Re: proxy memory objects patch

[Marcus Brinkmann]

On Wed, Nov 20, 2002 at 09:16:12PM -0500, Roland McGrath wrote:
> If you can implement the interface that takes many objects and ranges from
> each, please do that.  The reason to want this is so that we can use it to
> compose proxy memory objects that span adjacent memory objects of logically
> contiguous underlying objects like files and disks.  That lets us use the
> new io_map interface that I prefer, and keep the mmap atomic to the user,
> for 64-bit files/devices.  (Remember the RPCs were all updated except for
> io_map, which is still constrained to the first 4G of the file.)

We knew about this type of application before, it came up in the discussion
of supporting large files.

However, my motivation to implement this *now* is to fix a major security
hole in the Hurd.  I have a lot of things to do on my plate, but this issue
was lingering around for years now, so I decided that something must be done
about it.

I would like that you and Thomas define the exact interface you want it to
be, and I will adjust my patch to only acept a single real memory object without
range restriction within that interface.  That is the minimum feature we need
to fix the current security hole, and everything else that is more fancy like
this can be done later without affecting anything.  I don't have the time to
develop, debug and tune it until it makes everyone happy right now, sorry.

Thanks,
Marcus

-- 
`Rhubarb is no Egyptian god.' GNU      http://www.gnu.org    marcus@gnu.org
Marcus Brinkmann              The Hurd http://www.gnu.org/software/hurd/
Marcus.Brinkmann@ruhr-uni-bochum.de
http://www.marcus-brinkmann.de/