Re: [PATCH v3 08/15] perf: Use guard() for aux_mutex in perf_mmap()

[Lorenzo Stoakes <lorenzo.stoakes@oracle.com>]

On Tue, Aug 12, 2025 at 12:39:06PM +0200, Peter Zijlstra wrote:
> After duplicating the common code into the rb/aux branches is it
> possible to use a simple guard() for the aux_mutex. Making the aux
> branch self-contained.
>
> Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>

LGTM, I really am going to look into how we can use guards in mm to
simplify things... :)

So:

Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>

> ---
>  kernel/events/core.c |    6 +-----
>  1 file changed, 1 insertion(+), 5 deletions(-)
>
> --- a/kernel/events/core.c
> +++ b/kernel/events/core.c
> @@ -6975,7 +6975,6 @@ static int perf_mmap(struct file *file,
>  	struct perf_event *event = file->private_data;
>  	unsigned long vma_size, nr_pages;
>  	long user_extra = 0, extra = 0;
> -	struct mutex *aux_mutex = NULL;
>  	struct perf_buffer *rb = NULL;
>  	int ret, flags = 0;
>  	mapped_f mapped;
> @@ -7098,8 +7097,7 @@ static int perf_mmap(struct file *file,
>  		if (!rb)
>  			goto unlock;
>
> -		aux_mutex = &rb->aux_mutex;
> -		mutex_lock(aux_mutex);
> +		guard(mutex)(&rb->aux_mutex);
>
>  		aux_offset = READ_ONCE(rb->user_page->aux_offset);
>  		aux_size = READ_ONCE(rb->user_page->aux_size);
> @@ -7159,8 +7157,6 @@ static int perf_mmap(struct file *file,
>  	}
>
>  unlock:
> -	if (aux_mutex)
> -		mutex_unlock(aux_mutex);
>  	mutex_unlock(&event->mmap_mutex);
>
>  	if (ret)
>
>