Messages in this thread | | | Date | Fri, 2 May 2014 11:16:41 +0200 | From | Michal Hocko <> | Subject | Re: [PATCH v5] mm,writeback: fix divide by zero in pos_ratio_polynom |
| |
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
| |