[v2,11/51] memcg: implement mem_cgroup_css_from_page()

Commit Message

Tejun Heo May 27, 2015, 4:13 p.m. UTC

From 26bab580abfc441c841c1983469b8b86f5a8ef5c Mon Sep 17 00:00:00 2001
From: Tejun Heo <[e-mail]>
Date: Wed, 27 May 2015 12:08:29 -0400

Implement mem_cgroup_css_from_page() which returns the
cgroup_subsys_state of the memcg associated with a given page.  This
will be used by cgroup writeback support.

This function assumes that page->mem_cgroup association doesn't change
until the page is released, which is true on the default hierarchy as
long as mem_cgroup_migrate() is not used.  As the only user of
mem_cgroup_migrate() is FUSE which won't support cgroup writeback for
the time being, this works for now, and mem_cgroup_migrate() will soon
be updated so that the invariant actually holds.

v2: Trigger WARN if the function is used on the traditional
    hierarchies and add comment about the assumed invariant.

Signed-off-by: Tejun Heo <[e-mail]>
Cc: Johannes Weiner <[e-mail]>
Cc: Michal Hocko <[e-mail]>
---
Added WARN and comments.  No functional change.  git branches updated
accordingly.

Thanks.

 include/linux/memcontrol.h |  1 +
 mm/memcontrol.c            | 31 +++++++++++++++++++++++++++++++
 2 files changed, 32 insertions(+)

Comments

Johannes Weiner May 27, 2015, 5:09 p.m. UTC | #1

On Wed, May 27, 2015 at 12:13:44PM -0400, Tejun Heo wrote:
> From 26bab580abfc441c841c1983469b8b86f5a8ef5c Mon Sep 17 00:00:00 2001
> From: Tejun Heo <[e-mail]>
> Date: Wed, 27 May 2015 12:08:29 -0400
> 
> Implement mem_cgroup_css_from_page() which returns the
> cgroup_subsys_state of the memcg associated with a given page.  This
> will be used by cgroup writeback support.
> 
> This function assumes that page->mem_cgroup association doesn't change
> until the page is released, which is true on the default hierarchy as
> long as mem_cgroup_migrate() is not used.  As the only user of
> mem_cgroup_migrate() is FUSE which won't support cgroup writeback for
> the time being, this works for now, and mem_cgroup_migrate() will soon
> be updated so that the invariant actually holds.

Regular page migration uses mem_cgroup_migrate() as well, but it's not
a problem as it ensures that the old page doesn't have any outstanding
references at that point.

It's only replace_page_cache_page() that calls mem_cgroup_migrate() on
a live page breaking mem_cgroup_css_from_page().

So the page looks fine, I'd just update the culprit function in the
changelog and kerneldoc.
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to [e-mail]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Tejun Heo May 27, 2015, 5:48 p.m. UTC | #2

Hello,

On Wed, May 27, 2015 at 01:09:55PM -0400, Johannes Weiner wrote:
> Regular page migration uses mem_cgroup_migrate() as well, but it's not
> a problem as it ensures that the old page doesn't have any outstanding
> references at that point.

Ooh, I see.

> It's only replace_page_cache_page() that calls mem_cgroup_migrate() on
> a live page breaking mem_cgroup_css_from_page().
> 
> So the page looks fine, I'd just update the culprit function in the
> changelog and kerneldoc.

Alright, will update the comment and description.

Thanks.

Patch

diff --git a/include/linux/memcontrol.h b/include/linux/memcontrol.h
index 294498f..637ef62 100644
--- a/include/linux/memcontrol.h
+++ b/include/linux/memcontrol.h
@@ -115,6 +115,7 @@  static inline bool mm_match_cgroup(struct mm_struct *mm,
 }
 
 extern struct cgroup_subsys_state *mem_cgroup_css(struct mem_cgroup *memcg);
+extern struct cgroup_subsys_state *mem_cgroup_css_from_page(struct page *page);
 
 struct mem_cgroup *mem_cgroup_iter(struct mem_cgroup *,
 				   struct mem_cgroup *,
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index b22a92b..c76b85c 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -598,6 +598,37 @@  struct cgroup_subsys_state *mem_cgroup_css(struct mem_cgroup *memcg)
 	return &memcg->css;
 }
 
+/**
+ * mem_cgroup_css_from_page - css of the memcg associated with a page
+ * @page: page of interest
+ *
+ * This function is guaranteed to return a valid cgroup_subsys_state and
+ * the returned css remains associated with @page until it is released.
+ *
+ * This can only be used on the default hierarchy as @page's memcg
+ * association may change on the traditional hierarchies.  Use
+ * try_get_mem_cgroup_from_page() instead on the traditional hierarchies.
+ *
+ * XXX: The above comment isn't true yet as mem_cgroup_migrate() can modify
+ *      the association before @page is released even on the default
+ *      hierarchy; however, the current and planned usages don't mix the
+ *      the two functions and mem_cgroup_migrate() will soon be updated to
+ *      make the invariant actually true.
+ */
+struct cgroup_subsys_state *mem_cgroup_css_from_page(struct page *page)
+{
+	struct cgroup_subsys_state *css;
+
+	if (page->mem_cgroup)
+		css = &page->mem_cgroup->css;
+	else
+		css = &root_mem_cgroup->css;
+
+	WARN_ON_ONCE(!cgroup_on_dfl(css->cgroup));
+
+	return css;
+}
+
 static struct mem_cgroup_per_zone *
 mem_cgroup_page_zoneinfo(struct mem_cgroup *memcg, struct page *page)
 {