timers/migration: Spare write when nothing changed
authorAnna-Maria Behnsen <anna-maria@linutronix.de>
Tue, 16 Jul 2024 14:19:25 +0000 (16:19 +0200)
committerThomas Gleixner <tglx@linutronix.de>
Mon, 22 Jul 2024 16:03:34 +0000 (18:03 +0200)
The wakeup value is written unconditionally in tmigr_cpu_new_timer(). When
there was no new next timer expiry that needs to be propagated, then the
value that was read before is written. This is not required.

Move the write to the place where wakeup value is changed changed.

Signed-off-by: Anna-Maria Behnsen <anna-maria@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Frederic Weisbecker <frederic@kernel.org>
Link: https://lore.kernel.org/r/20240716-tmigr-fixes-v4-7-757baa7803fe@linutronix.de
kernel/time/timer_migration.c

index ca76120..9c15ae8 100644 (file)
@@ -1215,14 +1215,13 @@ u64 tmigr_cpu_new_timer(u64 nextexp)
                if (nextexp != tmc->cpuevt.nextevt.expires ||
                    tmc->cpuevt.ignore) {
                        ret = tmigr_new_timer(tmc, nextexp);
+                       /*
+                        * Make sure the reevaluation of timers in idle path
+                        * will not miss an event.
+                        */
+                       WRITE_ONCE(tmc->wakeup, ret);
                }
        }
-       /*
-        * Make sure the reevaluation of timers in idle path will not miss an
-        * event.
-        */
-       WRITE_ONCE(tmc->wakeup, ret);
-
        trace_tmigr_cpu_new_timer_idle(tmc, nextexp);
        raw_spin_unlock(&tmc->lock);
        return ret;