When pcpu_build_alloc_info() searches best_upa value, it ignores current value
if the number of waste units exceeds 1/3 of the number of total cpus. But the
comment on the code says that it will ignore if wastage is over 25%.
Modify the comment.
Signed-off-by: Namhyung Kim <[email protected]>
---
I used -U5 rather than 3 for you to see the mismatch at a glance. :-)
mm/percpu.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/mm/percpu.c b/mm/percpu.c
index 9581f84..5e78131 100644
--- a/mm/percpu.c
+++ b/mm/percpu.c
@@ -1174,11 +1174,11 @@ struct pcpu_alloc_info * __init pcpu_build_alloc_info(
allocs += this_allocs;
wasted += this_allocs * upa - group_cnt[group];
}
/*
- * Don't accept if wastage is over 25%. The
+ * Don't accept if wastage is over 1/3. The
* greater-than comparison ensures upa==1 always
* passes the following check.
*/
if (wasted > num_possible_cpus() / 3)
continue;
--
1.7.0.4
On 08/11/2010 04:19 AM, Namhyung Kim wrote:
> When pcpu_build_alloc_info() searches best_upa value, it ignores current value
> if the number of waste units exceeds 1/3 of the number of total cpus. But the
> comment on the code says that it will ignore if wastage is over 25%.
> Modify the comment.
>
> Signed-off-by: Namhyung Kim <[email protected]>
Applied to percpu#for-linus. Thanks.
--
tejun