Techwizz/kernel-fxtec-pro1x
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

perfcounters: release CPU context when exiting task counters

Browse source 
If counters are exiting via do_exit() not via filp close, then
the CPU context needs to be released - otherwise future percpu
counter creations might fail.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
This commit is contained in:
Ingo Molnar 2008-12-14 23:20:36 +01:00
parent 029af8c753
commit 0cc0c027d4
1 changed files with 12 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

14
kernel/perf_counter.c
View file

@ -1273,8 +1273,19 @@ __perf_counter_exit_task(struct task_struct *child,
local_irq_disable();
perf_flags = hw_perf_save_disable();
if (child_counter->state == PERF_COUNTER_STATE_ACTIVE)
if (child_counter->state == PERF_COUNTER_STATE_ACTIVE) {
struct perf_cpu_context *cpuctx;
cpuctx = &__get_cpu_var(perf_cpu_context);
child_counter->hw_ops->hw_perf_counter_disable(child_counter);
child_counter->state = PERF_COUNTER_STATE_INACTIVE;
child_counter->oncpu = -1;
cpuctx->active_oncpu--;
child_ctx->nr_active--;
}
list_del_init(&child_counter->list_entry);
hw_perf_restore(perf_flags);
@ -1539,4 +1550,3 @@ static int __init perf_counter_sysfs_init(void)
&perfclass_attr_group);
}
device_initcall(perf_counter_sysfs_init);