[tip:,sched/core] sched/numa: Move up the access pid reset logic

Message ID 169693102782.3135.18020196809878344961.tip-bot2@tip-bot2
State New
Headers
Series [tip:,sched/core] sched/numa: Move up the access pid reset logic |

Commit Message

tip-bot2 for Thomas Gleixner Oct. 10, 2023, 9:43 a.m. UTC
  The following commit has been merged into the sched/core branch of tip:

Commit-ID:     2e2675db1906ac04809f5399bf1f5e30d56a6f3e
Gitweb:        https://git.kernel.org/tip/2e2675db1906ac04809f5399bf1f5e30d56a6f3e
Author:        Raghavendra K T <raghavendra.kt@amd.com>
AuthorDate:    Tue, 10 Oct 2023 09:31:41 +01:00
Committer:     Ingo Molnar <mingo@kernel.org>
CommitterDate: Tue, 10 Oct 2023 11:10:01 +02:00

sched/numa: Move up the access pid reset logic

Recent NUMA hinting faulting activity is reset approximately every
VMA_PID_RESET_PERIOD milliseconds. However, if the current task has not
accessed a VMA then the reset check is missed and the reset is potentially
deferred forever. Check if the PID activity information should be reset
before checking if the current task recently trapped a NUMA hinting fault.

[ mgorman@techsingularity.net: Rewrite changelog ]

Suggested-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Raghavendra K T <raghavendra.kt@amd.com>
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20231010083143.19593-5-mgorman@techsingularity.net
---
 kernel/sched/fair.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)
  

Patch

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 31cfdb0..ce36969 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -3260,16 +3260,7 @@  static void task_numa_work(struct callback_head *work)
 			continue;
 		}
 
-		/* Do not scan the VMA if task has not accessed */
-		if (!vma_is_accessed(vma)) {
-			trace_sched_skip_vma_numa(mm, vma, NUMAB_SKIP_PID_INACTIVE);
-			continue;
-		}
-
-		/*
-		 * RESET access PIDs regularly for old VMAs. Resetting after checking
-		 * vma for recent access to avoid clearing PID info before access..
-		 */
+		/* RESET access PIDs regularly for old VMAs. */
 		if (mm->numa_scan_seq &&
 				time_after(jiffies, vma->numab_state->pids_active_reset)) {
 			vma->numab_state->pids_active_reset = vma->numab_state->pids_active_reset +
@@ -3278,6 +3269,12 @@  static void task_numa_work(struct callback_head *work)
 			vma->numab_state->pids_active[1] = 0;
 		}
 
+		/* Do not scan the VMA if task has not accessed */
+		if (!vma_is_accessed(vma)) {
+			trace_sched_skip_vma_numa(mm, vma, NUMAB_SKIP_PID_INACTIVE);
+			continue;
+		}
+
 		do {
 			start = max(start, vma->vm_start);
 			end = ALIGN(start + (pages << PAGE_SHIFT), HPAGE_SIZE);