Re: [PATCH v5] mm,writeback: fix divide by zero in pos_ratio_polynom

[Michal Hocko <mhocko@suse.cz>]

On Wed 30-04-14 16:42:55, Rik van Riel wrote:
> On Wed, 30 Apr 2014 13:13:53 -0700
> Andrew Morton <akpm@linux-foundation.org> wrote:
> 
> > This was a consequence of 64->32 truncation and it can't happen any
> > more, can it?
> 
> Andrew, this is cleaner indeed :)
> 
> Masayoshi-san, does the bug still happen with this version, or does
> this fix the problem?
> 
> ---8<---
> 
> Subject: mm,writeback: fix divide by zero in pos_ratio_polynom
> 
> It is possible for "limit - setpoint + 1" to equal zero, after
> getting truncated to a 32 bit variable, and resulting in a divide
> by zero error.
> 
> Using the fully 64 bit divide functions avoids this problem.
> 
> Also uninline pos_ratio_polynom, at Andrew's request.
> 
> Signed-off-by: Rik van Riel <riel@redhat.com>

This looks much better.
Reviewed-by: Michal Hocko <mhocko@suse.cz>

> ---
>  mm/page-writeback.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
> 
> diff --git a/mm/page-writeback.c b/mm/page-writeback.c
> index ef41349..a4317da 100644
> --- a/mm/page-writeback.c
> +++ b/mm/page-writeback.c
> @@ -593,14 +593,14 @@ unsigned long bdi_dirty_limit(struct backing_dev_info *bdi, unsigned long dirty)
>   * (5) the closer to setpoint, the smaller |df/dx| (and the reverse)
>   *     => fast response on large errors; small oscillation near setpoint
>   */
> -static inline long long pos_ratio_polynom(unsigned long setpoint,
> +static long long pos_ratio_polynom(unsigned long setpoint,
>  					  unsigned long dirty,
>  					  unsigned long limit)
>  {
>  	long long pos_ratio;
>  	long x;
>  
> -	x = div_s64(((s64)setpoint - (s64)dirty) << RATELIMIT_CALC_SHIFT,
> +	x = div64_s64(((s64)setpoint - (s64)dirty) << RATELIMIT_CALC_SHIFT,
>  		    limit - setpoint + 1);
>  	pos_ratio = x;
>  	pos_ratio = pos_ratio * x >> RATELIMIT_CALC_SHIFT;
> @@ -842,7 +842,7 @@ static unsigned long bdi_position_ratio(struct backing_dev_info *bdi,
>  	x_intercept = bdi_setpoint + span;
>  
>  	if (bdi_dirty < x_intercept - span / 4) {
> -		pos_ratio = div_u64(pos_ratio * (x_intercept - bdi_dirty),
> +		pos_ratio = div64_u64(pos_ratio * (x_intercept - bdi_dirty),
>  				    x_intercept - bdi_setpoint + 1);
>  	} else
>  		pos_ratio /= 4;
> 

-- 
Michal Hocko
SUSE Labs