Move memory controller allocations to their own slabs

Move the memory controller data structures page_cgroup and mem_cgroup_per_zone to their own slab caches. It saves space on the system, allocations are not necessarily pushed to order of 2 and should provide performance benefits.

Quoted 5 lines · On Tue, 11 Mar 2008 10:01:49 +0530 Balbir Singh <[email protected]>…

Move the memory controller data structures page_cgroup and mem_cgroup_per_zone to their own slab caches. It saves space on the system, allocations are not necessarily pushed to order of 2 and should provide performance benefits.

eh? Those structures are tiny. Which slab allocator has gone and used an order-2 allocation and for which structure did it (stupidly) do this?

Move the memory controller data structures page_cgroup and mem_cgroup_per_zone to their own slab caches. It saves space on the system, allocations are not necessarily pushed to order of 2 and should provide performance benefits. Users who disable the memory controller can also double check that the memory controller is not allocating page_cgroup's.

Signed-off-by Balbir Singh <[email protected]>

---

 linux/memcontrol.h |    0 
 mm/memcontrol.c    |   21 ++++++++++++++-------
 2 files changed, 14 insertions(+), 7 deletions(-)

diff -puN mm/memcontrol.c~memory-controller-move-to-own-slab mm/memcontrol.c

mm/memcontrol.c renamed from mm/memcontrol.c~memory-controller-move-to-own-slab +14 -7

@@ -26,6 +26,7 @@
 #include <linux/backing-dev.h>
 #include <linux/bit_spinlock.h>
 #include <linux/rcupdate.h>
+#include <linux/slab.h>
 #include <linux/swap.h>
 #include <linux/spinlock.h>
 #include <linux/fs.h>

@@ -35,6 +36,8 @@
 
 struct cgroup_subsys mem_cgroup_subsys;
 static const int MEM_CGROUP_RECLAIM_RETRIES = 5;
+static struct kmem_cache *page_cgroup_cache;
+static struct kmem_cache *mem_cgroup_per_zone_cache;
 
 /*
  * Statistics for memory cgroup.

@@ -560,7 +563,7 @@ retry:
 	}
 	unlock_page_cgroup(page);
 
-	pc = kzalloc(sizeof(struct page_cgroup), gfp_mask);
+	pc = kmem_cache_zalloc(page_cgroup_cache, gfp_mask);
 	if (pc == NULL)
 		goto err;
 

@@ -622,7 +625,7 @@ retry:
 		 */
 		res_counter_uncharge(&mem->res, PAGE_SIZE);
 		css_put(&mem->css);
-		kfree(pc);
+		kmem_cache_free(page_cgroup_cache, pc);
 		goto retry;
 	}
 	page_assign_page_cgroup(page, pc);

@@ -637,7 +640,7 @@ done:
 	return 0;
 out:
 	css_put(&mem->css);
-	kfree(pc);
+	kmem_cache_free(page_cgroup_cache, pc);
 err:
 	return -ENOMEM;
 }

@@ -695,7 +698,7 @@ void mem_cgroup_uncharge_page(struct pag
 		res_counter_uncharge(&mem->res, PAGE_SIZE);
 		css_put(&mem->css);
 
-		kfree(pc);
+		kmem_cache_free(page_cgroup_cache, pc);
 		return;
 	}
 

@@ -988,9 +991,10 @@ static int alloc_mem_cgroup_per_zone_inf
 	 *       function.
 	 */
 	if (node_state(node, N_HIGH_MEMORY))
-		pn = kmalloc_node(sizeof(*pn), GFP_KERNEL, node);
+		pn = kmem_cache_alloc_node(mem_cgroup_per_zone_cache,
+						GFP_KERNEL, node);
 	else
-		pn = kmalloc(sizeof(*pn), GFP_KERNEL);
+		pn = kmem_cache_alloc(mem_cgroup_per_zone_cache, GFP_KERNEL);
 	if (!pn)
 		return 1;
 

@@ -1008,7 +1012,7 @@ static int alloc_mem_cgroup_per_zone_inf
 
 static void free_mem_cgroup_per_zone_info(struct mem_cgroup *mem, int node)
 {
-	kfree(mem->info.nodeinfo[node]);
+	kmem_cache_free(mem_cgroup_per_zone_cache, mem->info.nodeinfo[node]);
 }
 
 static struct cgroup_subsys_state *

@@ -1020,6 +1024,9 @@ mem_cgroup_create(struct cgroup_subsys *
 	if (unlikely((cont->parent) == NULL)) {
 		mem = &init_mem_cgroup;
 		init_mm.mem_cgroup = mem;
+		page_cgroup_cache = KMEM_CACHE(page_cgroup, SLAB_PANIC);
+		mem_cgroup_per_zone_cache = KMEM_CACHE(mem_cgroup_per_zone,
+							SLAB_PANIC);
 	} else
 		mem = kzalloc(sizeof(struct mem_cgroup), GFP_KERNEL);
 

diff -puN include/linux/memcontrol.h~memory-controller-move-to-own-slab include/linux/memcontrol.h _

Signature · 10 lines

	Warm Regards,
	Balbir Singh
	Linux Technology Center
	IBM, ISTL

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

Quoted 6 lines · On Tue, 11 Mar 2008 10:01:49 +0530 Balbir Singh <[email protected]>…

Move the memory controller data structures page_cgroup and mem_cgroup_per_zone to their own slab caches. It saves space on the system, allocations are not necessarily pushed to order of 2 and should provide performance benefits. Users who disable the memory controller can also double check that the memory controller is not allocating page_cgroup's.

Quoted 2 lines

Signed-off-by Balbir Singh <[email protected]>

---

I think using its own kmem_cache for mem_cgroup_per_zone is a bit overkill.

Signed-off-by Balbir Singh <[email protected]>

---

 linux/memcontrol.h |    0 

 mm/memcontrol.c    |   21 ++++++++++++++-------
 2 files changed, 14 insertions(+), 7 deletions(-)

Hi

Quoted 4 lines

Signed-off-by Balbir Singh <[email protected]>

---

 linux/memcontrol.h |    0 

???
unnecessary hunk?
or diff comannd bug?