On Tue, Apr 22, 2014 at 6:34 PM, Michal Hocko <mhocko@xxxxxxx> wrote:
> Sorry, I should have been more specific that I was talking about
> mem_cgroup_reclaim_cookie path where the iteration for this particular
> zone and priority ended at the last node without finishing the full
> roundtrip last time. This new iteration (prev==NULL) wants to continue
> and it should start a new roundtrip.
>
> Makes sense?
Hi, Michal,
Good catch, it makes sense !
This reminds me of my draft edition of this patch, I specifically handle
this case as:
if (reclaim) {
if (!memcg ) {
iter->generation++;
if (!prev) {
memcg = root;
mem_cgroup_iter_update(iter, NULL,
memcg, root, seq);
goto out_unlock:
}
}
mem_cgroup_iter_update(iter, last_visited, memcg, root,
seq);
if (!prev && memcg)
reclaim->generation = iter->generation;
}
This is literally manual unwinding the second while loop, and thus omit
the while loop,
to save a mem_cgroup_iter_update() and a mem_cgroup_iter_update()
But it maybe a bit hard to read.
If it is OK, I could resend a new one.
Thanks,
Jianyu Zhan
--
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>