Re: [RFC] Convert pgtable cache to slab

["Martin K. Petersen" <mkp@sgi.com>]

>>>>> "Tony" = Luck, Tony <tony.luck@intel.com> writes:

>>> Perhaps "zero_cache" isn't as descriptive a name as it might be
>>> (not that I have any better suggestions :-(
>>  Maybe pgtable_zero_cache or something?

Tony> I thought along those lines at first, but actually this cache
Tony> would be useful for any allocation that needed a page that has
Tony> been pre-zeroed.  So I don't think that the "pgtable" part is
Tony> helpful.  Some sense of the fact that the objects are one page
Tony> big is what I was looking for.

Yeah.  I suspect that's why Bill named it that in the first place.

What about zeroed_page_cache?

Now, this opens up another can of worms in terms of where to
initialize that cache in the first place.  pgtable_cache_init() might
not be the best place if it's generally used.  Only requirement is
that it needs to be early.

-- 
Martin K. Petersen	Silicon Graphics, Inc.
mkp@sgi.com		http://www.sgi.com/