Re: [patch] hugepage allocator cleanup

[William Lee Irwin III <wli@holomorphy.com>]

On Wed, Jan 25, 2006 at 08:32:43AM -0800, William Lee Irwin III wrote:
>> Just yanking the page refcount affairs out of update_and_free_page()
>> should suffice. Could I get things trimmed down to that?
>> 

On Wed, Jan 25, 2006 at 05:52:08PM +0100, Nick Piggin wrote:
> I could remove the first set_page_count, and make the second conditional
> on the page having a zero refcount... for a 3-liner. But that's kind of
> ugly (if less intrusive), and it is adds seemingly nonsense code if one
> doesn't have the context of my out-of-tree patches.
> Hmm... it's obviously not 2.6.16 material so there is no rush to think
> it over. It is even simple enough that I don't mind carrying with my
> patchset indefinitely.

After I thought about it, alloc_fresh_huge_page() does enqueue pages
with refcounts of 1, where free_huge_page() (called from the freeing
hook in page[1].mapping) enqueues pages with refcounts of 0, so it
would actually make sense (and possibly prevent leaks) to take the
whole patch as-is.


-- wli

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>