Elad Efrat wrote:
> Jason Thorpe wrote:
> 
> 
>>On Dec 18, 2005, at 4:03 AM, Martin Husemann wrote:
> 
> 
>>>The mprotect changes are arch dependend, and as gimpy pointed out  would
>>>violate ABI on some archs. I don't see how implementing them (which 
>>>would
>>>be MD anyway) would cause any negative impact on other archs.
>>
>>
>>Yah, these would have to be implemented using new pmap hooks.
> 
> 
> Why wouuld these require pmap changes? they (should) be pure
> mmap/mprotect changes. Do you want to implement them in pmap because
> they are MD, or is there a technical need?

My proposal doesn't need pmap hooks.  However, VM placement does (see
PMAP_PREFER() for instance) for virtual caches.

-- 
Matt Thomas                     email: matt@3am-software.com
3am Software Foundry              www: http://3am-software.com/bio/matt/
Cupertino, CA              disclaimer: I avow all knowledge of this message.