diff options
| author | Lai Jiangshan <laijs@cn.fujitsu.com> | 2008-06-12 16:43:07 +0800 |
|---|---|---|
| committer | Ingo Molnar <mingo@elte.hu> | 2008-06-12 14:29:54 +0200 |
| commit | 7a232e0350940d2664f4de5cc3f0f443bae5062d (patch) | |
| tree | bb91a53faab7bf6e92df127f8bdb2225cf808925 /kernel | |
| parent | 2e084786f6fe052274f1dfa7c675fe4a02cacd6e (diff) | |
sched: 64-bit: fix arithmetics overflow
(overflow means weight >= 2^32 here, because inv_weigh = 2^32/weight)
A weight of a cfs_rq is the sum of weights of which entities
are queued on this cfs_rq, so it will overflow when there are
too many entities.
Although, overflow occurs very rarely, but it break fairness when
it occurs. 64-bits systems have more memory than 32-bit systems
and 64-bit systems can create more process usually, so overflow may
occur more frequently.
This patch guarantees fairness when overflow happens on 64-bit systems.
Thanks to the optimization of compiler, it changes nothing on 32-bit.
Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Diffstat (limited to 'kernel')
| -rw-r--r-- | kernel/sched.c | 9 |
1 files changed, 7 insertions, 2 deletions
diff --git a/kernel/sched.c b/kernel/sched.c index 6c1ecbdc0db9..eaf6751e7612 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1340,8 +1340,13 @@ calc_delta_mine(unsigned long delta_exec, unsigned long weight, { u64 tmp; - if (!lw->inv_weight) - lw->inv_weight = 1 + (WMULT_CONST-lw->weight/2)/(lw->weight+1); + if (!lw->inv_weight) { + if (BITS_PER_LONG > 32 && unlikely(lw->weight >= WMULT_CONST)) + lw->inv_weight = 1; + else + lw->inv_weight = 1 + (WMULT_CONST-lw->weight/2) + / (lw->weight+1); + } tmp = (u64)delta_exec * weight; /* |