Re: handling of Linux kernel vulnerabilities (was: CVE request - Linux kernel: VFAT slab-based buffer overflow)

[Greg KH <greg () kroah com>]

On Mon, Mar 04, 2013 at 06:57:23AM +0400, Solar Designer wrote:
> Greg,
>
> Note that I am not even asking you to reconsider.  I have little hope
> that you would, as you appeared to have a firm opinion on this.
> I merely mentioned this aspect, with no intent to prompt a discussion of
> it.  That said, I've commented inline, just to clear up your confusion.

Thanks for doing this.

> > You bring up a bunch of issues that
> > the distros need to consider, what can the Linux kernel security team do
> > differently?
>
> Post to oss-security on commit day.

You know why we will not do that, sorry.

> Optionally, also notify linux-distros a few days before the commit.

We don't usually have "days" before things are committed.  We find out
about a problem, we make up a fix, and it is committed.  Usually all
within 1-2 days.  Sometimes things take longer to fix, but usually it's
prettty fast.

> > > Overall, I think we should bite the bullet and accept sko's
> > > notifications to linux-distros, with a grace period of up to 7 days.
> > > Whenever a distro is ready to release an update, they should be able to
> > > insist on doing so within another 1 day, even if the initially planned
> > > grace period would expire later.  Would sko be OK with this?  Greg?
> >
> > Again, I don't think anyone that is part of security () kernel org minds
> > about having the issues publicized, after linux-distro has their time
> > to get things fixed and to their users.  If the linux-distro people care
> > about that, that does not seem to be a security () kernel org group issue,
> > right?
>
> Right, but since you previously refused to notify oss-security right
> away, I thought that you could possibly stipulate that you'd only keep
> notifying linux-distros if the linux-distros folks keep the issues from
> hitting oss-security for at least a certain amount of time, or at least
> until fixes are available (from at least one distro? from all?), or
> whatever.  If you're fine with letting linux-distros decide on this
> fully on their own, and you would not stop notifying linux-distros if
> you deem that they fully-disclose the issues publicly "too soon", that's
> great (and logical)!

As far as I am concerned, I trust linux-distros to manage this in a sane
and proper manner, and they can notify the world when they decide to do
so.  If that trust is somehow broken, we can revisit the issue in the
future.

thanks,

greg k-h