Natty Final Upload

Leann Ogasawara leann.ogasawara at canonical.com
Mon Apr 18 00:54:28 UTC 2011


On Sun, 2011-04-17 at 20:00 +0100, Andy Whitcroft wrote:
> During fridays release meeting report, I raised the possibility of a
> final kernel upload on Monday.  This primarily based on a pending KVM fix,
> without which installation under KVM cannot be localised.
> 
> I have reviewed all of the pending fixes on master-next (which includes
> 2.6.38.3 stable update).  There are a number of possible information
> leaks, but none which seem especially serious and could not wait for the
> first SRU kernel.  There are a few which are known to fix issues being
> reported currently:
> 
>  1) UBUNTU: SAUCE: kvm: fix push of wrong eip when doing softint,
>  2) rt2x00: Fix radio off hang issue for PCIE interface, and
>  3) Bluetooth: Fix HCI_RESET command synchonization.
> 
> The first of these is the KVM fix mentioned above, however having
> reviewed the KVM fix it seems that this only affects the host side and
> thus should only exhibit when the host is upgraded, this allows us to
> SRU this fix safely.
> 
> The second of these is a shutdown crash which while unfortunate should
> allow us to update the kernel as wireless functions normally until
> shutdown.  This should be SRUable.
> 
> The final one does does potentially affect a large number of bluetooth
> devices including keyboards and mice.  However it is not being widely
> reported and it is rare for the bluetooth keyboard to be your only
> input device.  It seems likely we can handle this one via SRU also.
> 
> Based on this unless there is any futher specific fixes which really are
> kitten killers, it does not seem necessary to upload again before release.
> Though we likely should prepare for an early SRU upload.
> 
> Comments?

Hi Andy,

I'd agree with your above analysis.  I say we just release note the
above issues and prep for an early SRU upload.

Leann





More information about the kernel-team mailing list