On Tue, Mar 26, 2002 at 07:43:37PM +0000, Charles M. Hannum wrote:
> 
> One idea I tossed around a while back was keeping track of pages that
> are zeroed but have not actually been written to.  This actually
> requires no more faults -- it's handled entirely by clearing the
> "zeroed" bit whenever the page is detected to be modified.

"How"? The modified bits I'm familiar with are implemented in the fault trap
handlers... which architectures have hardware implemented modified bits?

> This would allow us to, e.g., allocate stack pages in batches and then
> free up ones that aren't actually in use.

Regards,
	-is