Re: [PATCH 2/5] xen/spinlock: We don't need the old structure anymore

[Ramkumar Ramachandra]

Konrad Rzeszutek Wilk wrote:
> diff --git a/arch/x86/xen/spinlock.c b/arch/x86/xen/spinlock.c
> index 0438b93..71db82c 100644
> --- a/arch/x86/xen/spinlock.c
> +++ b/arch/x86/xen/spinlock.c
> @@ -81,7 +81,6 @@ static inline void spin_time_accum_blocked(u64 start)
>         spinlock_stats.time_blocked += delta;
>  }
>  #else  /* !CONFIG_XEN_DEBUG_FS */
> -#define TIMEOUT                        (1 << 10)

The timeout can be reduced, I think.

>  static inline void add_stats(enum xen_contention_stat var, u32 val)
>  {
>  }
> @@ -96,23 +95,6 @@ static inline void spin_time_accum_blocked(u64 start)
>  }
>  #endif  /* CONFIG_XEN_DEBUG_FS */
>
> -/*
> - * Size struct xen_spinlock so it's the same as arch_spinlock_t.
> - */
> -#if NR_CPUS < 256
> -typedef u8 xen_spinners_t;
> -# define inc_spinners(xl) \
> -       asm(LOCK_PREFIX " incb %0" : "+m" ((xl)->spinners) : : "memory");
> -# define dec_spinners(xl) \
> -       asm(LOCK_PREFIX " decb %0" : "+m" ((xl)->spinners) : : "memory");
> -#else
> -typedef u16 xen_spinners_t;
> -# define inc_spinners(xl) \
> -       asm(LOCK_PREFIX " incw %0" : "+m" ((xl)->spinners) : : "memory");
> -# define dec_spinners(xl) \
> -       asm(LOCK_PREFIX " decw %0" : "+m" ((xl)->spinners) : : "memory");
> -#endif
> -

Spinlocks on the filesystem help ensure consistency; otherwise, there
is a chance of a lot of noise coming through. Although NR_CPUS > 256,
very few CPUS are doing consistency checks.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/