Re: problems with hierarchy: L4 pagers

[Jonathan S. Shapiro]

On Mon, 2005-10-31 at 17:47 +0100, Espen Skoglund wrote:
> Remember
> that faults on endpoints (communication channels) are directed to the
> compager of the faulting thread (as opposed to memory pager---I can
> already hear Jonathan's teeth grinding at the word "pager" here), and
> the compager need not be the same as the memory pager.

At the moment, the only grinding of my teeth is frustration with myself
that I find it difficult to wrap my head around something that is
perfectly clear to you. I would not expect a second pager to introduce
any new architecture concerns that were already in the first.

I do find it interesting that faults are now being segregated to
different pager types. In a small way, this is similar to the type
distinction between memory faults and execution faults in EROS --
except, of course, that in our case there are no endpoint communication
faults that result from unmapping an endpoint.

I should expand on the similarity, but I need to prepare for a meeting.

shap