oprofile, ARM: Release resources on failure
authorRobert Richter <robert.richter@amd.com>
Wed, 29 Sep 2010 14:52:25 +0000 (16:52 +0200)
committerIngo Molnar <mingo@elte.hu>
Thu, 30 Sep 2010 07:14:25 +0000 (09:14 +0200)
This patch fixes a resource leak on failure, where the
oprofilefs and some counters may not released properly.

Signed-off-by: Robert Richter <robert.richter@amd.com>
Acked-by: Will Deacon <will.deacon@arm.com>
Cc: linux-arm-kernel@lists.infradead.org
Cc: <stable@kernel.org> # .35.x
LKML-Reference: <20100929145225.GJ13563@erda.amd.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
arch/arm/oprofile/common.c

index 0691176..72e09eb 100644 (file)
@@ -102,6 +102,7 @@ static int op_create_counter(int cpu, int event)
        if (IS_ERR(pevent)) {
                ret = PTR_ERR(pevent);
        } else if (pevent->state != PERF_EVENT_STATE_ACTIVE) {
+               perf_event_release_kernel(pevent);
                pr_warning("oprofile: failed to enable event %d "
                                "on CPU %d\n", event, cpu);
                ret = -EBUSY;
@@ -365,6 +366,7 @@ int __init oprofile_arch_init(struct oprofile_operations *ops)
        ret = init_driverfs();
        if (ret) {
                kfree(counter_config);
+               counter_config = NULL;
                return ret;
        }
 
@@ -402,7 +404,6 @@ void oprofile_arch_exit(void)
        struct perf_event *event;
 
        if (*perf_events) {
-               exit_driverfs();
                for_each_possible_cpu(cpu) {
                        for (id = 0; id < perf_num_counters; ++id) {
                                event = perf_events[cpu][id];
@@ -413,8 +414,10 @@ void oprofile_arch_exit(void)
                }
        }
 
-       if (counter_config)
+       if (counter_config) {
                kfree(counter_config);
+               exit_driverfs();
+       }
 }
 #else
 int __init oprofile_arch_init(struct oprofile_operations *ops)