memory-hotplug: add sysfs removable attribute for hotplug memory remove

Memory may be hot-removed on a per-memory-block basis, particularly on
POWER where the SPARSEMEM section size often matches the memory-block
size.  A user-level agent must be able to identify which sections of
memory are likely to be removable before attempting the potentially
expensive operation.  This patch adds a file called "removable" to the
memory directory in sysfs to help such an agent.  In this patch, a memory
block is considered removable if;

o It contains only MOVABLE pageblocks
o It contains only pageblocks with free pages regardless of pageblock type

On the other hand, a memory block starting with a PageReserved() page will
never be considered removable.  Without this patch, the user-agent is
forced to choose a memory block to remove randomly.

Sample output of the sysfs files:

./memory/memory0/removable: 0
./memory/memory1/removable: 0
./memory/memory2/removable: 0
./memory/memory3/removable: 0
./memory/memory4/removable: 0
./memory/memory5/removable: 0
./memory/memory6/removable: 0
./memory/memory7/removable: 1
./memory/memory8/removable: 0
./memory/memory9/removable: 0
./memory/memory10/removable: 0
./memory/memory11/removable: 0
./memory/memory12/removable: 0
./memory/memory13/removable: 0
./memory/memory14/removable: 0
./memory/memory15/removable: 0
./memory/memory16/removable: 0
./memory/memory17/removable: 1
./memory/memory18/removable: 1
./memory/memory19/removable: 1
./memory/memory20/removable: 1
./memory/memory21/removable: 1
./memory/memory22/removable: 1

Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

1 files changed, 60 insertions, 0 deletions

diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 93aba78dc8b..89fee2dcb03 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -523,6 +523,66 @@ EXPORT_SYMBOL_GPL(add_memory);
 
 #ifdef CONFIG_MEMORY_HOTREMOVE
 /*
+ * A free page on the buddy free lists (not the per-cpu lists) has PageBuddy
+ * set and the size of the free page is given by page_order(). Using this,
+ * the function determines if the pageblock contains only free pages.
+ * Due to buddy contraints, a free page at least the size of a pageblock will
+ * be located at the start of the pageblock
+ */
+static inline int pageblock_free(struct page *page)
+{
+	return PageBuddy(page) && page_order(page) >= pageblock_order;
+}
+
+/* Return the start of the next active pageblock after a given page */
+static struct page *next_active_pageblock(struct page *page)
+{
+	int pageblocks_stride;
+
+	/* Ensure the starting page is pageblock-aligned */
+	BUG_ON(page_to_pfn(page) & (pageblock_nr_pages - 1));
+
+	/* Move forward by at least 1 * pageblock_nr_pages */
+	pageblocks_stride = 1;
+
+	/* If the entire pageblock is free, move to the end of free page */
+	if (pageblock_free(page))
+		pageblocks_stride += page_order(page) - pageblock_order;
+
+	return page + (pageblocks_stride * pageblock_nr_pages);
+}
+
+/* Checks if this range of memory is likely to be hot-removable. */
+int is_mem_section_removable(unsigned long start_pfn, unsigned long nr_pages)
+{
+	int type;
+	struct page *page = pfn_to_page(start_pfn);
+	struct page *end_page = page + nr_pages;
+
+	/* Check the starting page of each pageblock within the range */
+	for (; page < end_page; page = next_active_pageblock(page)) {
+		type = get_pageblock_migratetype(page);
+
+		/*
+		 * A pageblock containing MOVABLE or free pages is considered
+		 * removable
+		 */
+		if (type != MIGRATE_MOVABLE && !pageblock_free(page))
+			return 0;
+
+		/*
+		 * A pageblock starting with a PageReserved page is not
+		 * considered removable.
+		 */
+		if (PageReserved(page))
+			return 0;
+	}
+
+	/* All pageblocks in the memory block are likely to be hot-removable */
+	return 1;
+}
+
+/*
  * Confirm all pages in a range [start, end) is belongs to the same zone.
  */
 static int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn)