On Tue, Jan 15, 2013 at 03:46:17PM +0000, Christoph Lameter wrote:
> On Tue, 15 Jan 2013, Joonsoo Kim wrote:
>
> > There is a subtle bug when calculating a number of acquired objects.
> > After acquire_slab() is executed at first, page->inuse is same as
> > page->objects, then, available is always 0. So, we always go next
> > iteration.
>
> page->inuse is always < page->objects because the partial list is not used
> for slabs that are fully allocated. page->inuse == page->objects means
> that no objects are available on the slab and therefore the slab would
> have been removed from the partial list.
Currently, we calculate "available = page->objects - page->inuse",
after acquire_slab() is called in get_partial_node().
In acquire_slab() with mode = 1, we always set new.inuse = page->objects.
So
acquire_slab(s, n, page, object == NULL);
if (!object) {
c->page = page;
stat(s, ALLOC_FROM_PARTIAL);
object = t;
available = page->objects - page->inuse;
!!!!!! available is always 0 !!!!!!
} else {
available = put_cpu_partial(s, page, 0);
stat(s, CPU_PARTIAL_NODE);
}
Therefore, "available > s->cpu_partial / 2" is always false and
we always go to second iteration.
This patch correct this problem.
> > After that, we don't need return value of put_cpu_partial().
> > So remove it.
>
> Hmmm... The code looks a bit easier to understand than what we have right now.
>
> Could you try to explain it better?
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@xxxxxxxxx. For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>