KAMEZAWA Hiroyuki wrote:
> spinlock is necessary when someone changes res_counter value.
> splited out from YAMAMOTO's background page reclaim for memory cgroup set.
> 
> Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
> From: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
> 
> 
>  kernel/res_counter.c |    5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> Index: linux-2.6.24-rc3-mm1/kernel/res_counter.c
> ===================================================================
> --- linux-2.6.24-rc3-mm1.orig/kernel/res_counter.c	2007-11-27 14:07:44.000000000 +0900
> +++ linux-2.6.24-rc3-mm1/kernel/res_counter.c	2007-11-27 14:09:40.000000000 +0900
> @@ -98,7 +98,7 @@
>  {
>  	int ret;
>  	char *buf, *end;
> -	unsigned long long tmp, *val;
> +	unsigned long long flags, tmp, *val;

Flags for irqsave should be unsigned long. No?

>  	buf = kmalloc(nbytes + 1, GFP_KERNEL);
>  	ret = -ENOMEM;
> @@ -121,9 +121,10 @@
>  		if (*end != '\0')
>  			goto out_free;
>  	}
> -
> +	spin_lock_irqsave(&counter->lock, flags);
>  	val = res_counter_member(counter, member);
>  	*val = tmp;
> +	spin_unlock_irqrestore(&counter->lock, flags);
>  	ret = nbytes;
>  out_free:
>  	kfree(buf);
> 
> 

_______________________________________________
Containers mailing list
Containers@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers