Re: [PATCH] in-core AFS multiplexor and PAG support

[David Howells]

> Right, if some process/user opens a file and then passes the descriptor
> to another process/user which closes it. The close should operate under
> the same permissions as the original opener.

As long as the token isn't explicitly withdrawn. With my token structure, I've
defined it such that if the list_head in the token struct is ever empty, then
the token is withdrawn.

Furthermore, I'm considering it such that the the filesystem will select a
token from the PAG's token ring in the file_operations->open method and will
attach it to the file->f_token at that point for quick reference later.

> If someone obtains my user id on in any way (i.e. weak password/
> bufferoverflow/ root exploit), he should not be allowed to use or access
> my tokens as he hasn't proven his identity. In this case he would either
> still be in his original process authentication group, or a new and
> empty PAG. But definitely not in any of my authentication groups.
> 
> Which is also why joining a PAG should never be allowed.

Someone asked for it, but I suspect if allowed at all it may be best that this
ability is governed by its own capability bit and also that the security
interface should be consulted.

David