Ubuntu Wiki

Logs

TZ UTC+1

05:02   BenC    ok, hello everyone
05:02   BenC    This is the Ubuntu Kernel Team meeting, our agenda is at https://wiki.ubuntu.com/KernelTeam/Meeting
05:03   BenC    Unfortunately, Kyle and Tim are at UbuntuLive and wont be attending, which shortens our agenda quite a bit
05:04   BenC    A few highlights of recent things we've been working on include UME kernel (amitk), and liveCD memory issues (pkl)
05:05   BenC    amitk: would you care to elaborate on some of the work you're doing with UME, and plans for power management improvements?
05:06   BenC    ah, amit has lost connection :)
05:07   BenC    While he's regaining that connection...any questions from community folks?
05:07   zul     not me
05:07   zul     I didnt prepare anything at least
05:07   BenC    this is going to be a short meeting, so we'll have time for this sort of thing :)
05:08   zul     oh yeah ill try to get to vserver this week what about openvz?
05:08   ScottK  Would anyone be willing to work on kernel bug triage while we wait?
05:09   BenC    zul: openvz looks like it may miss the gutsy release, but we are discussing possibilities with them
05:09   zul     ok
05:09   BenC    they are still based on 2.6.20 kernel, and the reject list is huge against 2.6.22
05:10   zul     and im not touching it ;)
05:10   amitk_  I dropped out...
05:10   BenC    ScottK: you're more than welcome to do that at any time :)
05:11   BenC    amitk: saw that on the other server...we were just discussing some open topics, but I had wondered if you wanted to give an update on UME and/or power savings work in gutsy
05:11   ScottK  BenC: Unfortunately for me, when it comes to the kernel, I'm more of a triagee than a triager.
05:12   BenC    ScottK: heh, a good time for that is bug days, in which we'll have a kernel team person actively helping and working on kernel bugs
05:13   ScottK  Sure.  Just thought maybe it would be a target of opportunity while waiting.  No problem.
05:13   amitk_  nothing big really - UME kernels seem to work on the menlow platform
05:13   amitk_  https://blueprints.launchpad.net/ubuntu/+spec/power-management-in-ubuntu has been modified to be more focused for gutsy
05:14   amitk_  and I am now working on splitting up the test cases for laptop-testing group in order to revive it
05:14   BenC    excellent, the laptop testing sounds really promising
05:15   BenC    amitk: have you contacted heno about the test infrastructure yet?
05:15   amitk_  no I haven't. I will do that right away
05:16   BenC    hopefully they've been working on the backend already, and it will just be a matter of plugging in the explanations and reporting procedures that you come up with
05:16   BenC    pkl_: you've been tasked with livecd memory usage problems, and with virtualization (mainly xen)...how's that coming?
05:17   amitk_  I would also like to put out a call for volunteers to do some powertop hacking (www.linuxpowertop.org)
05:17   pkl_    BenC: currently suspended, looking at the critical #126964 bug.
05:18   pkl_    That's the liveCD corruption and hang bug.
05:18   BenC    ah, right
05:18   pkl_    The hang I traced last week to a race I put into Squashfs in April.  Fixed.
05:18   BenC    been able to point the blame at anything specific yet?
05:19   BenC    good, nice work
05:19   pkl_    The corruption bug is much rarer, but very nasty.
05:20   pkl_    Unfortunately the available debug info proved impossible to track this down.
05:20   BenC    pkl_: would this benefit from crashdump?
05:20   BenC    I could probably write up something, and create an ISO which can handle crashdumps capturing
05:21   pkl_    I have put in a lot more non-invasive (i.e. no spewing rubbish to dmesg) logging via a psedo file (debugfs), and got some good debug info yesterday,
05:22   pkl_    I believe I know the execeptional set of circumstances which cause the bug.  It looks like an undiscovered race condition in the Squashfs fragment cache code.
=== neversfelde|mobi [n=neversfe@82.113.106.1] has joined #ubuntu-meeting
05:22   BenC    ok, if you need more extensive testing, would be good to talk heno about putting out a call for that
05:22   BenC    otherwise, keep up the good work you're already doing
05:22   pkl_    Debugging is currently ongoing.
05:23   pkl_    Yeah, I've been up to 6am most nights..
05:23   pkl_    As it's Squashfs I have a vested interest in fixing it, and a certain responsibility :)
05:24   BenC    pkl_: sleep becomes a luxury at times :)
05:25   BenC    Well, unless there's some other topics people want discussed, this meeting is at an end
05:25   BenC    I'll give a short pause from the chair...
05:26   BenC    And, that's it folks :)
05:27   BenC    thanks everyone, have a good week and so you all in two weeks for the next meeting