The patch titled
Subject: memcg: split part of memcg creation to css_online
has been added to the -mm tree. Its filename is
memcg-split-part-of-memcg-creation-to-css_online.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Glauber Costa <glommer@xxxxxxxxxxxxx>
Subject: memcg: split part of memcg creation to css_online
This patch is a preparatory work for later locking rework to get rid of
big cgroup lock from memory controller code.
The memory controller uses some tunables to adjust its operation. Those
tunables are inherited from parent to children upon children
intialization. For most of them, the value cannot be changed after the
parent has a new children.
cgroup core splits initialization in two phases: css_alloc and css_online.
After css_alloc, the memory allocation and basic initialization are done.
But the new group is not yet visible anywhere, not even for cgroup core
code. It is only somewhere between css_alloc and css_online that it is
inserted into the internal children lists. Copying tunable values in
css_alloc will lead to inconsistent values: the children will copy the old
parent values, that can change between the copy and the moment in which
the groups is linked to any data structure that can indicate the presence
of children.
Signed-off-by: Glauber Costa <glommer@xxxxxxxxxxxxx>
Acked-by: Michal Hocko <mhocko@xxxxxxx>
Cc: Tejun Heo <tj@xxxxxxxxxx>
Cc: Hiroyuki Kamezawa <kamezawa.hiroyuki@xxxxxxxxx>
Cc: Johannes Weiner <hannes@xxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/memcontrol.c | 62 +++++++++++++++++++++++++++++-----------------
1 file changed, 40 insertions(+), 22 deletions(-)
diff -puN mm/memcontrol.c~memcg-split-part-of-memcg-creation-to-css_online mm/memcontrol.c
--- a/mm/memcontrol.c~memcg-split-part-of-memcg-creation-to-css_online
+++ a/mm/memcontrol.c
@@ -6091,7 +6091,7 @@ err_cleanup:
static struct cgroup_subsys_state * __ref
mem_cgroup_css_alloc(struct cgroup *cont)
{
- struct mem_cgroup *memcg, *parent;
+ struct mem_cgroup *memcg;
long error = -ENOMEM;
int node;
@@ -6106,7 +6106,7 @@ mem_cgroup_css_alloc(struct cgroup *cont
/* root ? */
if (cont->parent == NULL) {
int cpu;
- parent = NULL;
+
if (mem_cgroup_soft_limit_tree_init())
goto free_out;
root_mem_cgroup = memcg;
@@ -6115,13 +6115,43 @@ mem_cgroup_css_alloc(struct cgroup *cont
&per_cpu(memcg_stock, cpu);
INIT_WORK(&stock->work, drain_local_stock);
}
- } else {
- parent = mem_cgroup_from_cont(cont->parent);
- memcg->use_hierarchy = parent->use_hierarchy;
- memcg->oom_kill_disable = parent->oom_kill_disable;
+
+ res_counter_init(&memcg->res, NULL);
+ res_counter_init(&memcg->memsw, NULL);
+ res_counter_init(&memcg->kmem, NULL);
}
- if (parent && parent->use_hierarchy) {
+ memcg->last_scanned_node = MAX_NUMNODES;
+ INIT_LIST_HEAD(&memcg->oom_notify);
+ atomic_set(&memcg->refcnt, 1);
+ memcg->move_charge_at_immigrate = 0;
+ mutex_init(&memcg->thresholds_lock);
+ spin_lock_init(&memcg->move_lock);
+
+ return &memcg->css;
+
+free_out:
+ __mem_cgroup_free(memcg);
+ return ERR_PTR(error);
+}
+
+static int
+mem_cgroup_css_online(struct cgroup *cont)
+{
+ struct mem_cgroup *memcg, *parent;
+ int error = 0;
+
+ if (!cont->parent)
+ return 0;
+
+ memcg = mem_cgroup_from_cont(cont);
+ parent = mem_cgroup_from_cont(cont->parent);
+
+ memcg->use_hierarchy = parent->use_hierarchy;
+ memcg->oom_kill_disable = parent->oom_kill_disable;
+ memcg->swappiness = mem_cgroup_swappiness(parent);
+
+ if (parent->use_hierarchy) {
res_counter_init(&memcg->res, &parent->res);
res_counter_init(&memcg->memsw, &parent->memsw);
res_counter_init(&memcg->kmem, &parent->kmem);
@@ -6142,18 +6172,9 @@ mem_cgroup_css_alloc(struct cgroup *cont
* much sense so let cgroup subsystem know about this
* unfortunate state in our controller.
*/
- if (parent && parent != root_mem_cgroup)
+ if (parent != root_mem_cgroup)
mem_cgroup_subsys.broken_hierarchy = true;
}
- memcg->last_scanned_node = MAX_NUMNODES;
- INIT_LIST_HEAD(&memcg->oom_notify);
-
- if (parent)
- memcg->swappiness = mem_cgroup_swappiness(parent);
- atomic_set(&memcg->refcnt, 1);
- memcg->move_charge_at_immigrate = 0;
- mutex_init(&memcg->thresholds_lock);
- spin_lock_init(&memcg->move_lock);
error = memcg_init_kmem(memcg, &mem_cgroup_subsys);
if (error) {
@@ -6163,12 +6184,8 @@ mem_cgroup_css_alloc(struct cgroup *cont
* call __mem_cgroup_free, so return directly
*/
mem_cgroup_put(memcg);
- return ERR_PTR(error);
}
- return &memcg->css;
-free_out:
- __mem_cgroup_free(memcg);
- return ERR_PTR(error);
+ return error;
}
static void mem_cgroup_css_offline(struct cgroup *cont)
@@ -6778,6 +6795,7 @@ struct cgroup_subsys mem_cgroup_subsys =
.name = "memory",
.subsys_id = mem_cgroup_subsys_id,
.css_alloc = mem_cgroup_css_alloc,
+ .css_online = mem_cgroup_css_online,
.css_offline = mem_cgroup_css_offline,
.css_free = mem_cgroup_css_free,
.can_attach = mem_cgroup_can_attach,
_
Patches currently in -mm which might be from glommer@xxxxxxxxxxxxx are
memcg-fix-typo-in-kmemcg-cache-walk-macro.patch
memcgvmscan-do-not-break-out-targeted-reclaim-without-reclaimed-pages.patch
memcg-reduce-the-size-of-struct-memcg-244-fold.patch
memcg-reduce-the-size-of-struct-memcg-244-fold-fix.patch
memcg-prevent-changes-to-move_charge_at_immigrate-during-task-attach.patch
memcg-split-part-of-memcg-creation-to-css_online.patch
memcg-fast-hierarchy-aware-child-test.patch
memcg-fast-hierarchy-aware-child-test-fix.patch
memcg-replace-cgroup_lock-with-memcg-specific-memcg_lock.patch
memcg-increment-static-branch-right-after-limit-set.patch
memcg-avoid-dangling-reference-count-in-creation-failure.patch
memcg-debugging-facility-to-access-dangling-memcgs.patch
memcg-debugging-facility-to-access-dangling-memcgs-fix.patch
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
