On Thu, 15 Feb 2007 23:06:45 -0800 Jeremy Fitzhardinge [off-list ref] wrote:

Andrew Morton wrote:

quoted

On Thu, 15 Feb 2007 18:25:00 -0800 Jeremy Fitzhardinge [off-list ref] wrote:

  

quoted

Add a new mm function apply_to_page_range()[...]

There was some discussion about this sort of thing last week.  The
consensus was that it's better to run the callback against a whole pmd's
worth of ptes, mainly to amortise the callback's cost (a lot).

It was implemented in
ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.20/2.6.20-mm1/broken-out/smaps-extract-pmd-walker-from-smaps-code.patch

Yes I was looking at that and wondering what the upshot would be.  I'll
have a closer look, but it seems like it should be usable.

It's a question of who-merges-first.  I wasn't planning on merging the
smaps stuff into 2.6.21.

Perhaps the best approach is to proceed as-is and clean things up once it's
all merged.

I guess your pte-at-a-time walker could be quite simply implemented underneath
the smaps pmd-at-a-time walker.