summaryrefslogtreecommitdiff
path: root/mm/memcontrol.c
diff options
context:
space:
mode:
authorBalbir Singh <balbir@linux.vnet.ibm.com>2009-09-23 15:56:42 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2009-09-24 07:20:59 -0700
commit0c3e73e84fe3f64cf1c2e8bb4e91e8901cbcdc38 (patch)
tree1c3e27fce3babd3703b34c9a353f253fcefb6325 /mm/memcontrol.c
parent4e41695356fb4e0b153be1440ad027e46e0a7ea2 (diff)
memcg: improve resource counter scalability
Reduce the resource counter overhead (mostly spinlock) associated with the root cgroup. This is a part of the several patches to reduce mem cgroup overhead. I had posted other approaches earlier (including using percpu counters). Those patches will be a natural addition and will be added iteratively on top of these. The patch stops resource counter accounting for the root cgroup. The data for display is derived from the statisitcs we maintain via mem_cgroup_charge_statistics (which is more scalable). What happens today is that, we do double accounting, once using res_counter_charge() and once using memory_cgroup_charge_statistics(). For the root, since we don't implement limits any more, we don't need to track every charge via res_counter_charge() and check for limit being exceeded and reclaim. The main mem->res usage_in_bytes can be derived by summing the cache and rss usage data from memory statistics (MEM_CGROUP_STAT_RSS and MEM_CGROUP_STAT_CACHE). However, for memsw->res usage_in_bytes, we need additional data about swapped out memory. This patch adds a MEM_CGROUP_STAT_SWAPOUT and uses that along with MEM_CGROUP_STAT_RSS and MEM_CGROUP_STAT_CACHE to derive the memsw data. This data is computed recursively when hierarchy is enabled. The tests results I see on a 24 way show that 1. The lock contention disappears from /proc/lock_stats 2. The results of the test are comparable to running with cgroup_disable=memory. Here is a sample of my program runs Without Patch Performance counter stats for '/home/balbir/parallel_pagefault': 7192804.124144 task-clock-msecs # 23.937 CPUs 424691 context-switches # 0.000 M/sec 267 CPU-migrations # 0.000 M/sec 28498113 page-faults # 0.004 M/sec 5826093739340 cycles # 809.989 M/sec 408883496292 instructions # 0.070 IPC 7057079452 cache-references # 0.981 M/sec 3036086243 cache-misses # 0.422 M/sec 300.485365680 seconds time elapsed With cgroup_disable=memory Performance counter stats for '/home/balbir/parallel_pagefault': 7182183.546587 task-clock-msecs # 23.915 CPUs 425458 context-switches # 0.000 M/sec 203 CPU-migrations # 0.000 M/sec 92545093 page-faults # 0.013 M/sec 6034363609986 cycles # 840.185 M/sec 437204346785 instructions # 0.072 IPC 6636073192 cache-references # 0.924 M/sec 2358117732 cache-misses # 0.328 M/sec 300.320905827 seconds time elapsed With this patch applied Performance counter stats for '/home/balbir/parallel_pagefault': 7191619.223977 task-clock-msecs # 23.955 CPUs 422579 context-switches # 0.000 M/sec 88 CPU-migrations # 0.000 M/sec 91946060 page-faults # 0.013 M/sec 5957054385619 cycles # 828.333 M/sec 1058117350365 instructions # 0.178 IPC 9161776218 cache-references # 1.274 M/sec 1920494280 cache-misses # 0.267 M/sec 300.218764862 seconds time elapsed Data from Prarit (kernel compile with make -j64 on a 64 CPU/32G machine) For a single run Without patch real 27m8.988s user 87m24.916s sys 382m6.037s With patch real 4m18.607s user 84m58.943s sys 50m52.682s With config turned off real 4m54.972s user 90m13.456s sys 50m19.711s NOTE: The data looks counterintuitive due to the increased performance with the patch, even over the config being turned off. We probably need more runs, but so far all testing has shown that the patches definitely help. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com> Cc: Prarit Bhargava <prarit@redhat.com> Cc: Andi Kleen <andi@firstfloor.org> Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Reviewed-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Paul Menage <menage@google.com> Cc: Li Zefan <lizf@cn.fujitsu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'mm/memcontrol.c')
-rw-r--r--mm/memcontrol.c121
1 files changed, 100 insertions, 21 deletions
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 011aba6cad70..2011f15278fd 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -70,6 +70,7 @@ enum mem_cgroup_stat_index {
MEM_CGROUP_STAT_PGPGIN_COUNT, /* # of pages paged in */
MEM_CGROUP_STAT_PGPGOUT_COUNT, /* # of pages paged out */
MEM_CGROUP_STAT_EVENTS, /* sum of pagein + pageout for internal use */
+ MEM_CGROUP_STAT_SWAPOUT, /* # of pages, swapped out */
MEM_CGROUP_STAT_NSTATS,
};
@@ -478,11 +479,24 @@ mem_cgroup_largest_soft_limit_node(struct mem_cgroup_tree_per_zone *mctz)
return mz;
}
+static void mem_cgroup_swap_statistics(struct mem_cgroup *mem,
+ bool charge)
+{
+ int val = (charge) ? 1 : -1;
+ struct mem_cgroup_stat *stat = &mem->stat;
+ struct mem_cgroup_stat_cpu *cpustat;
+ int cpu = get_cpu();
+
+ cpustat = &stat->cpustat[cpu];
+ __mem_cgroup_stat_add_safe(cpustat, MEM_CGROUP_STAT_SWAPOUT, val);
+ put_cpu();
+}
+
static void mem_cgroup_charge_statistics(struct mem_cgroup *mem,
struct page_cgroup *pc,
bool charge)
{
- int val = (charge)? 1 : -1;
+ int val = (charge) ? 1 : -1;
struct mem_cgroup_stat *stat = &mem->stat;
struct mem_cgroup_stat_cpu *cpustat;
int cpu = get_cpu();
@@ -1285,9 +1299,11 @@ static int __mem_cgroup_try_charge(struct mm_struct *mm,
VM_BUG_ON(css_is_removed(&mem->css));
while (1) {
- int ret;
+ int ret = 0;
unsigned long flags = 0;
+ if (mem_cgroup_is_root(mem))
+ goto done;
ret = res_counter_charge(&mem->res, PAGE_SIZE, &fail_res,
&soft_fail_res);
if (likely(!ret)) {
@@ -1347,6 +1363,7 @@ static int __mem_cgroup_try_charge(struct mm_struct *mm,
if (mem_cgroup_soft_limit_check(mem_over_soft_limit))
mem_cgroup_update_tree(mem_over_soft_limit, page);
}
+done:
return 0;
nomem:
css_put(&mem->css);
@@ -1419,9 +1436,12 @@ static void __mem_cgroup_commit_charge(struct mem_cgroup *mem,
lock_page_cgroup(pc);
if (unlikely(PageCgroupUsed(pc))) {
unlock_page_cgroup(pc);
- res_counter_uncharge(&mem->res, PAGE_SIZE, NULL);
- if (do_swap_account)
- res_counter_uncharge(&mem->memsw, PAGE_SIZE, NULL);
+ if (!mem_cgroup_is_root(mem)) {
+ res_counter_uncharge(&mem->res, PAGE_SIZE, NULL);
+ if (do_swap_account)
+ res_counter_uncharge(&mem->memsw, PAGE_SIZE,
+ NULL);
+ }
css_put(&mem->css);
return;
}
@@ -1498,7 +1518,8 @@ static int mem_cgroup_move_account(struct page_cgroup *pc,
if (pc->mem_cgroup != from)
goto out;
- res_counter_uncharge(&from->res, PAGE_SIZE, NULL);
+ if (!mem_cgroup_is_root(from))
+ res_counter_uncharge(&from->res, PAGE_SIZE, NULL);
mem_cgroup_charge_statistics(from, pc, false);
page = pc->page;
@@ -1517,7 +1538,7 @@ static int mem_cgroup_move_account(struct page_cgroup *pc,
1);
}
- if (do_swap_account)
+ if (do_swap_account && !mem_cgroup_is_root(from))
res_counter_uncharge(&from->memsw, PAGE_SIZE, NULL);
css_put(&from->css);
@@ -1588,9 +1609,11 @@ uncharge:
/* drop extra refcnt by try_charge() */
css_put(&parent->css);
/* uncharge if move fails */
- res_counter_uncharge(&parent->res, PAGE_SIZE, NULL);
- if (do_swap_account)
- res_counter_uncharge(&parent->memsw, PAGE_SIZE, NULL);
+ if (!mem_cgroup_is_root(parent)) {
+ res_counter_uncharge(&parent->res, PAGE_SIZE, NULL);
+ if (do_swap_account)
+ res_counter_uncharge(&parent->memsw, PAGE_SIZE, NULL);
+ }
return ret;
}
@@ -1779,7 +1802,10 @@ __mem_cgroup_commit_charge_swapin(struct page *page, struct mem_cgroup *ptr,
* This recorded memcg can be obsolete one. So, avoid
* calling css_tryget
*/
- res_counter_uncharge(&memcg->memsw, PAGE_SIZE, NULL);
+ if (!mem_cgroup_is_root(memcg))
+ res_counter_uncharge(&memcg->memsw, PAGE_SIZE,
+ NULL);
+ mem_cgroup_swap_statistics(memcg, false);
mem_cgroup_put(memcg);
}
rcu_read_unlock();
@@ -1804,9 +1830,11 @@ void mem_cgroup_cancel_charge_swapin(struct mem_cgroup *mem)
return;
if (!mem)
return;
- res_counter_uncharge(&mem->res, PAGE_SIZE, NULL);
- if (do_swap_account)
- res_counter_uncharge(&mem->memsw, PAGE_SIZE, NULL);
+ if (!mem_cgroup_is_root(mem)) {
+ res_counter_uncharge(&mem->res, PAGE_SIZE, NULL);
+ if (do_swap_account)
+ res_counter_uncharge(&mem->memsw, PAGE_SIZE, NULL);
+ }
css_put(&mem->css);
}
@@ -1859,9 +1887,14 @@ __mem_cgroup_uncharge_common(struct page *page, enum charge_type ctype)
break;
}
- res_counter_uncharge(&mem->res, PAGE_SIZE, &soft_limit_excess);
- if (do_swap_account && (ctype != MEM_CGROUP_CHARGE_TYPE_SWAPOUT))
- res_counter_uncharge(&mem->memsw, PAGE_SIZE, NULL);
+ if (!mem_cgroup_is_root(mem)) {
+ res_counter_uncharge(&mem->res, PAGE_SIZE, &soft_limit_excess);
+ if (do_swap_account &&
+ (ctype != MEM_CGROUP_CHARGE_TYPE_SWAPOUT))
+ res_counter_uncharge(&mem->memsw, PAGE_SIZE, NULL);
+ }
+ if (ctype == MEM_CGROUP_CHARGE_TYPE_SWAPOUT)
+ mem_cgroup_swap_statistics(mem, true);
mem_cgroup_charge_statistics(mem, pc, false);
ClearPageCgroupUsed(pc);
@@ -1952,7 +1985,9 @@ void mem_cgroup_uncharge_swap(swp_entry_t ent)
* We uncharge this because swap is freed.
* This memcg can be obsolete one. We avoid calling css_tryget
*/
- res_counter_uncharge(&memcg->memsw, PAGE_SIZE, NULL);
+ if (!mem_cgroup_is_root(memcg))
+ res_counter_uncharge(&memcg->memsw, PAGE_SIZE, NULL);
+ mem_cgroup_swap_statistics(memcg, false);
mem_cgroup_put(memcg);
}
rcu_read_unlock();
@@ -2464,20 +2499,64 @@ static int mem_cgroup_hierarchy_write(struct cgroup *cont, struct cftype *cft,
return retval;
}
+struct mem_cgroup_idx_data {
+ s64 val;
+ enum mem_cgroup_stat_index idx;
+};
+
+static int
+mem_cgroup_get_idx_stat(struct mem_cgroup *mem, void *data)
+{
+ struct mem_cgroup_idx_data *d = data;
+ d->val += mem_cgroup_read_stat(&mem->stat, d->idx);
+ return 0;
+}
+
+static void
+mem_cgroup_get_recursive_idx_stat(struct mem_cgroup *mem,
+ enum mem_cgroup_stat_index idx, s64 *val)
+{
+ struct mem_cgroup_idx_data d;
+ d.idx = idx;
+ d.val = 0;
+ mem_cgroup_walk_tree(mem, &d, mem_cgroup_get_idx_stat);
+ *val = d.val;
+}
+
static u64 mem_cgroup_read(struct cgroup *cont, struct cftype *cft)
{
struct mem_cgroup *mem = mem_cgroup_from_cont(cont);
- u64 val = 0;
+ u64 idx_val, val;
int type, name;
type = MEMFILE_TYPE(cft->private);
name = MEMFILE_ATTR(cft->private);
switch (type) {
case _MEM:
- val = res_counter_read_u64(&mem->res, name);
+ if (name == RES_USAGE && mem_cgroup_is_root(mem)) {
+ mem_cgroup_get_recursive_idx_stat(mem,
+ MEM_CGROUP_STAT_CACHE, &idx_val);
+ val = idx_val;
+ mem_cgroup_get_recursive_idx_stat(mem,
+ MEM_CGROUP_STAT_RSS, &idx_val);
+ val += idx_val;
+ val <<= PAGE_SHIFT;
+ } else
+ val = res_counter_read_u64(&mem->res, name);
break;
case _MEMSWAP:
- val = res_counter_read_u64(&mem->memsw, name);
+ if (name == RES_USAGE && mem_cgroup_is_root(mem)) {
+ mem_cgroup_get_recursive_idx_stat(mem,
+ MEM_CGROUP_STAT_CACHE, &idx_val);
+ val = idx_val;
+ mem_cgroup_get_recursive_idx_stat(mem,
+ MEM_CGROUP_STAT_RSS, &idx_val);
+ val += idx_val;
+ mem_cgroup_get_recursive_idx_stat(mem,
+ MEM_CGROUP_STAT_SWAPOUT, &idx_val);
+ val <<= PAGE_SHIFT;
+ } else
+ val = res_counter_read_u64(&mem->memsw, name);
break;
default:
BUG();