[02:22] <cjwatson> BenC: ubuntu-modules didn't actually appear in the NEW queue, so if you were intending it to be there you have a bug
[02:23] <cjwatson> oh, unless it's from the linux-ubuntu-modules source, which I guess it would have to be
[02:23] <cjwatson> duh. /me -> bed
[03:36] <BenC> cjwatson: heh, thanks
[03:37] <crimsun> BenC: you're listed as the feisty contact; should I still send you patches for feisty-updates?
[03:38] <BenC> crimsun: feisty -> pkl && kernel-team@
[03:38] <BenC> crimsun: [FEISTY]  in the subject too
[03:39] <crimsun> BenC: ok, thanks. I read a couple emails from k-t@ and was perplexed, since the wiki still mentions you being the contact
[03:39] <BenC> crimsun: Ah, I need to update that, thanks
[06:05] <BenC>       [WIRELESS]  cfg80211: New wireless config infrastructure.
[06:05] <BenC> yay
[06:09] <BenC>     Driver core: udev triggered device-<>driver binding
[07:17] <fabbione> BenC: still awake?
[07:23] <dade`> morning here
[07:31] <fabbione> BenC: kernel headers are busted.
[08:18] <rpereira> Hi. Does someone knows how to active CRT on an laptop with 945gm?
[08:18] <rpereira>  When I use i810switch: I'm getting this error
[08:18] <rpereira> :
[08:18] <rpereira>  PCI id of i810 is not recognized
[08:21] <dade`> wait
[08:21] <dade`> this is not an help channel
[08:21] <dade`> and even if it would be, your question is not kernel related
[08:22] <dade`> hm, "even if it was" I guess
[08:22] <dade`> anyway, look here https://help.ubuntu.com/community/MacBook
[08:22] <dade`> there are some example xorg configuration to enable dual monitor
[08:23] <dade`> and join #ubuntu for support
[08:33] <rpereira> Thank you dade` and sorry.
[10:31] <abogani> BenC, ping
[12:20] <abogani> Who is the kernel-team list moderator? 
[12:23] <abogani> lamont, Are you kernel-team list moderator?
[03:53] <lamont> hrm.. hard to answer people when they disappear
[04:49] <defendguin> something happened between release and now and now the laptop lid close doesn't suspend the laptop.  I have checked my settings in power manager and they are still set to suspend on lid close.
[05:23] <defendguin> how can i pin point what the problem is so i can put some decent information in the bug report
[05:37] <abogani> Who is the kernel-team list moderator? I should send a patch in the mailing list but email never reach it...
[05:39] <Nafallo> abogani: 15:53 < lamont> hrm.. hard to answer people when they disappear
[05:42] <abogani> Nafallo: Ok, thanks.
[06:49] <toocrazypt> hi
[07:02] <cjwatson> BenC: kernel headers seem badly broken
[07:03] <cjwatson> BenC: e.g. http://librarian.launchpad.net/7446222/buildlog_ubuntu-gutsy-amd64.busybox_1%3A1.1.3-4ubuntu1_FAILEDTOBUILD.txt.gz, http://librarian.launchpad.net/7451197/buildlog_ubuntu-gutsy-sparc.busybox_1%3A1.1.3-4ubuntu1_FAILEDTOBUILD.txt.gz, http://librarian.launchpad.net/7453000/buildlog_ubuntu-gutsy-i386.netcfg_1.37ubuntu1_FAILEDTOBUILD.txt.gz
[07:03] <BenC> cjwatson: missing build-dep for 'tr' command (coreutils
[07:03] <BenC> )
[07:03] <BenC> cjwatson: preparing an upload now
[07:03] <cjwatson> you don't need to build-dep on coreutils - it's Essential: yes
[07:03] <BenC> then something was broken on the buildd's
[07:04] <cjwatson> is it visible in the build log?
[07:04] <BenC> FNAME=$(patsubst $(INSTALL_HDR_PATH)/$(_dst)/%,%,$@)                    \
[07:04] <BenC> STUBDEF=__ASM_STUB_`echo $$FNAME | tr a-z.- A-Z__`;                     \
[07:04] <BenC> no, hidden by quite kbuild system :/
[07:04] <cjwatson> what might FNAME end up being?
[07:05] <cjwatson> (example)
[07:05] <BenC> stat.h
[07:05] <BenC> cjwatson: note, the linux-libc-dev built locally for me ended up just fine
[07:05] <BenC> I tested it before upload, and checked the ones I have built here
[07:05] <cjwatson> could it be locale-dependent?
[07:06] <BenC> I export LANG=C in debian/rules
[07:06] <cjwatson> LANG doesn't override, might be worth exporting LC_ALL=C instead
[07:06] <BenC> this is the same code in linux-source-2.6.20 too, so it's not anything new
[07:06] <cjwatson> just in case LC_COLLATE is set to something mad
[07:06] <cjwatson> ok
[07:06] <BenC> oh, right, we use LC_ALL=C
[07:07] <BenC> export DH_COMPAT=4
[07:07] <BenC> export LC_ALL=C
[07:07] <BenC> export SHELL=/bin/bash -e
[07:07] <cjwatson> in what file is the error visible?
[07:07] <BenC> all files in /usr/include/asm/
[07:08] <BenC> since they are stubs to asm-i386 and asm-x86_64
[07:08] <cjwatson> oh I see, just __ASM_STUB_ with no suffix
[07:08] <BenC> right
[07:08] <cjwatson> I think it's more likely to be buggy make syntax
[07:09] <BenC> why would it work on my build systems and not on the buildds?
[07:09] <BenC> the only difference is these are running feisty still
[07:09] <BenC> FNAME=$(patsubst $(INSTALL_HDR_PATH)/$(_dst)/%,%,$@)                    \
[07:09] <BenC> STUBDEF=__ASM_STUB_`echo $$FNAME | tr a-z.- A-Z__`;                     \
[07:09] <BenC> that's from 2.6.20...
[07:10] <BenC> exact same code
[07:10] <BenC> this is all in scripts/Makefile.headersinst btw
[07:10] <cjwatson> I don't know, just saying missing tr on the buildds is massively unlikely
[07:11] <BenC> what about a broken tr?
[07:11] <BenC> let me get gutsy coreutils
[07:11] <cjwatson>  coreutils | 5.97-5.2ubuntu3 |        feisty | source, amd64, i386, ia64, powerpc, sparc
[07:11] <cjwatson>  coreutils | 5.97-5.2ubuntu3 |         gutsy | source, amd64, i386, ia64, powerpc, sparc
[07:11] <cjwatson> identical
[07:11] <cjwatson> unless libc broke it
[07:11] <BenC> could be that too
[07:12] <BenC> let me get a gutsy chroot and try these commands
[07:12] <cjwatson> building a fix is going to be fun at this rate :-/
[07:12] <cjwatson> probably shouldn't have done a sync pass today with this broken
[07:14] <cjwatson> so ia64 and powerpc should still be fine
[07:17] <cjwatson> infinity: we'll need a manual bootstrap to get us out of this, I suspect
[07:19] <BenC> yeah, because kernel build does compile userspace stuff :/
[07:20] <cjwatson> well, unless the glibc build can be done with broken kernel headers
[07:21] <BenC> but still need to build new linux-libc-dev
[07:21] <cjwatson> oh, right, yeah
[07:30] <BenC> 10 more minutes till I get this updated gutsy chroot
[07:34] <cjwatson> I'm partway through a debootstrap now
[07:35] <cjwatson> # echo stat.h | tr a-z.- A-Z__
[07:35] <cjwatson> STAT_H
[07:35] <cjwatson> works fine here, though this is powerpc
[07:36] <BenC> doing mine on amd64...
[07:36] <cjwatson> version of make is the same in feisty and gutsy
[07:37] <cjwatson> likewise for both dash and bash
[07:37] <BenC> works for me too
[07:37] <BenC> let me run a headers_install and see what happens
[07:37] <cjwatson> is it possible that INSTALL_HDR_PATH is bust or something?
[07:38] <cjwatson> or something else that would cause FNAME to be empty
[07:40] <cjwatson> this does happen if that fragment of make is run in dash
[07:41] <BenC> works fine here locally
[07:41] <cjwatson> what's your /bin/sh?
[07:41] <BenC> bash
[07:41] <cjwatson> set it to dash as is now standard on Ubuntu, I bet it'll fail
[07:41] <cjwatson> the buildds only switched to that in gutsy
[07:42] <cjwatson> putting a semicolon after FNAME=... and before STUBDEF=... should fix it
[07:42] <cjwatson> yep, confirmed in a microtestcase
[07:42] <BenC> boom
[07:43] <BenC> SHELL=/bin/bash fixes it
[07:44] <cjwatson> I've mailed ubuntu-devel@ so fewer people will panic about build failures
[07:45] <cjwatson> I would recommend adding the semicolon too
[07:46] <cjwatson> from my reading of http://www.opengroup.org/onlinepubs/009695399/utilities/xcu_chap02.html#tag_02_09_01 it's actually a bash bug that it works
[07:46] <cjwatson> "Each variable assignment shall be expanded for tilde expansion, parameter expansion, command substitution, arithmetic expansion, and quote removal prior to assigning the value."
[07:46] <cjwatson> well, hmm, it doesn't say whether it goes expand expand assign assign or expand assign expand assign, so it's probably just unspecified and bash is entitled to its interpretation
[07:48] <cjwatson> I'll be out for most of the rest of the evening - I'd suggest mailing Adam instructions for recovery
[07:48] <BenC> ok
[08:06] <BenC> starting a build...initial packages look ok, but I have a big linux-2.6 sync to make sure is working
[10:08] <abogani> BenC: May i disturb you about kernel-team mailing list? I would write emails about -rt kernel and my spec (wiki.ubuntu.com/RealTime)... but no email reach the this mailing-list :-( 
[10:08] <BenC> kernel-team moderator will pick it up within a day or so
[10:11] <BenC> cjwatson: Uploading a package with just the SHELL= fix
[10:12] <BenC> current HEAD has some build issues, so just getting this out
[10:13] <abogani> BenC: It isn't first time... I attempted from january to write emails in kernel-team mailing-list (when i started my spec on realtime)...
[10:14] <BenC> abogani: are you sub'd to the list?
[10:14] <BenC> you should be sending emails from your subscribed address
[10:14] <abogani> BenC: Yes i read every email...
[10:14] <eean> I still have to cp ipw3945d-2.6.20-15-generic into existance
[10:14] <BenC> does it send you an email back saying you were moderated?
[10:15] <eean> otherwise my wireless card doesn't work
[10:15] <BenC> eean: maybe because you have to install linux-restricted-modules-2.6.20-15-generic
[10:15] <eean> maybe so!
[10:15] <BenC> install linux-generic to make sure you get it all, and stay updated
[10:15] <abogani> BenC: Yes but after nothing happen..
[10:16] <eean> BenC: yep, doing. 
[10:17] <eean> is there a way to go back and delete old kernels and their modules automatically?
[10:17] <BenC> automatically, no
[10:17] <eean> hmm
[10:18] <BenC> although, "sudo apt-get --purge autoremove" might do it
[10:18] <eean> I guess normally it doesn't build up so much
[10:18] <eean> like it did the weeks before feisty was released
[10:18] <Nafallo> BenC: I hope they are marked as manually installed.
[10:18] <BenC> yeah, normally you only see a problem when you follow the development cycle
[10:18] <eean> ah nifty
[10:18] <BenC> Nafallo: no, linux-image-generic pulls them in most times
[10:19] <Nafallo> BenC: yea, but the code for autoremove marks them manually installed IIRC.
[10:19] <Nafallo> BenC: if not, I would consider that a bug.
[10:19] <BenC> I wouldn't
[10:19] <Nafallo> nothing should EVER remove a kernel without me explicitly telling it to :-)
[10:19] <eean> autoremove is explicit
[10:19] <BenC> if you run "autoremove" you are telling it to :)
[10:20] <eean> doesn't happy by itself
[10:20] <eean> doesn't *happena
[10:20] <eean> blah ;)
[10:20] <Nafallo> hmm. some users might not check what packages will be removed :-P
[10:21] <Nafallo> I want that shiny cleanup tool I've seen specs about for a while now :-)
[10:21] <eean> and such users need old kernels hanging around?
[10:21] <Nafallo> I'm paranoid about kernels since I did an dist-upgrade that removed the running kernel and failed to install the new one ;-)
[10:21] <eean> ah well that would make sense
[10:22] <eean> ah suck! 
[10:23] <Nafallo> I had better words coming from me at the time :-P
[10:23] <eean> actually had a similar problem, I manually uninstalled old kernels and the new kernels initrd wasn't correctly installed
[10:23] <abogani> BenC: I don't want bother you with my stupid problem... Who can resolve my problem? _Only_ lamont?
[10:23] <BenC> abogani: and kyle, but he is getting ready to travel, so probably isn't around
[10:24] <BenC> abogani: best bet is to make sure your From matches the address you are sub'd to the list with
[10:26] <abogani> BenC: Yes it is..... :-(  In the meanwhile where i can send emails about -rt kernel?
[10:28] <BenC> abogani: depends on what your sending
[10:28] <abogani> BenC: Or You prefer this channel? :-)
[10:29] <BenC> what are you sending?
[10:29] <abogani>  BenC: Part cuted from my email :-) "Attached to this email the realtime flavour diff for the new Gutsy kernel build system. I removed all things non strictly related to realtime preemption patch (kvm/paravirt/hypercall)."
[10:30] <abogani> "I see, in gutsy's git tree, that BenC has already do all the works."
[10:30] <BenC> abogani: Ah, cool...I actually included ingo's stock 2.6.21-rt1 patch, but if you can email me the new patch directly, I can include it
[10:33] <abogani> BenC: Done.
[10:34] <BenC> abogani: thanks
[10:36] <abogani> BenC:  Where i can found git tree of the l-u-m and l-r-m ? I promised to UbuntuStudio guys all packages (headers, image, debug, all meta, l-u-m, l-r-m) for realtime kernel...
[10:36] <BenC> abogani: they are private, but you need not worry about that
[10:37] <BenC> lum and lrm are already done
[10:37] <BenC> linux-meta wont be uploaded for a few weeks
[10:39] <abogani> BenC: Thanks a lot! Kernel-team's people are incredible! :-)
[10:39] <BenC> np :)
[10:42] <abogani> BenC: Last thing... Who have kernel realtime's bugs assigned? 
[10:43] <abogani> BenC: Me, Kernel Team , UbuntuStudio Team ?!?
[10:58] <BenC> abogani: I was going to assign them to ubuntu-studio
[11:05] <abogani> BenC: I don't know if it is the right thing...
[11:06] <BenC> I would rather assign them to a team than a person
[11:06] <BenC> and preferably not the kernel team
[11:07] <BenC> ubuntu-studio has a vested interested in this patch
[11:08] <abogani> BenC: Ok i understand. Thanks for all!
[11:08] <BenC> thank you