Hi Greg,

On 14 February 2017 at 04:01, Greg KH <gregkh@google.com> wrote:
On Mon, Feb 13, 2017 at 03:42:42PM -0600, Tom Gall wrote:
> Hi All,
>
> As a sort of 'prototype`' email for reporting back on one of the
> potential sources for 4.9 kernel fixes, I went trolling into the
> gentoo repo. (gentoo-sources to be exact, I didn't look at hardened
> tho I will)
>
> For 4.9 it's a pretty short list of potential fixes. Being more
> verbose about this just because well first post as well get started,
> here's what I've whittled the list down to and recommendations for the
> two patches.
>
> Patch:  1510_fs-enable-link-security-restrictions-by-default.patch
> From:   http://sources.debian.net/src/linux/3.16.7-ckt4-3/debian/patches/debian/fs-enable-link-security-restrictions-by-default.patch/
> Desc:   Enable link security restrictions by default.
> Recommendation : Interesting but I have hard accepting this as a 'fix'
> as compared to a feature change.
>
> Patch:  2900_dev-root-proc-mount-fix.patch
> From:   https://bugs.gentoo.org/show_bug.cgi?id=438380
> Desc:   Ensure that /dev/root doesn't appear in /proc/mounts when
> bootint without an initramfs.
> Discussion: https://patchwork.kernel.org/patch/2076031/  (unresolved)
> Recommendation: no action

Wait, you do know the rules for the stable kernel trees, right?  Please
go read Documentation/stable_kernel_rules.txt.

In other words, I can't do anything with patches that are not already in
Linus's tree, that's just not how the development process works.  The
things you list above are all crazy things that are not accepted into
mainline, for good reasons.

Also, I wouldn't worry about Gentoo, they usually don't have many, if
any, bugfixes in their kernels, as I think you found out already :)

Now if someone could dig in the ubuntu or fedora or openSUSE kernels,
that would be useful...

So Arnd and I discussed this yesterday - he is going to dig at openSUSE, while I'll be hunting fedora to start with. Based on how many such changes, we will add a couple more distros (raspbian, lede-project seem interesting...).
Also, one idea I got was to have engineers working with our members CC this list on kernel backports that they push to their kernels - would you think that'd make any sense? 
  
thanks,

greg k-h
_______________________________________________
Lts-dev mailing list
Lts-dev@lists.linaro.org
https://lists.linaro.org/mailman/listinfo/lts-dev

Best,
~Sumit.