Re: [PATCH 0/2] ima: policy search speedup

[Mimi Zohar <zohar@linux.vnet.ibm.com>]

On Tue, 2012-12-11 at 08:59 -0800, Linus Torvalds wrote:
> On Tue, Dec 11, 2012 at 6:08 AM, Mimi Zohar <zohar@linux.vnet.ibm.com> wrote:
> > On Tue, 2012-12-11 at 14:51 +0200, Kasatkin, Dmitry wrote:
> >> >>
> >> >> Two months ago I was asking about it on mailing lists.
> >> >> Suggestion was not to use s_flags, but e.g. s_feature_flags.
> 
> Quite frankly, this seems stupid.
> 
> Without really knowing the problem space, the sane thing to do would
> seem to be inode->i_flags. At which point it's
> 
>  (a) faster to test (no need to dereference inode->i_sb)
> 
>  (b) matches what the integrity layer does with S_IMA (well, there the
> logic is reversed: S_IMA means that it has a integrity structure
> associated with it)
> 
>  (c) allows you to mark individual inodes as "no checking".

The appraisal policy is based on the object metadata, such as the uid,
so the result is static and can be cached.  The measurement policy, on
the other hand, is normally based on the subject (eg. who is
reading/executing) the file.  Knowledge of whether the file has been
measured is cached in the iint, but unlike the appraisal policy, not
whether it needs to be measured.  Having the flag on a per inode basis,
doesn't really help.

thanks,

Mimi