Kernel 2.4.2x and 2.6.x (x86 only) DoS

Apparently, crashing the 2.4.2x and 2.6.x kernels on x86 architecture is
trivial, from any user account. Not good news!

http://cve.mitre.org/cgi-bin/cvename...=CAN-2004-0554

The original kernel patch I saw only fixed a single instance of the bug,
and was not really a solution. Does anybody know if a more thorough kernel
patch has been released yet? This problem appears to be due to a rather
well known problem in interrupt handling, when floating point operations
are involved

http://www.aero.polimi.it/~rtai/docu...rticles/paolo-
floatingpoint.html

--
Jem Berkes
http://www.sysdesign.ca/

Jem Berkes wrote:
> Apparently, crashing the 2.4.2x and 2.6.x kernels on x86 architecture is
> trivial, from any user account. Not good news!


After compling and running this code: It will crash the kernel, but on
SMP kernels you have to rum multiple instances of it. It will kill 1 cpu
per instance.

> http://cve.mitre.org/cgi-bin/cvename...=CAN-2004-0554
>
> The original kernel patch I saw only fixed a single instance of the bug,
> and was not really a solution. Does anybody know if a more thorough kernel
> patch has been released yet? This problem appears to be due to a rather
> well known problem in interrupt handling, when floating point operations
> are involved
>
> http://www.aero.polimi.it/~rtai/docu...rticles/paolo-
> floatingpoint.html
>