perf_counter: More paranoia settings
Rename the perf_counter_priv knob to perf_counter_paranoia (because
priv can be read as private, as opposed to privileged) and provide
one more level:
0 - permissive
1 - restrict cpu counters to privilidged contexts
2 - restrict kernel-mode code counting and profiling
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Arnaldo Carvalho de Melo <acme@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 0c4bf86..344a659 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -916,9 +916,9 @@
#ifdef CONFIG_PERF_COUNTERS
{
.ctl_name = CTL_UNNUMBERED,
- .procname = "perf_counter_privileged",
- .data = &sysctl_perf_counter_priv,
- .maxlen = sizeof(sysctl_perf_counter_priv),
+ .procname = "perf_counter_paranoid",
+ .data = &sysctl_perf_counter_paranoid,
+ .maxlen = sizeof(sysctl_perf_counter_paranoid),
.mode = 0644,
.proc_handler = &proc_dointvec,
},