[PATCH] memcg: support hierarchical memory.numa_stats

[PATCH] memcg: support hierarchical memory.numa_stats

[Ying Han]

The memory.numa_stat is not currently hierarchical. Memory charged to the
children are not shown in parent's numa_stat.

This change adds the "hierarchical_" stats on top of all existing stats, and
it includes the sum of all children's values in addition to the value of
the memcg.

Tested: Create cgroup a, a/b and run workload under b. The values of b are
included in the "hierarchical_*" under a.

$ cat /dev/cgroup/memory/a/memory.numa_stat
total=0 N0=0 N1=0
file=0 N0=0 N1=0
anon=0 N0=0 N1=0
unevictable=0 N0=0 N1=0
hierarchical_total=262474 N0=262162 N1=312
hierarchical_file=247 N0=0 N1=247
hierarchical_anon=262227 N0=262162 N1=65
hierarchical_unevictable=0 N0=0 N1=0

$ cat /dev/cgroup/memory/a/b/memory.numa_stat
total=262474 N0=262162 N1=312
file=247 N0=0 N1=247
anon=262227 N0=262162 N1=65
unevictable=0 N0=0 N1=0
hierarchical_total=262474 N0=262162 N1=312
hierarchical_file=247 N0=0 N1=247
hierarchical_anon=262227 N0=262162 N1=65
hierarchical_unevictable=0 N0=0 N1=0

Signed-off-by: Ying Han <yinghan@google.com>
---
 Documentation/cgroups/memory.txt |  5 +++-
 mm/memcontrol.c                  | 65 ++++++++++++++++++++++++++++++++++++++++
 2 files changed, 69 insertions(+), 1 deletion(-)

diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt
index 8b8c28b..b519e74 100644
--- a/Documentation/cgroups/memory.txt
+++ b/Documentation/cgroups/memory.txt
@@ -568,7 +568,10 @@ node.  One of the use cases is evaluating application performance by
 combining this information with the application's CPU allocation.
 
 We export "total", "file", "anon" and "unevictable" pages per-node for
-each memcg.  The ouput format of memory.numa_stat is:
+each memcg and "hierarchical_" for sum of all hierarchical children's values
+in addition to the memcg's own value.
+
+The ouput format of memory.numa_stat is:
 
 total=<total pages> N0=<node 0 pages> N1=<node 1 pages> ...
 file=<total file pages> N0=<node 0 pages> N1=<node 1 pages> ...
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 2b55222..9d8cf25 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1177,6 +1177,32 @@ void mem_cgroup_iter_break(struct mem_cgroup *root,
 	     iter != NULL;				\
 	     iter = mem_cgroup_iter(NULL, iter, NULL))
 
+static unsigned long
+mem_cgroup_node_hierarchical_nr_lru_pages(struct mem_cgroup *memcg,
+				int nid, unsigned int lru_mask)
+{
+	u64 total = 0;
+	struct mem_cgroup *iter;
+
+	for_each_mem_cgroup_tree(iter, memcg)
+		total += mem_cgroup_node_nr_lru_pages(iter, nid, lru_mask);
+
+	return total;
+}
+
+static unsigned long
+mem_cgroup_hierarchical_nr_lru_pages(struct mem_cgroup *memcg,
+					unsigned int lru_mask)
+{
+	u64 total = 0;
+	struct mem_cgroup *iter;
+
+	for_each_mem_cgroup_tree(iter, memcg)
+	total += mem_cgroup_nr_lru_pages(iter, lru_mask);
+
+	return total;
+}
+
 void __mem_cgroup_count_vm_event(struct mm_struct *mm, enum vm_event_item idx)
 {
 	struct mem_cgroup *memcg;
@@ -5267,6 +5293,45 @@ static int memcg_numa_stat_show(struct cgroup *cont, struct cftype *cft,
 		seq_printf(m, " N%d=%lu", nid, node_nr);
 	}
 	seq_putc(m, '\n');
+
+	total_nr = mem_cgroup_hierarchical_nr_lru_pages(memcg, LRU_ALL);
+	seq_printf(m, "hierarchical_total=%lu", total_nr);
+	for_each_node_state(nid, N_HIGH_MEMORY) {
+		node_nr =
+			mem_cgroup_node_hierarchical_nr_lru_pages(memcg, nid,
+								LRU_ALL);
+		seq_printf(m, " N%d=%lu", nid, node_nr);
+	}
+	seq_putc(m, '\n');
+
+	file_nr = mem_cgroup_hierarchical_nr_lru_pages(memcg, LRU_ALL_FILE);
+	seq_printf(m, "hierarchical_file=%lu", file_nr);
+	for_each_node_state(nid, N_HIGH_MEMORY) {
+		node_nr = mem_cgroup_node_hierarchical_nr_lru_pages(memcg, nid,
+				LRU_ALL_FILE);
+		seq_printf(m, " N%d=%lu", nid, node_nr);
+	}
+	seq_putc(m, '\n');
+
+	anon_nr = mem_cgroup_hierarchical_nr_lru_pages(memcg, LRU_ALL_ANON);
+	seq_printf(m, "hierarchical_anon=%lu", anon_nr);
+	for_each_node_state(nid, N_HIGH_MEMORY) {
+		node_nr = mem_cgroup_node_hierarchical_nr_lru_pages(memcg, nid,
+				LRU_ALL_ANON);
+		seq_printf(m, " N%d=%lu", nid, node_nr);
+	}
+	seq_putc(m, '\n');
+
+	unevictable_nr = mem_cgroup_hierarchical_nr_lru_pages(memcg,
+						BIT(LRU_UNEVICTABLE));
+	seq_printf(m, "hierarchical_unevictable=%lu", unevictable_nr);
+	for_each_node_state(nid, N_HIGH_MEMORY) {
+		node_nr = mem_cgroup_node_hierarchical_nr_lru_pages(memcg, nid,
+				BIT(LRU_UNEVICTABLE));
+		seq_printf(m, " N%d=%lu", nid, node_nr);
+	}
+	seq_putc(m, '\n');
+
 	return 0;
 }
 #endif /* CONFIG_NUMA */
-- 
1.8.1.3

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

Re: [PATCH] memcg: support hierarchical memory.numa_stats

[Michal Hocko]

I am sorry but I didn't get to this sooner.

On Mon 08-04-13 14:58:46, Ying Han wrote:
> The memory.numa_stat is not currently hierarchical. Memory charged to the
> children are not shown in parent's numa_stat.
> 
> This change adds the "hierarchical_" stats on top of all existing stats, and
> it includes the sum of all children's values in addition to the value of
> the memcg.

OK, I guess it makes some sense to be consistent with what we have in
memory.stat file. We are using total_ prefix there for most things
though (except for the limit which uses hierarchical). I am not sure
total_total sounds that great... So maybe hierarchical_ wouldn't be that
bad in the end.

Few comments bellow but other than that
Acked-by: Michal Hocko <mhocko@suse.cz>

[...]
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -1177,6 +1177,32 @@ void mem_cgroup_iter_break(struct mem_cgroup *root,
>  	     iter != NULL;				\
>  	     iter = mem_cgroup_iter(NULL, iter, NULL))
>  
> +static unsigned long
> +mem_cgroup_node_hierarchical_nr_lru_pages(struct mem_cgroup *memcg,
> +				int nid, unsigned int lru_mask)
> +{
> +	u64 total = 0;
> +	struct mem_cgroup *iter;
> +
> +	for_each_mem_cgroup_tree(iter, memcg)
> +		total += mem_cgroup_node_nr_lru_pages(iter, nid, lru_mask);
> +
> +	return total;
> +}
> +
> +static unsigned long
> +mem_cgroup_hierarchical_nr_lru_pages(struct mem_cgroup *memcg,
> +					unsigned int lru_mask)
> +{
> +	u64 total = 0;
> +	struct mem_cgroup *iter;
> +
> +	for_each_mem_cgroup_tree(iter, memcg)
> +	total += mem_cgroup_nr_lru_pages(iter, lru_mask);

Indentation

> +
> +	return total;
> +}
> +

These do not need to be defined for !CONFIG_NUMA. I do not think this is
generally usable functionality. Just move it memcg_numa_stat_show

>  void __mem_cgroup_count_vm_event(struct mm_struct *mm, enum vm_event_item idx)
>  {
>  	struct mem_cgroup *memcg;
> @@ -5267,6 +5293,45 @@ static int memcg_numa_stat_show(struct cgroup *cont, struct cftype *cft,
>  		seq_printf(m, " N%d=%lu", nid, node_nr);
>  	}
>  	seq_putc(m, '\n');
> +
> +	total_nr = mem_cgroup_hierarchical_nr_lru_pages(memcg, LRU_ALL);
> +	seq_printf(m, "hierarchical_total=%lu", total_nr);
> +	for_each_node_state(nid, N_HIGH_MEMORY) {
> +		node_nr =
> +			mem_cgroup_node_hierarchical_nr_lru_pages(memcg, nid,
> +								LRU_ALL);
> +		seq_printf(m, " N%d=%lu", nid, node_nr);
> +	}
> +	seq_putc(m, '\n');
> +
> +	file_nr = mem_cgroup_hierarchical_nr_lru_pages(memcg, LRU_ALL_FILE);
> +	seq_printf(m, "hierarchical_file=%lu", file_nr);
> +	for_each_node_state(nid, N_HIGH_MEMORY) {
> +		node_nr = mem_cgroup_node_hierarchical_nr_lru_pages(memcg, nid,
> +				LRU_ALL_FILE);
> +		seq_printf(m, " N%d=%lu", nid, node_nr);
> +	}
> +	seq_putc(m, '\n');
> +
> +	anon_nr = mem_cgroup_hierarchical_nr_lru_pages(memcg, LRU_ALL_ANON);
> +	seq_printf(m, "hierarchical_anon=%lu", anon_nr);
> +	for_each_node_state(nid, N_HIGH_MEMORY) {
> +		node_nr = mem_cgroup_node_hierarchical_nr_lru_pages(memcg, nid,
> +				LRU_ALL_ANON);
> +		seq_printf(m, " N%d=%lu", nid, node_nr);
> +	}
> +	seq_putc(m, '\n');
> +
> +	unevictable_nr = mem_cgroup_hierarchical_nr_lru_pages(memcg,
> +						BIT(LRU_UNEVICTABLE));
> +	seq_printf(m, "hierarchical_unevictable=%lu", unevictable_nr);
> +	for_each_node_state(nid, N_HIGH_MEMORY) {
> +		node_nr = mem_cgroup_node_hierarchical_nr_lru_pages(memcg, nid,
> +				BIT(LRU_UNEVICTABLE));
> +		seq_printf(m, " N%d=%lu", nid, node_nr);
> +	}
> +	seq_putc(m, '\n');
> +
>  	return 0;
>  }
>  #endif /* CONFIG_NUMA */
> -- 
> 1.8.1.3
> 

-- 
Michal Hocko
SUSE Labs

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH 1/2] memcg: refactor mem_control_numa_stat_show()

[Greg Thelen]

Refactor mem_control_numa_stat_show() to use a new new stats structure
for smaller and simpler code.  This consolidates nearly identical
code.

          text      rodata     data       bss
before  5,913,438  1,695,438  756,608  1,712,128
after   5,912,734  1,695,502  756,608  1,712,128

Signed-off-by: Greg Thelen <gthelen@google.com>
Signed-off-by: Ying Han <yinghan@google.com>
---
 mm/memcontrol.c | 57 +++++++++++++++++++++++----------------------------------
 1 file changed, 23 insertions(+), 34 deletions(-)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 2b55222..e73526e 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -5228,45 +5228,34 @@ static int mem_cgroup_move_charge_write(struct cgroup *cgrp,
 static int memcg_numa_stat_show(struct cgroup *cont, struct cftype *cft,
 				      struct seq_file *m)
 {
+	struct numa_stat {
+		const char *name;
+		unsigned int lru_mask;
+	};
+
+	static const struct numa_stat stats[] = {
+		{ "total", LRU_ALL },
+		{ "file", LRU_ALL_FILE },
+		{ "anon", LRU_ALL_ANON },
+		{ "unevictable", BIT(LRU_UNEVICTABLE) },
+		{ NULL, 0 }  /* terminator */
+	};
+	const struct numa_stat *stat;
 	int nid;
-	unsigned long total_nr, file_nr, anon_nr, unevictable_nr;
-	unsigned long node_nr;
+	unsigned long nr;
 	struct mem_cgroup *memcg = mem_cgroup_from_cont(cont);
 
-	total_nr = mem_cgroup_nr_lru_pages(memcg, LRU_ALL);
-	seq_printf(m, "total=%lu", total_nr);
-	for_each_node_state(nid, N_MEMORY) {
-		node_nr = mem_cgroup_node_nr_lru_pages(memcg, nid, LRU_ALL);
-		seq_printf(m, " N%d=%lu", nid, node_nr);
-	}
-	seq_putc(m, '\n');
-
-	file_nr = mem_cgroup_nr_lru_pages(memcg, LRU_ALL_FILE);
-	seq_printf(m, "file=%lu", file_nr);
-	for_each_node_state(nid, N_MEMORY) {
-		node_nr = mem_cgroup_node_nr_lru_pages(memcg, nid,
-				LRU_ALL_FILE);
-		seq_printf(m, " N%d=%lu", nid, node_nr);
-	}
-	seq_putc(m, '\n');
-
-	anon_nr = mem_cgroup_nr_lru_pages(memcg, LRU_ALL_ANON);
-	seq_printf(m, "anon=%lu", anon_nr);
-	for_each_node_state(nid, N_MEMORY) {
-		node_nr = mem_cgroup_node_nr_lru_pages(memcg, nid,
-				LRU_ALL_ANON);
-		seq_printf(m, " N%d=%lu", nid, node_nr);
+	for (stat = stats; stat->name; stat++) {
+		nr = mem_cgroup_nr_lru_pages(memcg, stat->lru_mask);
+		seq_printf(m, "%s=%lu", stat->name, nr);
+		for_each_node_state(nid, N_MEMORY) {
+			nr = mem_cgroup_node_nr_lru_pages(memcg, nid,
+							  stat->lru_mask);
+			seq_printf(m, " N%d=%lu", nid, nr);
+		}
+		seq_putc(m, '\n');
 	}
-	seq_putc(m, '\n');
 
-	unevictable_nr = mem_cgroup_nr_lru_pages(memcg, BIT(LRU_UNEVICTABLE));
-	seq_printf(m, "unevictable=%lu", unevictable_nr);
-	for_each_node_state(nid, N_MEMORY) {
-		node_nr = mem_cgroup_node_nr_lru_pages(memcg, nid,
-				BIT(LRU_UNEVICTABLE));
-		seq_printf(m, " N%d=%lu", nid, node_nr);
-	}
-	seq_putc(m, '\n');
 	return 0;
 }
 #endif /* CONFIG_NUMA */
-- 
1.8.2.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH 2/2] memcg: support hierarchical memory.numa_stats

[Greg Thelen]

From: Ying Han <yinghan@google.com>

The memory.numa_stat file was not hierarchical.  Memory charged to the
children was not shown in parent's numa_stat.

This change adds the "hierarchical_" stats to the existing stats.  The
new hierarchical stats include the sum of all children's values in
addition to the value of the memcg.

Tested: Create cgroup a, a/b and run workload under b.  The values of
b are included in the "hierarchical_*" under a.

$ cat /dev/cgroup/memory/a/memory.numa_stat
total=0 N0=0 N1=0 N2=0 N3=0
file=0 N0=0 N1=0 N2=0 N3=0
anon=0 N0=0 N1=0 N2=0 N3=0
unevictable=0 N0=0 N1=0 N2=0 N3=0
hierarchical_total=21395 N0=0 N1=16 N2=21379 N3=0
hierarchical_file=21368 N0=0 N1=0 N2=21368 N3=0
hierarchical_anon=27 N0=0 N1=16 N2=11 N3=0
hierarchical_unevictable=0 N0=0 N1=0 N2=0 N3=0

$ cat /dev/cgroup/memory/a/b/memory.numa_stat
total=21395 N0=0 N1=16 N2=21379 N3=0
file=21368 N0=0 N1=0 N2=21368 N3=0
anon=27 N0=0 N1=16 N2=11 N3=0
unevictable=0 N0=0 N1=0 N2=0 N3=0
hierarchical_total=21395 N0=0 N1=16 N2=21379 N3=0
hierarchical_file=21368 N0=0 N1=0 N2=21368 N3=0
hierarchical_anon=27 N0=0 N1=16 N2=11 N3=0
hierarchical_unevictable=0 N0=0 N1=0 N2=0 N3=0

Signed-off-by: Ying Han <yinghan@google.com>
Signed-off-by: Greg Thelen <gthelen@google.com>
---
 mm/memcontrol.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index e73526e..f0ec99d 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -5244,6 +5244,7 @@ static int memcg_numa_stat_show(struct cgroup *cont, struct cftype *cft,
 	int nid;
 	unsigned long nr;
 	struct mem_cgroup *memcg = mem_cgroup_from_cont(cont);
+	struct mem_cgroup *iter;
 
 	for (stat = stats; stat->name; stat++) {
 		nr = mem_cgroup_nr_lru_pages(memcg, stat->lru_mask);
@@ -5256,6 +5257,21 @@ static int memcg_numa_stat_show(struct cgroup *cont, struct cftype *cft,
 		seq_putc(m, '\n');
 	}
 
+	for (stat = stats; stat->name; stat++) {
+		nr = 0;
+		for_each_mem_cgroup_tree(iter, memcg)
+			nr += mem_cgroup_nr_lru_pages(iter, stat->lru_mask);
+		seq_printf(m, "hierarchical_%s=%lu", stat->name, nr);
+		for_each_node_state(nid, N_MEMORY) {
+			nr = 0;
+			for_each_mem_cgroup_tree(iter, memcg)
+				nr += mem_cgroup_node_nr_lru_pages(
+					iter, nid, stat->lru_mask);
+			seq_printf(m, " N%d=%lu", nid, nr);
+		}
+		seq_putc(m, '\n');
+	}
+
 	return 0;
 }
 #endif /* CONFIG_NUMA */
-- 
1.8.2.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

[PATCH 2/2 v2] memcg: support hierarchical memory.numa_stats

[Greg Thelen]

From: Ying Han <yinghan@google.com>

The memory.numa_stat file was not hierarchical.  Memory charged to the
children was not shown in parent's numa_stat.

This change adds the "hierarchical_" stats to the existing stats.  The
new hierarchical stats include the sum of all children's values in
addition to the value of the memcg.

Tested: Create cgroup a, a/b and run workload under b.  The values of
b are included in the "hierarchical_*" under a.

$ cat /dev/cgroup/memory/a/memory.numa_stat
total=0 N0=0 N1=0 N2=0 N3=0
file=0 N0=0 N1=0 N2=0 N3=0
anon=0 N0=0 N1=0 N2=0 N3=0
unevictable=0 N0=0 N1=0 N2=0 N3=0
hierarchical_total=21395 N0=0 N1=16 N2=21379 N3=0
hierarchical_file=21368 N0=0 N1=0 N2=21368 N3=0
hierarchical_anon=27 N0=0 N1=16 N2=11 N3=0
hierarchical_unevictable=0 N0=0 N1=0 N2=0 N3=0

$ cat /dev/cgroup/memory/a/b/memory.numa_stat
total=21395 N0=0 N1=16 N2=21379 N3=0
file=21368 N0=0 N1=0 N2=21368 N3=0
anon=27 N0=0 N1=16 N2=11 N3=0
unevictable=0 N0=0 N1=0 N2=0 N3=0
hierarchical_total=21395 N0=0 N1=16 N2=21379 N3=0
hierarchical_file=21368 N0=0 N1=0 N2=21368 N3=0
hierarchical_anon=27 N0=0 N1=16 N2=11 N3=0
hierarchical_unevictable=0 N0=0 N1=0 N2=0 N3=0

Signed-off-by: Ying Han <yinghan@google.com>
Signed-off-by: Greg Thelen <gthelen@google.com>
---
Changelog:
- v2: add documentation

 Documentation/cgroups/memory.txt |  5 ++++-
 mm/memcontrol.c                  | 16 ++++++++++++++++
 2 files changed, 20 insertions(+), 1 deletion(-)

diff --git a/Documentation/cgroups/memory.txt b/Documentation/cgroups/memory.txt
index 8b8c28b..b519e74 100644
--- a/Documentation/cgroups/memory.txt
+++ b/Documentation/cgroups/memory.txt
@@ -568,7 +568,10 @@ node.  One of the use cases is evaluating application performance by
 combining this information with the application's CPU allocation.
 
 We export "total", "file", "anon" and "unevictable" pages per-node for
-each memcg.  The ouput format of memory.numa_stat is:
+each memcg and "hierarchical_" for sum of all hierarchical children's values
+in addition to the memcg's own value.
+
+The ouput format of memory.numa_stat is:
 
 total=<total pages> N0=<node 0 pages> N1=<node 1 pages> ...
 file=<total file pages> N0=<node 0 pages> N1=<node 1 pages> ...
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index e73526e..f0ec99d 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -5244,6 +5244,7 @@ static int memcg_numa_stat_show(struct cgroup *cont, struct cftype *cft,
 	int nid;
 	unsigned long nr;
 	struct mem_cgroup *memcg = mem_cgroup_from_cont(cont);
+	struct mem_cgroup *iter;
 
 	for (stat = stats; stat->name; stat++) {
 		nr = mem_cgroup_nr_lru_pages(memcg, stat->lru_mask);
@@ -5256,6 +5257,21 @@ static int memcg_numa_stat_show(struct cgroup *cont, struct cftype *cft,
 		seq_putc(m, '\n');
 	}
 
+	for (stat = stats; stat->name; stat++) {
+		nr = 0;
+		for_each_mem_cgroup_tree(iter, memcg)
+			nr += mem_cgroup_nr_lru_pages(iter, stat->lru_mask);
+		seq_printf(m, "hierarchical_%s=%lu", stat->name, nr);
+		for_each_node_state(nid, N_MEMORY) {
+			nr = 0;
+			for_each_mem_cgroup_tree(iter, memcg)
+				nr += mem_cgroup_node_nr_lru_pages(
+					iter, nid, stat->lru_mask);
+			seq_printf(m, " N%d=%lu", nid, nr);
+		}
+		seq_putc(m, '\n');
+	}
+
 	return 0;
 }
 #endif /* CONFIG_NUMA */
-- 
1.8.2.1

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>