[cryptography] kernel.org hack and kernel integrity

Douglas Huff dhuff at jrbobdobbs.org
Sat Sep 3 02:18:55 EDT 2011

The short answer is "tweak[ing] dates on commits" would change the commit id
which would break any other existing trees dependant on that commit and
someone would notice. Really that simple.

Douglas Huff
On Sep 2, 2011 9:19 PM, "Jeffrey Walton" <noloader at gmail.com> wrote:
> Am I the only guy who finds the kernel integrity assurances suspect [1]:
> However, it's also useful to note that the potential damage of cracking
> kernel.org is far less than typical software repositories. That's because
> kernel development takes place using the git distributed revision control
> system, designed by Linus Torvalds. For each of the nearly 40,000 files
> in the Linux kernel, a cryptographically secure SHA-1 hash is calculated
> to uniquely define the exact contents of that file.
> I did see the claims that git had security related design goals
> (wikipedia). Unfortunately, the wikipedia reference points to a
> Torvalds talk at Google where he claims "security is distributed. and
> I trust 5, 10, 0r 15 developers [sic]" [2] (among his other ramblings
> and bashings). So its not clear to me how Torvalds trust a few people,
> therefore integrity is assured. And naively, I would also expect that
> the ability to do things like "tweak dates on commits" would help hide
> malicious behavior [3].
> Could anyone explain git's security assurances to a non-git layman?
> [1] http://kernel.org/
> [2] http://www.youtube.com/watch?v=4XpnKHJAok8, 27:43
> [3]
> _______________________________________________
> cryptography mailing list
> cryptography at randombit.net
> http://lists.randombit.net/mailman/listinfo/cryptography
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.randombit.net/pipermail/cryptography/attachments/20110903/880ea00f/attachment.html>

More information about the cryptography mailing list