Some areas use preempt_disable() + preempt_enable() to safely access
per-CPU data. The PREEMPT_RT folks have shown this can also be done by
keeping preemption enabled and instead disabling migration (and acquiring a
sleepable lock, if relevant).
Introduce a helper which checks whether the current task can be migrated
elsewhere, IOW if it is pinned to its local CPU in the current
context. This can help determining if per-CPU properties can be safely
accessed.
Note that CPU affinity is not checked here, as a preemptible task can have
its affinity changed at any given time (including if it has
PF_NO_SETAFFINITY, when hotplug gets involved).
Signed-off-by: Valentin Schneider <valentin.schneider@xxxxxxx>
---
include/linux/sched.h | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index debc960f41e3..8ba7b4a7ee69 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1715,6 +1715,16 @@ static inline bool is_percpu_thread(void)
#endif
}
+/* Is the current task guaranteed to stay on its current CPU? */
+static inline bool migratable(void)
+{
+#ifdef CONFIG_SMP
+ return preemptible() && !current->migration_disabled;
+#else
+ return true;
+#endif
+}
+
/* Per-process atomic flags. */
#define PFA_NO_NEW_PRIVS 0 /* May not gain new privileges. */
#define PFA_SPREAD_PAGE 1 /* Spread page cache over cpuset */
--
2.25.1
