bug-classpath
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Bug classpath/23921] gnu.classpath access restrictions are not enforced


From: gnu_andrew at member dot fsf dot org
Subject: [Bug classpath/23921] gnu.classpath access restrictions are not enforced by default
Date: 6 Oct 2007 22:54:32 -0000


------- Comment #1 from gnu_andrew at member dot fsf dot org  2007-10-06 22:54 
-------
We ran into a problem related to this with JikesRVM.  JikesRVM does have
protection on the java and gnu.classpath namespaces, which means that in
previous releases (it's fixed in svn, but via a hack) it would block running
any of the classpath tools or examples.  We now block only:

       "Lgnu/classpath/debug/".getBytes(),
       "Lgnu/classpath/jdwp/".getBytes(),
       "Lgnu/classpath/NotImplementedException".getBytes(),
       "Lgnu/classpath/Pair".getBytes(),
       "Lgnu/classpath/Pointer".getBytes(),
       "Lgnu/classpath/Pointer32".getBytes(),
       "Lgnu/classpath/Pointer64".getBytes(),
       "Lgnu/classpath/ServiceFactory".getBytes(),
       "Lgnu/classpath/ServiceProviderLoadingAction".getBytes(),
       "Lgnu/classpath/SystemProperties".getBytes(),

Obviously this is nasty if a class is added that needs protection.  We should
decide which of these should be protected and create an internal subpackage as
dalibor suggests.  Failing that, examples and tools both need to be moved
outside the gnu.classpath umbrella (which works for tools but not really for
examples).

The other issue is applying the protection at the Classpath level.  The VM
still needs to be able to extend this security.  JikesRVM protects its own
internal classes org.jikesrvm and org.mmtk as well.  Is it worth moving part of
the solution to Classpath, given the VM must do something anyway?


-- 

gnu_andrew at member dot fsf dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |gnu_andrew at member dot fsf
                   |dot org                     |dot org
           Severity|normal                      |enhancement
             Status|UNCONFIRMED                 |ASSIGNED
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2007-10-06 22:54:31
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23921





reply via email to

[Prev in Thread] Current Thread [Next in Thread]