lkml.org 
[lkml]   [2010]   [Nov]   [27]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [thisops uV2 06/10] vmstat: Use this_cpu_inc_return for vm statistics
From
On Fri, Nov 26, 2010 at 11:09 PM, Christoph Lameter <cl@linux.com> wrote:
> this_cpu_inc_return() saves us a memory access there. Code
> size does not change.
>
> V1->V2:
>        - Fixed the location of the __per_cpu pointer attributes
>        - Sparse checked
>
> Signed-off-by: Christoph Lameter <cl@linux.com>
>
> ---
>  mm/vmstat.c |   20 ++++++++------------
>  1 file changed, 8 insertions(+), 12 deletions(-)
>
> Index: linux-2.6/mm/vmstat.c
> ===================================================================
> --- linux-2.6.orig/mm/vmstat.c  2010-11-26 10:35:45.000000000 -0600
> +++ linux-2.6/mm/vmstat.c       2010-11-26 10:45:49.000000000 -0600
> @@ -167,8 +167,8 @@ static void refresh_zone_stat_thresholds
>  void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
>                                int delta)
>  {
> -       struct per_cpu_pageset * __percpu pcp = zone->pageset;
> -       s8 * __percpu p = pcp->vm_stat_diff + item;
> +       struct per_cpu_pageset __percpu *pcp = zone->pageset;
> +       s8 __percpu *p = pcp->vm_stat_diff + item;
>        long x;
>        long t;
>

Shouldn't this hunk be in the other vmstat patch?

> @@ -223,13 +223,11 @@ EXPORT_SYMBOL(mod_zone_page_state);
>  */
>  void __inc_zone_state(struct zone *zone, enum zone_stat_item item)
>  {
> -       struct per_cpu_pageset * __percpu pcp = zone->pageset;
> -       s8 * __percpu p = pcp->vm_stat_diff + item;
> +       struct per_cpu_pageset __percpu *pcp = zone->pageset;
> +       s8 __percpu *p = pcp->vm_stat_diff + item;
>        s8 v, t;
>
> -       __this_cpu_inc(*p);
> -
> -       v = __this_cpu_read(*p);
> +       v = __this_cpu_inc_return(*p);
>        t = __this_cpu_read(pcp->stat_threshold);
>        if (unlikely(v > t)) {
>                s8 overstep = t >> 1;
> @@ -247,13 +245,11 @@ EXPORT_SYMBOL(__inc_zone_page_state);
>
>  void __dec_zone_state(struct zone *zone, enum zone_stat_item item)
>  {
> -       struct per_cpu_pageset * __percpu pcp = zone->pageset;
> -       s8 * __percpu p = pcp->vm_stat_diff + item;
> +       struct per_cpu_pageset __percpu *pcp = zone->pageset;
> +       s8 __percpu *p = pcp->vm_stat_diff + item;
>        s8 v, t;
>
> -       __this_cpu_dec(*p);
> -
> -       v = __this_cpu_read(*p);
> +       v = __this_cpu_dec_return(*p);
>        t = __this_cpu_read(pcp->stat_threshold);
>        if (unlikely(v < - t)) {
>                s8 overstep = t >> 1;

For the rest of this patch:

Reviewed-by: Pekka Enberg <penberg@kernel.org>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2010-11-27 09:11    [W:0.125 / U:0.936 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site