Ubuntu Kernel Team Meeting Minutes - 07/06/2010

Brad Figg brad.figg at canonical.com
Tue Jul 6 18:22:06 UTC 2010


= Meeting Minutes =
[[http://irclogs.ubuntu.com/2010/07/06/%23ubuntu-meeting.txt|IRC Log of the meeting.]]
[[http://voices.canonical.com/kernelteam|Meeting minutes.]]

== Agenda ==
[[https://wiki.ubuntu.com/KernelTeam/Meeting#Tues, 06 Jul, 2010|20100706 Meeting Agenda]]


=== [TOPIC] Open action: manjo to send a firewire patch with test info ===
tgardner uploaded the latest package of module-init-tools which contains the switch. pgraner and I have tested the new stack, and I have also tested this in Lucid and found to perform better than the old stack. pgraner did tests with dvgrab, kino, and 
files transfer across HDDs, and all checked out ok. Thanks to pgraner for all the testing. Jfo will issue a call for testing soon. Since this item is all done we can have this item removed from the meeting agenda.

=== [TOPIC] Open action: JFo to put out a CFT on new firewire stack ===
Will send out today.

=== ARM Status  ===
  * Marvel (mvl-dove)
    * Nothing new this week
  * Freescale (fsl-imx51)
    * Nothing new this week
  * Texas Instruments (ti-omap)
    * CONFIG   : Activate block IO actions, tracing required by system-tap
    * CONFIG   : Increase default VRAM allocation size to 32MB
    * PATCH    : Bug fixing for segfault
    * PATCH    : Video display fix on Beagleboard
    * ON GOING : Submitted patches for building Syslink as a module need to be modified
    * ON GOING : Auto-load Syslink modules on OMAP4 architectures is nearly complete
    * ON GOING : Still trying to figure out why disabling camera power domain on OMAP3 prevents daisy chain read in Ubuntu system
    * ON GOING : May fixed DSS2 oops. QA has an image they need to look at. mpoirier has been testing a lot in his environment without re-occurrence
    * ON GOING : rtg will be looking at IGEPv2 board - mpoirier shipped the board yesterday

=== Release Metrics ===
==== Bugs ====
||<-3 tablestyle="width: 100%;")>                                                                                                                                ||
||<:>                   ||<:> '''Alpha 3 Milestoned Bugs (51 across all packages (up 34))''' ||<:> '''Release Targeted Bugs (107 across all packages (up 21))''' ||
||linux                 ||<:> 1                                                              ||<:> 1                                                             ||
||linux-ti-omap         ||<:> 0 (-2)                                                         ||<:> 2 (-2)                                                        ||
||linux-meta-ti-omap    ||<:> 1                                                              ||<:> 2                                                             ||

I will add mvl and freescale back for next week's metrics
==== Milestoned Features ====
  * 14 blueprints (including HWE Blueprints)

====  Bugs with Patches Attached:115 (up 5)  ====
  * [[https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bugs?field.has_patch=on|Launchpad report]]
  * [[http://qa.ubuntu.com/reports/ogasawara/csv-stats/bugs-with-patches/linux/|Breakdown by status]]

=== Blueprint: kernel-maverick-apparmor  ===
Fixed Bug #581525 both kernel and user space and am about to push out test kernels for reporters to verify. Have a patch for Bug #599450 that needs to be tested. Bug #602261 has been reported as a potential AA issue, looking into it.  We have some compiler 
patches that reduce memory usage that will get synced with alpha3. Looking into duplicating and the bug and seeing if it is AA.
  * [[https://launchpad.net/bugs/581525|Launchpad bug 581525 in apparmor (Ubuntu) "Lucid: system becomes unstable randomly, seems related with apparmor" [Undecided,New] ]]
  * [[https://launchpad.net/bugs/599450|Launchpad bug 599450 in linux (Ubuntu Maverick) "[apparmor] getattr handled incorrectly in 2.6.35-6.7" [High,New] ]]
  * [[https://launchpad.net/bugs/602261|Launchpad bug 602261 in linux (Ubuntu) "Thrashing and OOM during upgrade from 10.04 to Maverick" [Undecided,Incomplete] ]]
(apw) Notes that that bug occurs when running lucid (pre-update kernel). (jjohansen) Agrees, though it is likely userspace component since the compiler consumes 80-100MB.

=== Blueprint: kernel-maverick-firewire-stack  ===
See open action item above. This blueprint topic will be removed from future agendas.

=== Blueprint: kernel-maverick-misc  ===
Some work on cleaning up lintian errors is ongoing.  Nothing else to report.

=== Blueprint: kernel-maverick-new-kernel-on-lts  ===
Nothing new this week.

=== Blueprint: kernel-maverick-pv-ops-ec2-kernel  ===
Have update Xen patchset, that I will start testing today. Amazon have opened their pv-grub based loader for general use so I will be revisiting the pv-ops based kernel.

=== Blueprint: kernel-maverick-ubuntu-delta-review  ===
Since we won't be using union-mounts in Maverick, we're planning an aufs update for Alpha3.

=== Blueprint: kernel-maverick-union-mounts  ===
Based on the lack of feedback we will assume staying with aufs2 for Maverick.  Preparations to update aufs2 are in progress. This blueprint topic will be removed from future agendas.

=== Blueprint: kernel-maverick-config-review  ===
We have 2 misc bugs to close for Alpha3.  Both are already assigned to a kernel dev and In Progress.

=== Blueprint: kernel-maverick-bug-handling  ===
Nothing new this week.

=== Blueprint: kernel-maverick-upstart  ===
The ureadahead patches remain untested.  Will bring this up with foundations.

=== Blueprint: kernel-maverick-bios-test-automation  ===
Changes to Firmware Test Suite this week:
  * Additional Tests:
    * thermal_trip: Check Thermal Zone trip mechanism.
    * cpufreq:      Check P-states across all CPUs.
    * bios_info:    Gather BIOS version information.
    * microcode:    Check if latest microcode is loaded.
    * maxfreq:      Check CPU max freq against BIOS setting.
  * Other changes:
    * hibernate:    Fix S4 detection bug, check for hibernation phases.
    * lid:          Detect GPE activity on lid events.
    * WMI:          Only dump driver writing advice text out once.
    * ACPI tables:  Extract ACPI tables without the need for acpidump.
  * * Misc:
    * Wrote Python version of acpidump for inclusion into apport
    * Maverick PPA builds for new code.

=== Status: Maverick  ===
linux-2.6.35-6.9 was uploaded immediately upon Alpha2's release.  I expect the upstream 2.6.35-rc4 kernel to land relatively soon.  We'll of course rebase our Maverick kernel at that time.
Alpha 3 is Thurs Aug 5th which is about a month away.  We are towering over the Alpha 3 burn down chart's trend line, so please review the list of work items for Alpha 3 and start closing out tasks:
  * http://people.canonical.com/~pitti/workitems/maverick/canonical-kernel-team-maverick-alpha-3.html
  * https://wiki.ubuntu.com/KernelTeam/ReleaseStatus/Maverick#Milestone maverick-alpha-3

=== Security & bugfix kernels - Karmic/Jaunty/Intrepid/Hardy/Others  ===
  * Dapper:      2.6.15-55.84  (security)
  * Hardy:       2.6.24-28.70  (security)
                 2.6.24-28.71  (proposed)[18]  0 / 1 verifications done
  * Jaunty:      2.6.28-19.61  (security)
  * Karmic:      2.6.31-22.60  (security)
                 2.6.31-22.61  (waiting for approval)
     - mvl-dove  2.6.31-214.28 (security)
                 2.6.31-214.29 (waiting for approval)
     - fsl-imx51 2.6.31-112.28 (security)
                 2.6.31-112.29 (waiting for approval)
     - ec2       2.6.31-307.15 (security)
                 2.6.31-307.16 (waiting for approval)
  * Lucid:       2.6.32-23.37  (updates)
                 2.6.32-24.17  (proposed)[ 1]  0/ 4 verifications done
     - LBM       2.6.32-24.38  (proposed)[ 0]  0/ 1 verifications done
     - mvl-dove  2.6.32-206.19 (updates)
     - fsl-imx51 2.6.31-608.15 (updates)
     - ti-omap   2.6.33-502.8  (updates)
     - qcm-msm   2.6.31-802.5  (updates)
     - ec2       2.6.32-307.12 (updates)

Lucid has been reuploaded with a proposed kernel that addresses a lot of ext4 issues. If all goes to plan this gets moved to updates soon enough to have a security update before Jul-28 which is the first point release.

Karmic is still not accepted because of the abstracted.debian backport.  As there is a security update pending we need to eihter get someone take the risk and accept it into proposed, or we need to revert everything and get the minimal changes uploaded. 
Though the security update would be based on the version currently in updates, I think it would require a very careful merging back and together with the branches needing special care cause the whole validation of the builds being void.

||smb       ||tgardner, pgraner Would you think that Steve would take that risk if you asked him?||
||tgardner  ||possibly||
||tgardner  ||especially after we show that the binary packages are identical||
||smb       ||tgardner, Would you take the action of talking to him, then?||
||tgardner  ||smb, sure||
||smb       ||ok thanks||

=== Incoming Bugs and Regressions ===
==== Incoming Bugs: ====
||<-2 tablestyle="width:  35%;")>                   ||
||<:> '''Version'''      ||<:> '''Count'''          ||
||Maverick               ||<:>50 (+22)              ||
||Lucid                  ||<:>1028 (+48)            ||

==== Current regression stats: ====
||<-5 tablestyle="width: 100%;")>                                                                                                   ||
||<:> '''Version'''     ||<:> '''Potential'''      ||<:> '''Update'''         ||<:> '''Release'''        ||<:> '''Proposed'''       ||
|| maverick             ||<:> 27 (+15)             ||<:>                      ||<:>                      ||<:>                      ||
-- 
Brad Figg brad.figg at canonical.com http://www.canonical.com




More information about the kernel-team mailing list