[eros-arch] Re: Problem with the semantics of address translation
and keeper invocation
digitale at digitaleric.net
Fri Sep 2 23:32:57 EDT 2005
On Fri, 2005-09-02 at 23:03, Charles Landau wrote:
> At 10:53 PM -0400 9/2/05, Eric Northup wrote:
> >I don't quite understand the case where the node defining the whole
> >address space is not directly named by the address space root
> >capability. Even if there are several layers of non-translating/wrapper
> >nodes, doesn't the first wrapper still define the address space?
> In KeyKOS, a segment can be larger than the size of an address space,
> and a domain root can point to such a segment (though there's little
> reason to do so). I'm not sure if this is allowed in EROS/CapROS. The
> first node with SSC/BLSS less than or equal to 6 (corresponding to
> 2^32) defines the whole address space.
Thanks for the explanation.
EROS/CapROS support this as well. The main use case I've heard Shap
mention is for a huge address space used as a background space.
Because EROS does not share the top-level page tables, it does not
matter how the top of that page table lines up with the Node path, so
the distinction between the "first Node in the address space path" and
"the Node translating the top bit of the machine's virtual address
space" is not important.
More information about the eros-arch