Locking and/ or running inside interrupt handler will not lead to an
atomic section on PREEMPT_RT. The RCU read section needs to be
considered because all locks, which become sleeping locks on
PREEMPT_RT, start a RCU read section. Scheduling/ sleeping while within
a RCU read section is invalid.
Check for also for RCU read section in stop_timeout() to determine if it
is safe to sleep.
Signed-off-by: Sebastian Andrzej Siewior <bige...@linutronix.de>
---
drivers/gpu/drm/i915/gt/intel_engine_cs.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/i915/gt/intel_engine_cs.c
b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
index b721bbd233567..f5a6143ea8a24 100644
--- a/drivers/gpu/drm/i915/gt/intel_engine_cs.c
+++ b/drivers/gpu/drm/i915/gt/intel_engine_cs.c
@@ -1609,7 +1609,7 @@ u64 intel_engine_get_last_batch_head(const struct
intel_engine_cs *engine)
static unsigned long stop_timeout(const struct intel_engine_cs *engine)
{
- if (in_atomic() || irqs_disabled()) /* inside atomic preempt-reset? */
+ if (in_atomic() || irqs_disabled() || rcu_preempt_depth()) /* inside
atomic preempt-reset? */
return 0;
/*
--
2.50.0
