[01:04] <crimsun> TheMuso: in jaunty's new PM world, we should investigate handling legacy (non-PM-enabled) sound modules for suspend (& resume) by unloading+alsactl store (& loading+alsactl restore) as per Takashi's recent e-mail on alsa-devel.  (Yes, the drivers should be updated, really...)
[01:05] <crimsun> there's some elbow grease involved, but I'm happy to look into it this cycle
[01:08] <TheMuso> crimsun: Ok, I'm happy to help where necessary.,
[01:08] <TheMuso> crimsun: I've also been attempting to play with a managable source package for alsa modules in a separate package. Haven't really come up with anything yet, short of using the script in alsa-driver to copy the files into alsa-kernel from a git tree.
[01:10] <TheMuso> crimsun: In that case, I'll subscribe to any pm related specs.
[02:01] <zoopster> If a kernel bug affecting ubuntu has been fixed upstream how can I get my hands on a compiled kernel with the patch to test?
[02:24] <TheMuso> zoopster: It depends on what version of the kernel the bug was fixed in. If its part of 2.6.27, then likely enough it may be included in a future kernel update for intrepid.
[02:27] <zoopster> TheMuso: thanks. the patches were put upstream for 2.6.27 and launchpad shows fix released from upstream...was unsure how that trickled down and how long that took. I didn't want to recompile a kernel with the patches if that was already done somewhere else
[11:45] <tseliot> amitk: I forgot to send a message (which reports a problem with the headers of kernel 2.6.28) with my @ubuntu.com alias and therefore it's waiting for moderation
[12:00] <amitk> tseliot: I don't have moderator access of the list. You might have to wait for BenC or rtg.
[12:01] <tseliot> amitk: ah, ok, thanks
[12:14] <snowblink> hi - can I ask why dapper's latest kernel package appears to be 2.6.15.54, but contains 2.6.15.53?
[13:26] <smb_tp> snowblink, where did you find 2.6.15.54?
[13:30] <snowblink> smb_tp, http://archive.ubuntu.com/ubuntu/dists/dapper-updates/main/binary-i386/Packages.gz
[13:38] <smb_tp> snowblink, This is the version number of the linux-meta package. The upload counter is just incidentally one above the counter of the kernel image itself
[13:40] <snowblink> smb_tp, ah ok. Thanks - it was a little confusing.
[13:41] <smb_tp> Yes, normally this number is much smaller than that of the kernel itself, but for some odd reason it jumped to this area some time ago. Now both are quite close which is a bit confusing
[15:09] <tseliot> rtg, amitk: is what I reported in the mailing list a known problem?
[15:10] <rtg> tseliot: I'm just looking at it. I think its likely a packaging problem.
[15:11] <tseliot> rtg: ok, that would be good news
[15:12] <_ruben> from what i've seen the asm/ dir got butchered in .28 (or .27 already perhaps) .. seen its implications in various out of tree modules
[15:12] <_ruben> which isnt an ubuntu specific thing, but upstream change
[15:13] <tseliot> _ruben: if this is true then other files should not include headers from asm
[15:14] <_ruben> well, its not completely gone, but lots of stuff has been (re)moved (to other places) .. i dont know any details, just an observation :)
[15:14] <tseliot> ah, ok
[15:20] <rtg> tseliot: I think I have to agree with _ruben. it looks like stuff has really moved around.
[15:20] <tseliot> rtg: shall I file a bug report then?
[15:21] <rtg> tseliot: its not really a bug per se, its more of a third party module problem. don't you think?
[15:21] <rtg> I guess you could file a bug against nVidia
[15:22] <tseliot> rtg: but the compilation test only includes this:
[15:22] <tseliot> #include <linux/version.h> #include <linux/utsname.h>
[15:22] <tseliot> and this fails
[15:22] <tseliot> I don't think it's nvidia specific
[15:23] <tseliot> rtg: example: /lib/modules/2.6.28-1-ub-generic/build/include/linux/sched.h:48:36: error: asm/param.h: No such file or directory
[15:23] <rtg> tseliot: hmm, that does seem straightforward. I wonder if BenC knows 'cause I haven't followed the .28 development cycle very closely.
[15:24] <tseliot> ok
[15:26] <BenC> tseliot: are you sure you are using standard build procedures?
[15:27] <BenC> tseliot: are there broken symlinks in there?
[15:30] <tseliot> BenC: I'm using Nvidia's makefile as usual. Nvidia's conftest.sh tries to build the test file with cc -D__KERNEL__ -DKBUILD_BASENAME="#conftest12386" -DKBUILD_MODNAME="#conftest12386" -nostdinc -isystem /usr/lib/gcc/i486-linux-gnu/4.3.2/include -I/lib/modules/2.6.28-1-ub-generic/build/include/asm/mach-default  -I/lib/modules/2.6.28-1-ub-generic/build/include -o conftest12386 conftest12386.c
[15:30] <BenC> tseliot: looks like an nvidia bug
[15:31] <BenC> tseliot: it's missing -I/.../arch/x86/include
[15:31] <BenC> tseliot: note, fglrx built fine with latest header packages...I didn't test nvidia
[15:32] <tseliot> BenC: ok, let me try again
[15:46] <rtg> BenC: apw and I were wondering why the build number isn't exposed in /proc/version ?
[15:47] <rtg> if it were, then we could drop /proc/version_signature
[15:49] <apw> i presume its not in the main version number to maintain the location of the modules through a complete abi cycle
[15:55] <tseliot> BenC: I get this error: http://pastebin.com/m7302cc4b
[16:15] <tseliot> BenC: nevermind, I had to add .../arch/x86/include/asm/mach-default too
[16:15] <tseliot> I'll report the problem to NVIDIA
[16:15] <tseliot> thanks for your time
[16:35] <lamont> how do I get a cli kvm host to keep running when the window is iconified or on a diff workspace???
[16:37] <rtg> soren: ^^^
[16:43] <cjb> Hi, something between intrepid 2.6.27-2 (which works fine) and 2.6.27-7 gives me two-second lag on my ssh connections on an otherwise fast (500KB/sec) cable line.  Anyone know anything about that?
[16:43] <cjb> HTTP seems unaffected.  It's quite odd.
[16:44] <rtg> cjb: 'GSSAPIAuthentication no' ?
[16:45] <cjb>  /etc/ssh/ssh_config says GSSAPIAuthentication yes
[16:45] <rtg> cjb: try setting it in ~/.ssh/config
[16:45] <cjb> So I guess not.  Does that have some recently-regressed kernel interaction, then?
[16:46] <rtg> cjb: not to my knowledge
[16:46] <cjb> oh, ok.
[17:36] <rtg> apw: I noticed actual activity in Linus' repo this morning, so he must be back from scuba diving.
[17:37] <apw> rtg wondered why he was so quiet, having a holiday, outrageous
[17:54] <lamont> damn fiordland
[18:26] <laga> lamont: he was in maui apparently