Re: Cookie jar patch for Supexec/Super patch?

[Hartmut Keller <Keller@informatik.uni-stuttgart.de>]

>>>>> Andreas Schwab <schwab@issan.informatik.uni-dortmund.de> writes:

> Don't forget the non-GEM programs that never heard about MiNT, or those
> that don't have the sources available or are compiled with strange
> compilers.  I'd guess that nearly all of them will want to access TOS
> variables.

Really? I have written lots of programs myself, but I felt the need to
access system variables only in very rare cases. So I think your
assumption is a very pessimistic one.

>>>>> Sven Karlsson <sven@it.lth.se> had an idea a few days ago:

> And add a flag in the prg header to control super/supexec behaivour.

Yes, this flag would control whether the program is secure (new
version) or not (default, all old programs). Then we can add a switch
to mint that allows execution of insecure programs (then all programs
can be used) or not (then all suspicious accesses like calls to
super/supexec are denied). When you want a secure system, you simply
switch to secure mode and then obviously only secure programs keep
working.

Instead of setting it in the prg-header, we can also set it somewhere
else, e.g. at a defined location in the text or data section of the
program, guarded by some magic value to avoid misinterpretation of old
programs.


  Hartmut


+-------------------------+-----------------------------------------------+
|  Hartmut Keller         |  Internet: keller@informatik.uni-stuttgart.de | 
|  Inst. fuer Informatik  +-----------------------------------------------+
|  Breitwiesenstr. 20-22  |  "If there's only one reason to live in this  |
|  70565 Stuttgart        |   world, I'll find it!"      (All About Eve)  |
+-------------------------+-----------------------------------------------+