[PATCH] lowmemorykiller: prevent multiple instances of low memory killer

[Dan Carpenter]

On Mon, Apr 15, 2013 at 03:03:29PM +0200, Oskar Andero wrote:
> From: Snild Dolkow <snild.dolkow at sonymobile.com>
> 
> Running multiple instances of LMK is not useful since it will try to
> kill the same process.
> 
> This patch adds a spinlock to prevent multiple instances of the LMK
> running at the same time. Uses spin_trylock and return on failure to
> avoid blocking.
> 
> Cc: Greg Kroah-Hartman <gregkh at linuxfoundation.org>
> Cc: Brian Swetland <swetland at google.com>
> Reviewed-by: Radovan Lekanovic <radovan.lekanovic at sonymobile.com>
> Signed-off-by: Snild Dolkow <snild.dolkow at sonymobile.com>
> Signed-off-by: Oskar Andero <oskar.andero at sonymobile.com>
> ---
>  drivers/staging/android/lowmemorykiller.c | 9 +++++++++
>  1 file changed, 9 insertions(+)
> 
> diff --git a/drivers/staging/android/lowmemorykiller.c b/drivers/staging/android/lowmemorykiller.c
> index 3b91b0f..0b19353 100644
> --- a/drivers/staging/android/lowmemorykiller.c
> +++ b/drivers/staging/android/lowmemorykiller.c
> @@ -38,6 +38,7 @@
>  #include <linux/rcupdate.h>
>  #include <linux/profile.h>
>  #include <linux/notifier.h>
> +#include <linux/spinlock.h>
>  
>  static uint32_t lowmem_debug_level = 2;
>  static short lowmem_adj[6] = {
> @@ -57,6 +58,8 @@ static int lowmem_minfree_size = 4;
>  
>  static unsigned long lowmem_deathpending_timeout;
>  
> +#define LMK_BUSY (-1)

Where is lowmem_shrink called from?  I only see shrink called from
the bcache sysfs handler __bch_cache_set().  The return value isn't
checked there.

Up to now this function has only returns positive numbers.

There isn't a place which check LMK_BUSY so maybe it's best to just
return zero?

regards,
dan carpenter