[PATCH] mm/vmpressure: scale window size based on machine memory

From: Benjamin Lee McQueen

Date: Thu Feb 26 2026 - 17:23:29 EST


vmpressure's window size was fixed at 512 pages regardless of machine size.

This fix, from a TODO scales the window size based on RAM size of the machine.
The method of scaling is similar to vmstat's scaling but based on RAM alone.

Signed-off-by: Benjamin Lee McQueen <mcq@xxxxxxxxxxx>
---
mm/vmpressure.c | 15 ++++++++++++---
1 file changed, 12 insertions(+), 3 deletions(-)

diff --git a/mm/vmpressure.c b/mm/vmpressure.c
index 3fbb86996c4d..1d7b5d9e4351 100644
--- a/mm/vmpressure.c
+++ b/mm/vmpressure.c
@@ -32,11 +32,20 @@
* As the vmscan reclaimer logic works with chunks which are multiple of
* SWAP_CLUSTER_MAX, it makes sense to use it for the window size as well.
*
- * TODO: Make the window size depend on machine size, as we do for vmstat
- * thresholds. Currently we set it to 512 pages (2MB for 4KB pages).
+ * The window size scales with machine memory using logarithmic scaling
+ * Smaller machines get smaller window size, while larger ones, for
+ * example, get proportionally larger window size.
*/
-static const unsigned long vmpressure_win = SWAP_CLUSTER_MAX * 16;
+static unsigned long vmpressure_win;

+static int __init vmpressure_win_init(void)
+{
+ unsigned long mem = totalram_pages() >> (27 - PAGE_SHIFT);
+
+ vmpressure_win = SWAP_CLUSTER_MAX * max(16UL, (unsigned long)fls(mem) * 8);
+ return 0;
+}
+core_initcall(vmpressure_win_init);
/*
* These thresholds are used when we account memory pressure through
* scanned/reclaimed ratio. The current values were chosen empirically. In
--
2.53.0