Request pull

[Wang, Yong Y]

> -----Original Message-----
> From: Tim Gardner [mailto:tcanonical at tpi.com]
> Sent: 2008年5月13日 22:06
> To: Wang, Yong Y
> Cc: kernel-team at lists.ubuntu.com
> Subject: Re: Request pull
>
> Wang, Yong Y wrote:
> > Hi kernel team,
> >
> > The following changes since commit
> > 20edcf831060d7e81599001858e66730f47f1d7b:
> >   Tim Gardner (1):
> >         [ALSA] hda-codec - Add docking-station mic input
> for Thinkpad
> > X61
> >
> > are available in the git repository at:
> >
> >   git://kernel.ubuntu.com/ywang/ubuntu-hardy-lum.git master
> >
> > Yong Wang (1):
> >       UBUNTU: Update e1000/e1000e to v8.0.1/v0.2.9.5
> >
> > BTW, pls disable e1000/e1000e in ubuntu-hardy.git if this
> pull request
> > can be accepted.
> >
> > Thanks
> > -Yong
> >
>
> Yong - How does this driver relate to the upstream kernel version?
> Frankly, there are very few bug reports against the Hardy e1000/e1000e
> driver so I am reluctant to supersede the kernel version.
>

These are the latest e1000/e1000e drivers released by Intel and have been well tested on a wide range of kernels (2.6.18 to 2.6.26, IIRC). I believe they are going to make into .26 mainline kernel.

> Are there platforms that require this update? If so, I'm happy to add
> the driver to LUM for just those PCI identifiers that do not conflict
> with the kernel version. Alternatively, will the kernel
> driver work with the addition of new PCI identifiers?
>

Yes, Intel's next generation mobile platform, codenamed Montevina, requires this new driver to enable NIC. That should count as SRU justification according to https://wiki.ubuntu.com/StableReleaseUpdates.

> FYI - If there is driver and bus/device identifier overlap between the
> kernel and LUM, the LUM driver gets preference unless it is
> specifically
> black listed.
>

Is this behavior automatic? I used to see Kyle comment out MODULE_DEVICE_TABLE in corresponding file in kernel tree under such circumstance.

Thanks
-Yong