[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

MMU support questions



On Mar 28, 2011, at 4:25 , Tobias Schoofs wrote:

> Is it wise to protect memory on a per-task basis? This would lead to a 
> concept of heavy task where each context-switch includes an address 
> space switch.
I'll read the rest of your response carefully later to understand it, but I want to respond to this right away.

I don't think the overhead of switching a few TLB entries to enable / disable the mapping of a stack and heap region associated with a task will cost much, especially compared to the additional robustness it will provide.  We aren't talking about a private virtual address space, but two TLB entries.  That behavior would have to be selectable anyway, many applications would fail if their stack was suddenly private.  I now need to go back and read what you wrote to see what use of the MMU you're proposing that will be inherently light-weight.

Peter
-----------------
Peter Dufault
HD Associates, Inc.      Software and System Engineering