Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 442c9137 authored by Christoph Lameter's avatar Christoph Lameter Committed by Linus Torvalds
Browse files

[PATCH] More page migration: do not inc/dec rss counters 



If we install a migration entry then the rss not really decreases since the
page is just moved somewhere else.  We can save ourselves the work of
decrementing and later incrementing which will just eventually cause cacheline
bouncing.

Signed-off-by: default avatarChristoph Lameter <clameter@sgi.com>
Signed-off-by: default avatarAndrew Morton <akpm@osdl.org>
Signed-off-by: default avatarLinus Torvalds <torvalds@osdl.org>
parent 6c5240ae

mm/migrate.c

+0 −1
Original line number Diff line number Diff line
@@ -164,7 +164,6 @@ static void remove_migration_pte(struct vm_area_struct *vma, unsigned long addr, 
	if (!is_migration_entry(entry) || migration_entry_to_page(entry) != old)

		goto out;



	inc_mm_counter(mm, anon_rss);

	get_page(new);

	pte = pte_mkold(mk_pte(new, vma->vm_page_prot));

	if (is_write_migration_entry(entry))

mm/rmap.c

+1 −1
Original line number Diff line number Diff line
@@ -595,6 +595,7 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma, 
					list_add(&mm->mmlist, &init_mm.mmlist);

				spin_unlock(&mmlist_lock);

			}

			dec_mm_counter(mm, anon_rss);

		} else {

			/*

			 * Store the pfn of the page in a special migration
@@ -606,7 +607,6 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma, 
		}

		set_pte_at(mm, address, pte, swp_entry_to_pte(entry));

		BUG_ON(pte_file(*pte));

		dec_mm_counter(mm, anon_rss);

	} else

		dec_mm_counter(mm, file_rss);