[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Tue, 11 Nov 2008 12:10:39 +0900
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>
To: Balbir Singh <balbir@...ux.vnet.ibm.com>
Cc: linux-mm@...ck.org, YAMAMOTO Takashi <yamamoto@...inux.co.jp>,
Paul Menage <menage@...gle.com>, lizf@...fujitsu.com,
linux-kernel@...r.kernel.org,
Nick Piggin <nickpiggin@...oo.com.au>,
David Rientjes <rientjes@...gle.com>,
Pavel Emelianov <xemul@...nvz.org>,
Dhaval Giani <dhaval@...ux.vnet.ibm.com>,
Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [RFC][mm] [PATCH 4/4] Memory cgroup hierarchy feature selector
(v2)
On Sat, 08 Nov 2008 14:41:13 +0530
Balbir Singh <balbir@...ux.vnet.ibm.com> wrote:
>
> Don't enable multiple hierarchy support by default. This patch introduces
> a features element that can be set to enable the nested depth hierarchy
> feature. This feature can only be enabled when the cgroup for which the
> feature this is enabled, has no children.
>
> Signed-off-by: Balbir Singh <balbir@...ux.vnet.ibm.com>
> ---
>
IMHO, permission to this is not enough.
I think following is sane way.
==
When parent->use_hierarchy==1.
my->use_hierarchy must be "1" and cannot be tunrned to be "0" even if no children.
When parent->use_hierarchy==0
my->use_hierarchy can be either of "0" or "1".
this value can be chagned if we don't have children
==
Thanks,
-Kame
> mm/memcontrol.c | 43 ++++++++++++++++++++++++++++++++++++++++++-
> 1 file changed, 42 insertions(+), 1 deletion(-)
>
> diff -puN mm/memcontrol.c~memcg-add-hierarchy-selector mm/memcontrol.c
> --- linux-2.6.28-rc2/mm/memcontrol.c~memcg-add-hierarchy-selector 2008-11-08 14:09:33.000000000 +0530
> +++ linux-2.6.28-rc2-balbir/mm/memcontrol.c 2008-11-08 14:10:53.000000000 +0530
> @@ -137,6 +137,11 @@ struct mem_cgroup {
> * reclaimed from.
> */
> struct mem_cgroup *last_scanned_child;
> + /*
> + * Should the accounting and control be hierarchical, per subtree?
> + */
> + unsigned long use_hierarchy;
> +
> };
> static struct mem_cgroup init_mem_cgroup;
>
> @@ -1079,6 +1084,33 @@ out:
> return ret;
> }
>
> +static u64 mem_cgroup_hierarchy_read(struct cgroup *cont, struct cftype *cft)
> +{
> + return mem_cgroup_from_cont(cont)->use_hierarchy;
> +}
> +
> +static int mem_cgroup_hierarchy_write(struct cgroup *cont, struct cftype *cft,
> + u64 val)
> +{
> + int retval = 0;
> + struct mem_cgroup *mem = mem_cgroup_from_cont(cont);
> +
> + if (val == 1) {
> + if (list_empty(&cont->children))
> + mem->use_hierarchy = val;
> + else
> + retval = -EBUSY;
> + } else if (val == 0) {
> + if (list_empty(&cont->children))
> + mem->use_hierarchy = val;
> + else
> + retval = -EBUSY;
> + } else
> + retval = -EINVAL;
> +
> + return retval;
> +}
> +
> static u64 mem_cgroup_read(struct cgroup *cont, struct cftype *cft)
> {
> return res_counter_read_u64(&mem_cgroup_from_cont(cont)->res,
> @@ -1213,6 +1245,11 @@ static struct cftype mem_cgroup_files[]
> .name = "stat",
> .read_map = mem_control_stat_show,
> },
> + {
> + .name = "use_hierarchy",
> + .write_u64 = mem_cgroup_hierarchy_write,
> + .read_u64 = mem_cgroup_hierarchy_read,
> + },
> };
>
> static int alloc_mem_cgroup_per_zone_info(struct mem_cgroup *mem, int node)
> @@ -1289,9 +1326,13 @@ mem_cgroup_create(struct cgroup_subsys *
> parent = mem_cgroup_from_cont(cont->parent);
> if (!mem)
> return ERR_PTR(-ENOMEM);
> + mem->use_hierarchy = parent->use_hierarchy;
> }
>
> - res_counter_init(&mem->res, parent ? &parent->res : NULL);
> + if (parent && parent->use_hierarchy)
> + res_counter_init(&mem->res, &parent->res);
> + else
> + res_counter_init(&mem->res, NULL);
>
> for_each_node_state(node, N_POSSIBLE)
> if (alloc_mem_cgroup_per_zone_info(mem, node))
> _
>
> --
> Balbir
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists