[xfs-masters] Re: [interesting] smattering of possible memory ordering bugs

[Dave Kleikamp <shaggy@xxxxxxxxxxxxxxxxxx>]

Memory barriers aren't one of my strengths, but this appears correct.

Acked-by: Dave Kleikamp <shaggy@xxxxxxxxxxxxxxxxxx>

On Fri, 2007-10-26 at 12:09 +1000, Nick Piggin wrote:
> Index: linux-2.6/fs/jfs/jfs_metapage.c
> ===================================================================
> --- linux-2.6.orig/fs/jfs/jfs_metapage.c
> +++ linux-2.6/fs/jfs/jfs_metapage.c
> @@ -39,11 +39,12 @@ static struct {
>  #endif
>  
>  #define metapage_locked(mp) test_bit(META_locked, &(mp)->flag)
> -#define trylock_metapage(mp) test_and_set_bit(META_locked,
> &(mp)->flag)
> +#define trylock_metapage(mp) test_and_set_bit_lock(META_locked,
> &(mp)->flag)
>  
>  static inline void unlock_metapage(struct metapage *mp)
>  {
> -       clear_bit(META_locked, &mp->flag);
> +       clear_bit_unlock(META_locked, &mp->flag);
> +       smp_mb__after_clear_bit();
>         wake_up(&mp->wait);
>  }
>   

Thanks,
Shaggy
-- 
David Kleikamp
IBM Linux Technology Center