Re: make swappiness safer to use

On Wed, Aug 01, 2007 at 09:19:25AM +0800, Fengguang Wu wrote:

On Wed, Aug 01, 2007 at 01:02:51AM +0200, Andrea Arcangeli wrote:

quoted

diff --git a/mm/vmscan.c b/mm/vmscan.c

[...]

quoted

@@ -912,6 +913,44 @@ static void shrink_active_list(unsigned 
 		 * altogether.
 		 */
 		swap_tendency = mapped_ratio / 2 + distress + sc->swappiness;
+
+		/*
+		 * If there's huge imbalance between active and inactive
+		 * (think active 100 times larger than inactive) we should
+		 * become more permissive, or the system will take too much
+		 * cpu before it start swapping during memory pressure.
+		 * Distress is about avoiding early-oom, this is about
+		 * making swappiness graceful despite setting it to low
+		 * values.
+		 *
+		 * Avoid div by zero with nr_inactive+1, and max resulting
+		 * value is vm_total_pages.
+		 */
+		imbalance = zone_page_state(zone, NR_ACTIVE) /
+                        (zone_page_state(zone, NR_INACTIVE) + 1);
+
+		/*
+		 * Reduce the effect of imbalance if swappiness is low,
+		 * this means for a swappiness very low, the imbalance
+		 * must be much higher than 100 for this logic to make
+		 * the difference.
+		 *
+		 * Max temporary value is vm_total_pages*100.
+		 */
+		imbalance *= (vm_swappiness + 1) / 100;

                             ~~~~~~~~~~~~~~~~~~~~~~~~~ It will be zero!

You know what, while editing I even wrote "imbalance = imbalance *
((vm_swappiness + 1) / 100" because I felt it was different, then I
edited it back to the cut-and-pasted version thinking it was actually
the same, but it's not.

Better to scale it up before the division:
		imbalance *= (vm_swappiness + 1) * 1024 / 100;

quoted

+
+		/*
+		 * If not much of the ram is mapped, makes the imbalance
+		 * less relevant, it's high priority we refill the inactive
+		 * list with mapped pages only in presence of high ratio of
+		 * mapped pages.
+		 *
+		 * Max temporary value is vm_total_pages*100.
+		 */
+		imbalance *= mapped_ratio / 100;

		imbalance *= mapped_ratio * 1024 / 100;

quoted

+		/* apply imbalance feedback to swap_tendency */
+		swap_tendency += imbalance;

		swap_tendency += imbalance / 1024 / 1024;

No this is exactly what we want to avoid. We must not allow more than
100 times higher the number of pages.

The fix is only to do "imbalance = imbalance *" instead of "imbalance
*=".

--
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>