Re: perf: regression with PERF_EVENT_IOC_REFRESH

[Peter Zijlstra <peterz@infradead.org>]

On Tue, 2011-05-24 at 17:11 +0200, Peter Zijlstra wrote:
> On Tue, 2011-05-24 at 11:04 -0400, Vince Weaver wrote:
> > As an aside, I also wasted 6 hours last night finding out that you don't 
> > get signaled on overflow if you don't have a ring-buffer mmap()ed, even
> > if you never read from the buffer and you only are interested in counting
> > the number of overflows. 
> 
> That sounds like something we could fix, let me investigate.

Does the below cure this?

---
 kernel/events/core.c |    5 ++++-
 1 files changed, 4 insertions(+), 1 deletions(-)

diff --git a/kernel/events/core.c b/kernel/events/core.c
index c09767f..bd1ba5e 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -5018,9 +5018,12 @@ static int __perf_event_overflow(struct perf_event *event, int nmi,
 		event->pending_kill = POLL_HUP;
 		if (nmi) {
 			event->pending_disable = 1;
+			event->pending_wakeup = 1;
 			irq_work_queue(&event->pending);
-		} else
+		} else {
 			perf_event_disable(event);
+			perf_event_wakeup(event);
+		}
 	}
 
 	if (event->overflow_handler)