Re: [RFC 3/3] irq: Privatize irq_common_data::state_use_accessors

[Peter Zijlstra <peterz@infradead.org>]

On Sat, Dec 12, 2015 at 10:56:02AM +0800, Boqun Feng wrote:
> According to Peter Zijlstra, irq_common_data::state_use_accessors is not
> designed for public use. Therefore make it private so that people who
> write code accessing it directly will get blamed by sparse.
> 
> Signed-off-by: Boqun Feng <boqun.feng@gmail.com>
> ---
>  include/linux/irq.h | 4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/include/linux/irq.h b/include/linux/irq.h
> index 3c1c967..0b8f273 100644
> --- a/include/linux/irq.h
> +++ b/include/linux/irq.h
> @@ -137,7 +137,7 @@ struct irq_domain;
>   * @msi_desc:		MSI descriptor
>   */
>  struct irq_common_data {
> -	unsigned int		state_use_accessors;
> +	unsigned int		__private state_use_accessors;
>  #ifdef CONFIG_NUMA
>  	unsigned int		node;
>  #endif
> @@ -208,7 +208,7 @@ enum {
>  	IRQD_FORWARDED_TO_VCPU		= (1 << 20),
>  };
>  
> -#define __irqd_to_state(d)		((d)->common->state_use_accessors)
> +#define __irqd_to_state(d) ACCESS_PRIVATE((d)->common, state_use_accessors)
>  
>  static inline bool irqd_is_setaffinity_pending(struct irq_data *d)
>  {

We should probably #undef that one once we're done with it.. Thomas?