Re: [PATCH 2/2] perf_counter: optimize for perf_event_init_task()

[Peter Zijlstra <peterz@infradead.org>]

On Wed, 2009-09-23 at 16:13 +0800, Xiao Guangrong wrote:
> We can traverse ctx->group_list to get all group leader, it should be safe
> since we hold ctx->mutex

I don't think we need that WARN_ON_ONCE there.

> Signed-off-by: Xiao Guangrong <xiaoguangrong@cn.fujitsu.com>
> ---
>  kernel/perf_event.c |    5 ++---
>  1 files changed, 2 insertions(+), 3 deletions(-)
> 
> diff --git a/kernel/perf_event.c b/kernel/perf_event.c
> index 9ca975a..4e6e822 100644
> --- a/kernel/perf_event.c
> +++ b/kernel/perf_event.c
> @@ -4786,9 +4786,8 @@ int perf_event_init_task(struct task_struct *child)
>  	 * We dont have to disable NMIs - we are only looking at
>  	 * the list, not manipulating it:
>  	 */
> -	list_for_each_entry_rcu(event, &parent_ctx->event_list, event_entry) {
> -		if (event != event->group_leader)
> -			continue;
> +	list_for_each_entry(event, &parent_ctx->group_list, group_entry) {
> +		WARN_ON_ONCE(event != event->group_leader);
>  
>  		if (!event->attr.inherit) {
>  			inherited_all = 0;