Re: [RFC] 2.4.11-pre4 patch

From: jmjonesat_private
Date: Mon Oct 08 2001 - 12:50:29 PDT

  • Next message: Chris Wright: "Re: [RFC] 2.4.11-pre4 patch"

    On Mon, 8 Oct 2001, Stephen Smalley wrote:
    
    > 
    > On Fri, 5 Oct 2001, Chris Wright wrote:
    > 
    > > i think, once we reach a stable snapshot, we'll need to break the patch
    > > apart and present it to subsystem maintainers for comments.  for
    > > example, i'd like create a LSM VFS patch and propose it to the
    > > linux-fsdevel list.  i don't believe we can present the entire LSM patch
    > > and expect any useful repsonses.
    > 
    > Are you sure that this is necessary?   The SubmittingPatches document says
    > to separate each logical change into its own patch.  LSM is a single
    > logical change, i.e. the addition of security hooks to the kernel,
    > although it spans many different kernel subsystems.  Also, since LSM was
    > created specifically in response to Linus' remarks at the 2.5 kernel
    > summit about what he would be willing to consider for inclusion in the
    > mainstream Linux kernel, wouldn't it be better to send the entire patch to
    > him and to lkml for initial consideration?
    
    You know, I have to agree with Mr. Smalley.  This patch exhibits
    significant effort to "reduce" it to make it as small as possible without
    sacrificing functionality.  Along the way, there have been "you can do
    that with THIS part, so we don't need THAT thing" discussions.
    
    If you divide it any way at all, I think you will be opening the "what if"
    cases for a number of access-restriction questions.  If you CAN break off
    part of it and it doesn't do that... LSM needs to reduce farther to meet
    that objective.
    
    Perhaps it would be more useful to discuss the division you have in
    mind...  what functional divisions do you believe could stand on their own
    against the "light of scrutiny"?  Or, rather, what are the "subsystems" of
    the LSM patches that you envision addressing?  It's ALL
    access-restriction, the only concievable divisions are based on "access to
    what?"  I am very much in disbelief that you can separate access-to-files
    from access-to-sockets (for example) and still have an effectively
    restrictive policy. 
    
    Seriously, this is a tight patch,
    J. Melvin Jones
    
    > 
    > --
    > Stephen D. Smalley, NAI Labs
    > ssmalleyat_private
    > 
    > 
    > 
    > 
    > _______________________________________________
    > linux-security-module mailing list
    > linux-security-moduleat_private
    > http://mail.wirex.com/mailman/listinfo/linux-security-module
    > 
    
    
    |>------------------------------------------------------
    ||  J. MELVIN JONES            jmjonesat_private 
    |>------------------------------------------------------
    ||  Microcomputer Systems Consultant  
    ||  Software Developer
    ||  Web Site Design, Hosting, and Administration
    ||  Network and Systems Administration
    |>------------------------------------------------------
    ||  http://www.jmjones.com/
    |>------------------------------------------------------
    
    
    
    
    _______________________________________________
    linux-security-module mailing list
    linux-security-moduleat_private
    http://mail.wirex.com/mailman/listinfo/linux-security-module
    



    This archive was generated by hypermail 2b30 : Mon Oct 08 2001 - 12:52:00 PDT