Re: [Use cpuops V1 09/11] Connector: Use this_cpu operations

[Tejun Heo]

On 12/06/2010 06:16 PM, Christoph Lameter wrote:
> get_seq can benefit from this_cpu_operations. Address calculation is avoided
> and the increment is done using an xadd.
> 
> Cc: Scott James Remnant <scott@xxxxxxxxxx>
> Cc: Mike Frysinger <vapier@xxxxxxxxxx>
> Signed-off-by: Christoph Lameter <cl@xxxxxxxxx>
> 
> ---
>  drivers/connector/cn_proc.c |    5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> Index: linux-2.6/drivers/connector/cn_proc.c
> ===================================================================
> --- linux-2.6.orig/drivers/connector/cn_proc.c	2010-11-30 09:38:33.000000000 -0600
> +++ linux-2.6/drivers/connector/cn_proc.c	2010-11-30 09:39:38.000000000 -0600
> @@ -43,9 +43,10 @@ static DEFINE_PER_CPU(__u32, proc_event_
>  
>  static inline void get_seq(__u32 *ts, int *cpu)
>  {
> -	*ts = get_cpu_var(proc_event_counts)++;
> +	preempt_disable();
> +	*ts = __this_cpu_inc(proc_event_counts);

Eh?  __this_cpu_inc() evaluates to the incremented value?

-- 
tejun
--
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/