul
[05:14] <fabbione> Mithrandir: xen packages are in experimental already
[05:14] <fabbione> Mithrandir: the userland and the boot loader at least
[05:15] <fabbione> Mithrandir: you will still need the kernels tho
[05:15] <fabbione> lamont: ping?
[05:18] <lamont> fabbione: yes?
[05:18] <fabbione> hey dude
[05:18] <fabbione> you almost totally disappeared from earth :)
[05:33] <zul> its way too hot to sleep
[06:04] <jbailey> fabbione: Hey, bored? =)
[06:04] <fabbione> jbailey: gimme some crack :)
[06:06] <jbailey> fabbione: I uploaded 0.7 of initramfs-tools, wondering if you could try it on your sparc boxes.
[06:06] <fabbione> s/boxes/box
[06:06] <fabbione> and no.. 2.6.12 is no go yet on sparc
[06:06] <fabbione> sorry :(
[06:07] <jbailey> No worries.  It does work with older kernels still.
[06:07] <jbailey> Just need to specify which modules to load into the initramfs
[06:07] <fabbione> hmm ok..
[06:07] <fabbione> i will try that soon
[06:07] <fabbione> not today probably
[06:07] <jbailey> No worries. =)
[06:08] <jbailey> When you do try it, let me know and I can tell you what configurations I need most tested. =)
[06:09] <fabbione> jbailey: let's try to schedule it together
[06:09] <fabbione> since we can share a screen session on the console
[06:09] <jbailey> Serial console?
[06:09] <fabbione> yup
[06:10] <jbailey> Coo.
[06:10] <fabbione> it doesn't have anything else this sparc :)
[06:10] <jbailey> I should also get off my ass and buy a switch.
[06:10] <fabbione> it's a U1
[06:10] <fabbione> brb i need to wake up my wife
[06:21] <fabbione> re
[09:21] <fabbione> hey chmj 
[09:21] <fabbione> chmj: it would be a good idea if you can complete the uscan-extmod thingy asap
[09:22] <fabbione> specially the file with the data info
[09:22] <Mithrandir> fabbione: I was thinking about the kernel packaging
[09:23] <fabbione> Mithrandir: and didn't you get a nightmare?
[09:23] <Mithrandir> fabbione: no, should I?
[09:23] <fabbione> ehhe
[09:23] <chmj> fabbione, its complete, just the url file need to be reformated 
[09:24] <fabbione> chmj: ok can you please finish it?
[09:25] <fabbione> chmj: you also told me that the url file is not completed, but only a "sample" of the external-drivers
[09:25] <fabbione> Mithrandir: and what are you toughts?
[09:25] <Mithrandir> fabbione: on the xen build process?  It's utterly crackful ATM.
[09:26] <fabbione> Mithrandir: why?
[09:26] <Mithrandir> it downloads random stuff off the intarweb
[09:26] <fabbione> oh yeah, but that's easy to patch/fix
[09:28] <Mithrandir> I found nightly snapshots so broken sourcepuller isn't that big of a problem
[09:29] <chmj> fabbione, will do 
[09:29] <fabbione> chmj: thanks
[09:31] <fabbione> bah this udeb stuff is a real mess
[09:31] <fabbione> too many files in too many different places
[09:31] <fabbione> this doesn't work
[01:07] <fabbione> this is an interesting failure
[01:08] <fabbione> ppc kernel does build properly on the buildd
[01:08] <fabbione> it fails on davis
[01:08] <fabbione> anybody can do a manual build at home?
[01:08] <fabbione> both dpkg-buildpackage -rfakeroot -uc -us -b -B
[01:08] <fabbione> and fakeroot make -f debian/rules binary-arch
[01:08] <fabbione> please?
[01:29] <chmj> ppc ?
[01:29] <chmj> I don have 
[02:11] <jbailey> fabbione: Still need a ppc build?
[02:11] <fabbione> jbailey: pretty much yes
[02:12] <jbailey> What package to pull down?
[02:13] <fabbione> linux-source-2.6.12
[02:13] <fabbione> actually ...
[02:13] <fabbione> yes please do so
[02:15] <jbailey> Great.  Hopefully it'll be quick.
[02:15] <jbailey> Is the SMP aware?
[02:38] <zul> morning
[02:59] <jbailey> Heya Chuck
[03:10] <zul> hey jeff
[04:02] <zul> how the hell do you import a directory into baz?
[04:03] <chmj> import or add ? 
[04:04] <zul> import
[04:07] <zul> oops...never mind figured it out
[04:28] <jbailey> zul: bzr add directory
[04:28] <jbailey> zul: =)
[04:31] <zul> meh..
[04:36] <zul> fabbione: you can get my first version of linux-non-supported-modules in my baz archive under linux-non-supported-modules--pre1,1--2.6.23 and the external drivers that i used is http://zulinux.homelinux.net/arch/lnsm/
[04:36] <zul> its very very beta though
[04:42] <chmj> jbailey, I just commited, can you please check that nothing is broken ?
[04:42] <jbailey> chmj: Committed what to where, sorry?
[04:42] <zul> aiieeeeeeeeee...:)
[04:42] <chmj> kernel-team 
[04:42] <chmj> committed kernel-team@ubuntu.com--2005/kernel-debian--pre1,3--2.6.11.93--patch-10
[04:43] <zul> uh...2.6.11.94 should it be?
[04:43] <chmj> arg ! 
[04:44] <chmj> i knew i missed something 
[04:44] <zul> well at least you didnt break it yet :)
[04:48] <chmj> what are those numbers btw -> kernel-team@ubuntu.com--2005/kernel-debian--pre(.*)--
[04:49] <zul> pre-release
[04:49] <zul> jbailey: im still getting thoese i486-linux-gnu-gcc: command not found
[04:50] <jbailey> chmj: I don't do kernel-team stuff really.  Chuck or Fabio is a better bet.
[04:50] <jbailey> That and I don't know how to work baz. =)
[04:50] <zul> lovely...something is still fucked..
[04:51] <jbailey> zul: For local testing, make a symlink.  It should get overwritten when the new gcc is in.
[04:51] <zul> okie dokie
[04:51] <jbailey> doko: This is true, yes?
[04:52] <chmj> I desperately need bandwidth 
[04:52] <zul> chmj: where do you live?
[04:52] <jbailey> chmj: Can't work remotely on one of the Canonical boxes?
[04:53] <chmj> zul, South Africa 
[04:53] <doko> jbailey: yes
[04:53] <chmj> jbailey, too slow 
[04:53] <zul> chmj: cool i lived in east africa for a couple of years
[04:54] <chmj> zul, no, not cool, not cool at all
[04:54] <chmj> not if you are in IT and need bandwith 
[04:56] <zul> heh..
[04:56] <zul> wheee....kernel-package cant determine subarch now
[05:05] <zul> jbailey: new kernel-package fixes the i486 stuff
[05:50] <svenl> fabbione: do you include a patch that fixes the sleep issues on powerbook in the latest linux-images ?
[05:50] <svenl> or not yet.
[05:58] <fabbione> zul: no it's a gcc issue
[05:59] <fabbione> svenl: mostlikely not. i am tracking upstream atm and what benh push me
[06:00] <fabbione> zul: in a second
[06:00] <svenl> fabbione: so, question is, did benh push you something or push something upstream for this powerbook don recover from sleep problem yet ? :)
[06:00] <fabbione> zul: i will look at the package
[06:01] <svenl> fabbione: but as i guess the reply is no, i rather not use that kernel just yet on my powerboook :)
[06:05] <zul> fabbione:   http://incoming.debian.org/kernel-package_9.001_i386.changes
[06:07] <fabbione> zul: that's weird, because there are other packages failing for that reason
[06:07] <lamont__> while mumbling something about new drivers magically showing up in the config files...
[06:08] <fabbione> anyway it's tomorrow work
[06:08] <lamont__> r2[45] 00, hostap, etc
[06:08] <lamont__> etc== squashfs, iirc
[06:09] <fabbione> lamont__: i did probably add most of them automatically
[06:09] <fabbione> zul: but but... did you make an entire new package for l-n-s-m ????
[06:10] <zul> no i didnt..the tar.gz has the drivers that i used minus the debian directory the debian directory is in baz
[06:12] <fabbione> zul: hmm ok.. i will look at it tomorrow
[06:12] <zul> builds for me on 386 at least..
[06:14] <fabbione> zul: ok thanks
[06:15] <fabbione> lamont__: anyway i plan very soon to do the magic config allignment
[06:16] <fabbione> because it's too much of a mess right now
[06:17] <fabbione> zul: so you did create another package :)
[06:17] <fabbione> ok.. we need to find a way to integrate that into the kernel build system
[06:17] <zul> kind of yeah..
[06:17] <zul> its based off linux-restricted..
[06:18] <fabbione> zul: right, i understand and the idea is good
[06:19] <zul> buuuuut....:)
[06:20] <lamont__> fabbione: I'll commit my changes shortly- waiting for at least one of the 4 kernels to build
[06:20] <lamont__> should be another 10-20 minutes
[06:20] <fabbione> lamont__: sure.. i am done for today
[06:20] <fabbione> 9 hours of d-i sync is a pain
[06:20] <fabbione> and i am not even half way
[06:30] <fabbione> cya tomorrow guys
[06:30] <fabbione> :)
[07:35] <mjg59> Hmm. Need to play with the dynamic-HZ stuff at some stage.
[09:13] <zul> yay another interview
[09:21] <jbailey> fabbione: The kernel build has died with it prompting me for:
[09:22] <jbailey> Thermal Management Support (TAU) [N/y/?]  (NEW)
[09:22] <fabbione> jbailey: hmmmmm
[09:22] <fabbione> same as davis
[09:22] <fabbione> so now the question is... why the fuck does it build on the buildd???
[09:22] <jbailey> No tty, probably
[09:22] <fabbione> i don't think that's the problem
[09:22] <fabbione> if you just hit enter it will fail later
[09:23] <jbailey> Do you want me to do so?
[09:23] <fabbione> if you want to try
[09:23] <jbailey> No skin off my teeth. =)
[09:23] <fabbione> i wonder if that can be a consequence of running a 64bit kernel
[09:23] <fabbione> i never saw that using a 32bit one
[09:23] <fabbione> jbailey: if you could try to reboot your ppc later in 32 bit mode
[09:23] <fabbione> and rebuild, that would be nice
[09:24] <jbailey> Sure.
[09:31] <jbailey> Ugh.  glibc testsuite managed to do something nasty to the ppc64 kernel.
[09:31] <fabbione> jbailey: nasty to what level?
[09:31] <Mithrandir> buggy kernels.  Gotta love 'em
[09:31] <lamont__> fabbione: kernle 4 of 4 building now... I think I'm going to commit
[09:32] <fabbione> lamont__: great
[09:32] <jbailey> fabbione: Can't run sudo, during reboot init couldn't should down the MTA, had to hit the power button.
[09:32] <fabbione> ah
[09:32] <fabbione> anything in the logs?
[09:32] <lamont__> _GAH_ - lossy link to the house today.  interactive performance sucks
[09:32] <jbailey> fabbione: It's just coming back up.
[09:32] <fabbione> jbailey: i am really not 100% sure how stable is 12rc6 on ppc64
[09:33] <jbailey> Well, tst-ifaddrs randomly hangs at the best of times.  I've just never seen it take out the box before.
[09:34] <dilinger> lamont__: salting the holy penguin pee is the trick for keeping 'em stable?
[09:34] <jbailey> dilinger: It's like beer nuts.  It makes it drink more.
[09:34] <jbailey> With alcohol, code flows.
[09:36] <jbailey> Nothing in syslog
[09:36] <zul> jbailey: yeah /* XXX - drunk fix later */
[09:36] <jbailey> /* XXX - here be pr0n */ char *firmware_blob {  ...
[09:37] <zul> brb...need to reboot
[09:37] <dilinger> jbailey: feh, more like /* we're not allowed to redistribute this, but we can fix it later... */ char *firmware_blob = ...
[09:40] <lamont__> dilinger: nah, you just have to take them with a grain of salt sometimes...
[09:41] <lamont__> oops.
[09:43] <zul> this is interesting http://kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=8d5f7b4353dae4c7ee342c61303372fd996ca161
[09:48] <dilinger> hm
[09:48] <dilinger> won't that keep artifacts, then?  shouldn't it check if fb console has been initialized, and memset only if that's the case?
[09:49] <lamont__> * committed kernel-team@ubuntu.com--2005/kernel-debian--pre1,2--2.6.11.94--patch-18
[09:54] <zul> fabbione: ping
[09:57] <dilinger> zul: fyi, looks like neofb does the same thing
[09:57] <dilinger> at least in 2.6.11
[09:59] <fabbione> zul: pong
[10:01] <zul> fabbione: some of the external drivers have a ./Configure or ./configure and sometimes just a make how do you suppose we deal with them?
[10:01] <fabbione> zul: we will see it tomorrow really .. i am tired :)
[10:01] <zul> ok i was just doing some work on it though
[10:14] <jbailey> Wow
[10:14] <jbailey> Uncompressed initramfs = 7megs right now
[10:15] <jbailey> gzip'd: 3.2 megs
[10:15] <jbailey> lzma'd: 1.57 megs
[10:16] <jbailey> Oh great kernel gods, please provide. kthxbye
[10:18] <Mithrandir> who cares about the size of the initramfs?
[10:18] <jbailey> Mithrandir: Some ppc arch's can't cope with a  >4mb initramfs, and anyone downloading it off of a tftp server.
[10:18] <Mithrandir> it's the difference between 0.2 and 0.3 seconds or so of boot time.
[10:18] <Mithrandir> ah, ok.
[10:19] <Mithrandir> ppc is dead, though. ;-)
[10:19] <jbailey> In fact the boottime would be longer locally.  It took it a good little while to compress it.
[10:19] <Mithrandir> I guess I'm spoiled with amd64.
[10:19] <jbailey> "It's not dead yet!"
[10:19] <jbailey> Mithrandir: This is a dual ppc64 machine. =)
[10:20] <Mithrandir> I'm looking forward to playing with a dual 275 AMD64 machine.
[10:20] <Mithrandir> that means four cores.
[10:20] <Mithrandir> hm, we probably want Ubuntu to support that well, so I should be able to play with that on work time.
[10:20] <TMM> hi all!
[10:20] <TMM> I'm having a little problem, I build my own kernel with kernel-package from the sources apt-getted (which claim to have ubuntu patches) I copy the config from /boot, build the kernel with make-kpkg --initrd buildpackage (no changes) and my vesafb won't work with that kernel... shouldn't it be identical to the ones provided with ubuntu?
[10:21] <TMM> I was trying to add bootsplash support, and devise some initrd magic to load and set up the correct framebuffer driver (perhaps for inculsion in breezy, haven't checked with anyone or anything, just my fancy) but I'm getting stuck at even building a functioning kernel :)
[10:22] <jbailey> That reminds me that I should poke sladen about bootsplash stuff.
[10:22] <TMM> I'm currently not an ubuntu dev, just aspiring to be one with this little thing :)
[10:22] <jbailey> TMM: I doin't know how far the bootsplash stuff has gotten.
[10:23] <TMM> but, for now I can't even figure out how to build a 'standard' ubuntu kernel from source apparently
[10:23] <jbailey> TMM: But there's a group of people who are quite passionate about it that are probably worth checking in with.
[10:23] <jbailey> Hmm, dunno.  I don't build kernels. =)  
[10:23] <TMM> in my experience, even the friendly bootup procedure of ubuntu (not too verbose) scares people 
[10:24] <TMM> and I've got ubuntu (previously debian sid) running on quite a few luser desktops :)
[10:27] <TMM> somehow people consider it to look old fashoned
[10:29] <TMM> so, noone that can shed some light on my situation then? :)
[10:30] <TMM> am I missing some initrd stuff from ubuntu itself? I can't check because for some reason, my harddisk freezes as soon as I try to loopmount an initrd
[10:30] <TMM> yeah, weird, I know
[10:30] <jbailey> Is this on the stock kernel or a custom one?
[10:30] <jbailey> I don't think you should need to make a custom kernel at all for bootsplash stuff.
[10:31] <TMM> I'm pretty sure the bootsplash stuff isn't included in the stock ubuntu kernels
[10:31] <TMM> and, even then, why can't I build a working kernel image from the ubuntu sources? :)
[10:36] <jbailey> IIRC, the planned bootsplash stuff is entirely in userspace.
[10:36] <TMM> hmm
[10:36] <TMM> well, then this won't be much help
[10:36] <TMM> even then... I'm stuck with this problem :)
[10:41] <jbailey> kernels are possibly not the best place to start hacking. =)
[10:41] <jbailey> Are you willing to consider doing other hacking?
[10:42] <TMM> ofcourse
[10:42] <jbailey> Have you heard of the MOTUs?
[10:42] <TMM> but, this problem kind of has me a bit baffled
[10:42] <TMM> no, what is a MOTU? sounds like a swearword
[10:42] <jbailey> They're the folks who help keep the "universe" section in good shape.  
[10:42] <jbailey> "Masters Of The Universe"
[10:43] <TMM> ah
[10:43] <TMM> is there bootsplash stuff in there?
[10:43] <jbailey> There's alot of tasks to do there, and the heads of the MOTU are really cool people who can help you find something fun that's the right amount of challenge.
[10:43] <jbailey> sladen is the one who's generally doing the bootsplash stuff, and yes, he's in there.
[10:44] <jbailey> #ubuntu-motu is the name of the channel.
[10:44] <TMM> ok!
[10:50] <dilinger> jbailey: sorry for taking so long to respond, i totally forgot about the email
[10:50] <dilinger> jbailey: anyways, check your inbox
[10:53] <jbailey> dilinger: It's okay.  I havne't been the speediest either. =)