From 35282a2de4e5e4e173ab61aa9d7015886021a821 Mon Sep 17 00:00:00 2001 From: Brice Goglin Date: Tue, 16 Jun 2009 15:32:43 -0700 Subject: [PATCH] migration: only migrate_prep() once per move_pages() migrate_prep() is fairly expensive (72us on 16-core barcelona 1.9GHz). Commit 3140a2273009c01c27d316f35ab76a37e105fdd8 improved move_pages() throughput by breaking it into chunks, but it also made migrate_prep() be called once per chunk (every 128pages or so) instead of once per move_pages(). This patch reverts to calling migrate_prep() only once per chunk as we did before 2.6.29. It is also a followup to commit 0aedadf91a70a11c4a3e7c7d99b21e5528af8d5d ("mm: move migrate_prep out from under mmap_sem"). This improves migration throughput on the above machine from 600MB/s to 750MB/s. Signed-off-by: Brice Goglin Acked-by: Christoph Lameter Cc: KOSAKI Motohiro Cc: Heiko Carstens Cc: Nick Piggin Cc: Hugh Dickins Cc: Rik van Riel Cc: Lee Schermerhorn Reviewed-by: KAMEZAWA Hiroyuki Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- mm/migrate.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/mm/migrate.c b/mm/migrate.c index 5a24923e7fd..939888f9dda 100644 --- a/mm/migrate.c +++ b/mm/migrate.c @@ -820,7 +820,6 @@ static int do_move_page_to_node_array(struct mm_struct *mm, struct page_to_node *pp; LIST_HEAD(pagelist); - migrate_prep(); down_read(&mm->mmap_sem); /* @@ -907,6 +906,9 @@ static int do_pages_move(struct mm_struct *mm, struct task_struct *task, pm = (struct page_to_node *)__get_free_page(GFP_KERNEL); if (!pm) goto out; + + migrate_prep(); + /* * Store a chunk of page_to_node array in a page, * but keep the last one as a marker