Stacking - anyone care how to report module id's?

Previous message: linux-helpat_private: "GOODBYE from linuxat_private"
Next in thread: Greg KH: "Re: Stacking - anyone care how to report module id's?"
Reply: Greg KH: "Re: Stacking - anyone care how to report module id's?"
Reply: Crispin Cowan: "Re: Stacking - anyone care how to report module id's?"
Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

dwheelerat_private

I'm still interested in completing a stacking module, but
I have two questions for everyone:

1. I need the "module id" of a stacked module, if it has one,
    to implement returns from sys_security.  There are two options:
    A. Add it to the struct security_ops.  This makes module id of the
    primary module visible to "everyone", which might be useful for
    some queries. It requires modification of the verification #define.
    B. Add it as a parameter to mod_reg_security().  The advantages here
    are that, in theory, a module could call mod_reg_security multiple
    times with different module id's.

I'm currently leaning toward option B (modifying mod_reg_security).
Any comments?

2. I have completely FAILED to find out if pointer writes are
    atomic (and if not, where), and I seem to have trouble posting to
    the linux mailing list.  I'm currently thinking of just writing
    the code and making that assumption (that pointer writes
    are atomic).  If anyone can help me, that'd be great.

    Would anyone (Greg KH?) be willing to ask the question for me on
    the linux kernel mailing list, since I can't seem to post on it?
    ("Are pointer writes atomic on all architectures?  If not, on which
     architectures is this assumption true or not true?")

Thanks.

--- David A. Wheeler
     dwheelerat_private

_______________________________________________
linux-security-module mailing list
linux-security-moduleat_private
http://mail.wirex.com/mailman/listinfo/linux-security-module