.NET support in UNIX
Marius van OersMcAfee Avert
Technical stream: Thursday 25 Sept 2003, 15.40-16.20.
.NET support in UNIX
With .NET, instead of native code it is also possible to provide code in an intermediate form,
using the Intermediate Language, IL. When the user launches a file the IL code is compiled by
JIT (Just In Time Compiling) in memory into native code, and the native code is executed.
The disadvantage is speed-decrease, the advantage is that applications can run on more
OS/versions so eliminating the need for a specific application per operating system
(WinME/2K/XP etc).
The use of Unix/Linux as operating system is increasing rapidly on areas like File, Mail and
Web-servers. Here the range of operating systems/flavors/kernel versions is even bigger
(SunOS/Linux RedHat/Suse, BeOS etc) Applications can be OS/flavor/kernel version specific so
to make/distribute code for Unix is not easy.
So in both the Windows and Unix worlds, native code is faster but limits the distribution.
.NET addresses this for the Microsoft Windows world, and .NET may also be supported in the
Unix world. Currently there are projects going on (example Mono) in an effort to create an
open source implementation of the .NET framework.
With this development environment it will allow applications developed for .NET to run on
Windows and Unix/Linux-based operating systems. Malware (virus/Trojan) writers may move to the
.NET environment, thus possibly creating cross-platform malicious binaries.
As the output binaries don't contain native code but (abstract) IL code, this also poses another
issue. So .NET support might introduce security issues in the Windows and Unix worlds. This
paper will mainly focus on the .NET support in Unix and the possible security issues related to
that.
