[Mono-devel-list] Updated CAS patch for LinkDemand

[Sebastien Pouliot]

> -----Original Message-----
> From: mono-devel-list-admin at lists.ximian.com
> [mailto:mono-devel-list-admin at lists.ximian.com]On Behalf Of Paolo Molaro
> Sent: 13 fevrier 2005 11:34
> To: mono-devel-list at lists.ximian.com
> Subject: Re: [Mono-devel-list] Updated CAS patch for LinkDemand
> 
> 
> On 02/10/05 Sebastien Pouliot wrote:
> > > Uhm, when char is signed it may be possible to get a public key 
> > 
> > Oops.
> > 
> > > (though with likely quite a bit of cpu time) that gives 4 as the sum.
> > 
> > Actually not so much cpu as the 16 bytes are just an (empty) header.
> 
> But the result would likely not be a valid public key, right?

Interpreting the header results in a 0 bits public key - so it cannot be, 
crypto speaking, used to validate anything.

> The public key would have to be correct wrt the signing of the
> assembly and give 4 as a sum. Please correct me if I'm wrong.

However if the same code was used to detect the ECMA key then the runtime
public key would be used (instead of the 0 bit public key). In this case
any header (with a sum of four would be valid) would be correct as long as
the runtime private key has signed the assembly.

Unless of course the assembly signature isn't verified ;-)

Sebastien