From: Peter Zijlstra <peterz@xxxxxxxxxxxxx> commit d6097c9e4454adf1f8f2c9547c2fa6060d55d952 upstream. Unless the very next line is schedule(), or implies it, one must not use preempt_enable_no_resched(). It can cause a preemption to go missing and thereby cause arbitrary delays, breaking the PREEMPT=y invariant. Link: http://lkml.kernel.org/r/20190423200318.GY14281@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx Cc: Waiman Long <longman@xxxxxxxxxx> Cc: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx> Cc: Ingo Molnar <mingo@xxxxxxxxxx> Cc: Will Deacon <will.deacon@xxxxxxx> Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx> Cc: the arch/x86 maintainers <x86@xxxxxxxxxx> Cc: Davidlohr Bueso <dave@xxxxxxxxxxxx> Cc: Tim Chen <tim.c.chen@xxxxxxxxxxxxxxx> Cc: huang ying <huang.ying.caritas@xxxxxxxxx> Cc: Roman Gushchin <guro@xxxxxx> Cc: Alexei Starovoitov <ast@xxxxxxxxxx> Cc: Daniel Borkmann <daniel@xxxxxxxxxxxxx> Cc: stable@xxxxxxxxxxxxxxx Fixes: 2c2d7329d8af ("tracing/ftrace: use preempt_enable_no_resched_notrace in ring_buffer_time_stamp()") Signed-off-by: Peter Zijlstra (Intel) <peterz@xxxxxxxxxxxxx> Signed-off-by: Steven Rostedt (VMware) <rostedt@xxxxxxxxxxx> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- kernel/trace/ring_buffer.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -700,7 +700,7 @@ u64 ring_buffer_time_stamp(struct ring_b preempt_disable_notrace(); time = rb_time_stamp(buffer); - preempt_enable_no_resched_notrace(); + preempt_enable_notrace(); return time; }
