No, unfortunately I don't have that data yet. I need to set up a selinux-capable system in our testing pool, then I will generate numbers. thanks, -serge PS - the filename of the tarball of patches on sf.net is stacker- patches-feb22-05.tar.gz On Tue, 2005-02-22 at 10:25 -0500, Stephen Smalley wrote: > On Mon, 2005-02-21 at 12:48 -0600, Serge Hallyn wrote: > > www.sf.net/projects/lsm-stacker has a new update of the stacker > > patches. I've gone back to most things lockless for the sake of > > performance. See item # 4 under the updates for the RCU (-less) > > info, and the file stacker_rcu_test.txt (in the latest > > sourceforge release for some performance data) . See item # 2 for > > the rwlock (-less) info. Both require some extra knowledge from > > modules writers which is not yet reflected in > > Documentation/security-stacking.txt. > > Also, does your performance data include a comparison of a vanilla > 2.6.11-rc4-bk9 kernel (with SELinux+capabilities stacked in the existing > manner) vs. with your patches and SELinux+cap_stack stacked via stacker? > That is what we truly want to know - what overhead will this impose on > existing SELinux systems. > -- Serge Hallyn <serue@private>
This archive was generated by hypermail 2.1.3 : Tue Feb 22 2005 - 07:58:41 PST