On Tue, Aug 16, 2016 at 08:15:36AM +0200, Vlastimil Babka wrote:

On 08/16/2016 08:15 AM, Joonsoo Kim wrote:

quoted

On Wed, Aug 10, 2016 at 11:12:23AM +0200, Vlastimil Babka wrote:

quoted

--- a/include/linux/compaction.h
+++ b/include/linux/compaction.h

@@ -58,6 +58,22 @@ enum compact_result {

struct alloc_context; /* in mm/internal.h */

+/*
+ * Number of free order-0 pages that should be available above given watermark
+ * to make sure compaction has reasonable chance of not running out of free
+ * pages that it needs to isolate as migration target during its work.
+ */
+static inline unsigned long compact_gap(unsigned int order)
+{
+	/*
+	 * Although all the isolations for migration are temporary, compaction
+	 * may have up to 1 << order pages on its list and then try to split
+	 * an (order - 1) free page. At that point, a gap of 1 << order might
+	 * not be enough, so it's safer to require twice that amount.
+	 */
+	return 2UL << order;
+}

I agree with this wrapper function but there is a question.

Could you elaborate more on this code comment? Freescanner could keep
COMPACT_CLUSTER_MAX freepages on the list. It's not associated with
requested order at least for now. Why compact_gap is 2UL << order in
this case?

It's true that for high enough order, COMPACT_CLUSTER_MAX might be
more limiting than 1 << order. But then it also helps to have more

AFAIK, regardless of order, migration scanner isolates
COMPACT_CLUSTER_MAX pages. And, freepage scanner isolates
more than nr_migratepages freepages.

free pages for probability of compaction success, so I don't think
it's worth complicating the compact_gap() formula.

I agree that it's not worth complicating the compact_gap() formula but
it would be better to fix the comment?

Thanks.

--
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>