Skip to contentSkip to conversation

perf, x86: Fix double disable calls

tip:perf/urgent tip-bot for Peter Zijlstra 1 participant 1 message Mar 10, 2010

Subsystems (2)
lore
Share & tools
lore
#0 lore raw
Commit-ID:  19925ce778f9fc371b9607625de3bff04c60121e
Gitweb:     http://git.kernel.org/tip/19925ce778f9fc371b9607625de3bff04c60121e
Author:     Peter Zijlstra <[email protected]>
AuthorDate: Sat, 6 Mar 2010 13:20:40 +0100
Committer:  Ingo Molnar <[email protected]>
CommitDate: Wed, 10 Mar 2010 13:22:33 +0100

perf, x86: Fix double disable calls

hw_perf_enable() would disable events that were not yet enabled.

This causes problems with code that assumes that ->enable/->disable calls are balanced (like the LBR code does).

What happens is that we disable newly added counters that match their previous assignment, even though they are not yet programmed on the hardware.

Avoid this by only doing the first pass over the existing events.

LKML-Reference: <new-submission>

---
 arch/x86/kernel/cpu/perf_event.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)
arch/x86/kernel/cpu/perf_event.c +2 -1
index 071c840..045cc0b 100644
@@ -802,6 +802,7 @@ void hw_perf_enable(void)
 		return;
 
 	if (cpuc->n_added) {
+		int n_running = cpuc->n_events - cpuc->n_added;
 		/*
 		 * apply assignment obtained either from
 		 * hw_perf_group_sched_in() or x86_pmu_enable()
@@ -809,7 +810,7 @@ void hw_perf_enable(void)
 		 * step1: save events moving to new counters
 		 * step2: reprogram moved events into new counters
 		 */
-		for (i = 0; i < cpuc->n_events; i++) {
+		for (i = 0; i < n_running; i++) {
 
 			event = cpuc->event_list[i];
 			hwc = &event->hw;