The patch titled
memcg: add the pagefault count into memcg stats
has been removed from the -mm tree. Its filename was
add-the-pagefault-count-into-memcg-stats.patch
This patch was dropped because it was withdrawn
The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/
------------------------------------------------------
Subject: memcg: add the pagefault count into memcg stats
From: Ying Han <yinghan@xxxxxxxxxx>
Two new stats in per-memcg memory.stat which tracks the number of page
faults and number of major page faults.
"pgfault"
"pgmajfault"
They are different from "pgpgin"/"pgpgout" stat which count number of
pages charged/discharged to the cgroup and have no meaning of reading/
writing page to disk.
It is valuable to track the two stats for both measuring application's
performance as well as the efficiency of the kernel page reclaim path.
Counting pagefaults per process is useful, but we also need the aggregated
value since processes are monitored and controlled in cgroup basis in
memcg.
Functional test: check the total number of pgfault/pgmajfault of all
memcgs and compare with global vmstat value:
$ cat /proc/vmstat | grep fault
pgfault 1070751
pgmajfault 553
$ cat /dev/cgroup/memory.stat | grep fault
pgfault 1071138
pgmajfault 553
total_pgfault 1071142
total_pgmajfault 553
$ cat /dev/cgroup/A/memory.stat | grep fault
pgfault 199
pgmajfault 0
total_pgfault 199
total_pgmajfault 0
Performance test: run page fault test(pft) wit 16 thread on faulting in
15G anon pages in 16G container. There is no regression noticed on the
"flt/cpu/s"
Sample output from pft:
TAG pft:anon-sys-default:
Gb Thr CLine User System Wall flt/cpu/s fault/wsec
15 16 1 0.67s 233.41s 14.76s 16798.546 266356.260
+-------------------------------------------------------------------------+
N Min Max Median Avg Stddev
x 10 16682.962 17344.027 16913.524 16928.812 166.5362
+ 10 16695.568 16923.896 16820.604 16824.652 84.816568
No difference proven at 95.0% confidence
[akpm@xxxxxxxxxxxxxxxxxxxx: fix build]
[hughd@xxxxxxxxxx: shmem fix]
Signed-off-by: Ying Han <yinghan@xxxxxxxxxx>
Acked-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Cc: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx>
Reviewed-by: Minchan Kim <minchan.kim@xxxxxxxxx>
Cc: Daisuke Nishimura <nishimura@xxxxxxxxxxxxxxxxx>
Acked-by: Balbir Singh <balbir@xxxxxxxxxxxxxxxxxx>
Signed-off-by: Hugh Dickins <hughd@xxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
fs/ncpfs/mmap.c | 2 +
include/linux/memcontrol.h | 7 +++++
mm/filemap.c | 1
mm/memcontrol.c | 47 +++++++++++++++++++++++++++++++++++
mm/memory.c | 2 +
mm/shmem.c | 11 ++++----
6 files changed, 65 insertions(+), 5 deletions(-)
diff -puN fs/ncpfs/mmap.c~add-the-pagefault-count-into-memcg-stats fs/ncpfs/mmap.c
--- a/fs/ncpfs/mmap.c~add-the-pagefault-count-into-memcg-stats
+++ a/fs/ncpfs/mmap.c
@@ -16,6 +16,7 @@
#include <linux/mman.h>
#include <linux/string.h>
#include <linux/fcntl.h>
+#include <linux/memcontrol.h>
#include <asm/uaccess.h>
#include <asm/system.h>
@@ -92,6 +93,7 @@ static int ncp_file_mmap_fault(struct vm
* -- wli
*/
count_vm_event(PGMAJFAULT);
+ mem_cgroup_count_vm_event(area->vm_mm, PGMAJFAULT);
return VM_FAULT_MAJOR;
}
diff -puN include/linux/memcontrol.h~add-the-pagefault-count-into-memcg-stats include/linux/memcontrol.h
--- a/include/linux/memcontrol.h~add-the-pagefault-count-into-memcg-stats
+++ a/include/linux/memcontrol.h
@@ -20,6 +20,8 @@
#ifndef _LINUX_MEMCONTROL_H
#define _LINUX_MEMCONTROL_H
#include <linux/cgroup.h>
+#include <linux/vm_event_item.h>
+
struct mem_cgroup;
struct page_cgroup;
struct page;
@@ -147,6 +149,7 @@ unsigned long mem_cgroup_soft_limit_recl
unsigned long *total_scanned);
u64 mem_cgroup_get_limit(struct mem_cgroup *mem);
+void mem_cgroup_count_vm_event(struct mm_struct *mm, enum vm_event_item idx);
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
void mem_cgroup_split_huge_fixup(struct page *head, struct page *tail);
#endif
@@ -355,6 +358,10 @@ static inline void mem_cgroup_split_huge
{
}
+static inline
+void mem_cgroup_count_vm_event(struct mm_struct *mm, enum vm_event_item idx)
+{
+}
#endif /* CONFIG_CGROUP_MEM_CONT */
#if !defined(CONFIG_CGROUP_MEM_RES_CTLR) || !defined(CONFIG_DEBUG_VM)
diff -puN mm/filemap.c~add-the-pagefault-count-into-memcg-stats mm/filemap.c
--- a/mm/filemap.c~add-the-pagefault-count-into-memcg-stats
+++ a/mm/filemap.c
@@ -1661,6 +1661,7 @@ int filemap_fault(struct vm_area_struct
/* No page in the page cache at all */
do_sync_mmap_readahead(vma, ra, file, offset);
count_vm_event(PGMAJFAULT);
+ mem_cgroup_count_vm_event(vma->vm_mm, PGMAJFAULT);
ret = VM_FAULT_MAJOR;
retry_find:
page = find_get_page(mapping, offset);
diff -puN mm/memcontrol.c~add-the-pagefault-count-into-memcg-stats mm/memcontrol.c
--- a/mm/memcontrol.c~add-the-pagefault-count-into-memcg-stats
+++ a/mm/memcontrol.c
@@ -94,6 +94,8 @@ enum mem_cgroup_events_index {
MEM_CGROUP_EVENTS_PGPGIN, /* # of pages paged in */
MEM_CGROUP_EVENTS_PGPGOUT, /* # of pages paged out */
MEM_CGROUP_EVENTS_COUNT, /* # of pages paged in/out */
+ MEM_CGROUP_EVENTS_PGFAULT, /* # of page-faults */
+ MEM_CGROUP_EVENTS_PGMAJFAULT, /* # of major page-faults */
MEM_CGROUP_EVENTS_SOFT_KSWAPD_STEAL, /* # of pages reclaimed from */
/* soft reclaim in background reclaim */
MEM_CGROUP_EVENTS_SOFT_KSWAPD_SCAN, /* # of pages scanned from */
@@ -593,6 +595,16 @@ static void mem_cgroup_swap_statistics(s
this_cpu_add(mem->stat->count[MEM_CGROUP_STAT_SWAPOUT], val);
}
+void mem_cgroup_pgfault(struct mem_cgroup *mem, int val)
+{
+ this_cpu_add(mem->stat->events[MEM_CGROUP_EVENTS_PGFAULT], val);
+}
+
+void mem_cgroup_pgmajfault(struct mem_cgroup *mem, int val)
+{
+ this_cpu_add(mem->stat->events[MEM_CGROUP_EVENTS_PGMAJFAULT], val);
+}
+
static unsigned long mem_cgroup_read_events(struct mem_cgroup *mem,
enum mem_cgroup_events_index idx)
{
@@ -847,6 +859,33 @@ static inline bool mem_cgroup_is_root(st
return (mem == root_mem_cgroup);
}
+void mem_cgroup_count_vm_event(struct mm_struct *mm, enum vm_event_item idx)
+{
+ struct mem_cgroup *mem;
+
+ if (!mm)
+ return;
+
+ rcu_read_lock();
+ mem = mem_cgroup_from_task(rcu_dereference(mm->owner));
+ if (unlikely(!mem))
+ goto out;
+
+ switch (idx) {
+ case PGMAJFAULT:
+ mem_cgroup_pgmajfault(mem, 1);
+ break;
+ case PGFAULT:
+ mem_cgroup_pgfault(mem, 1);
+ break;
+ default:
+ BUG();
+ }
+out:
+ rcu_read_unlock();
+}
+EXPORT_SYMBOL(mem_cgroup_count_vm_event);
+
/*
* Following LRU functions are allowed to be used without PCG_LOCK.
* Operations are called by routine of global LRU independently from memcg.
@@ -3824,6 +3863,8 @@ enum {
MCS_PGPGIN,
MCS_PGPGOUT,
MCS_SWAP,
+ MCS_PGFAULT,
+ MCS_PGMAJFAULT,
MCS_SOFT_KSWAPD_STEAL,
MCS_SOFT_KSWAPD_SCAN,
MCS_SOFT_DIRECT_STEAL,
@@ -3850,6 +3891,8 @@ struct {
{"pgpgin", "total_pgpgin"},
{"pgpgout", "total_pgpgout"},
{"swap", "total_swap"},
+ {"pgfault", "total_pgfault"},
+ {"pgmajfault", "total_pgmajfault"},
{"soft_kswapd_steal", "total_soft_kswapd_steal"},
{"soft_kswapd_scan", "total_soft_kswapd_scan"},
{"soft_direct_steal", "total_soft_direct_steal"},
@@ -3890,6 +3933,10 @@ mem_cgroup_get_local_stat(struct mem_cgr
s->stat[MCS_SOFT_DIRECT_STEAL] += val;
val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_SOFT_DIRECT_SCAN);
s->stat[MCS_SOFT_DIRECT_SCAN] += val;
+ val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_PGFAULT);
+ s->stat[MCS_PGFAULT] += val;
+ val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_PGMAJFAULT);
+ s->stat[MCS_PGMAJFAULT] += val;
/* per zone stat */
val = mem_cgroup_get_local_zonestat(mem, LRU_INACTIVE_ANON);
diff -puN mm/memory.c~add-the-pagefault-count-into-memcg-stats mm/memory.c
--- a/mm/memory.c~add-the-pagefault-count-into-memcg-stats
+++ a/mm/memory.c
@@ -2874,6 +2874,7 @@ static int do_swap_page(struct mm_struct
/* Had to read the page from swap area: Major fault */
ret = VM_FAULT_MAJOR;
count_vm_event(PGMAJFAULT);
+ mem_cgroup_count_vm_event(mm, PGMAJFAULT);
} else if (PageHWPoison(page)) {
/*
* hwpoisoned dirty swapcache pages are kept for killing
@@ -3413,6 +3414,7 @@ int handle_mm_fault(struct mm_struct *mm
__set_current_state(TASK_RUNNING);
count_vm_event(PGFAULT);
+ mem_cgroup_count_vm_event(mm, PGFAULT);
/* do counter updates before entering really critical section. */
check_sync_rss_stat(current);
diff -puN mm/shmem.c~add-the-pagefault-count-into-memcg-stats mm/shmem.c
--- a/mm/shmem.c~add-the-pagefault-count-into-memcg-stats
+++ a/mm/shmem.c
@@ -1304,12 +1304,10 @@ repeat:
swappage = lookup_swap_cache(swap);
if (!swappage) {
shmem_swp_unmap(entry);
+ spin_unlock(&info->lock);
/* here we actually do the io */
- if (type && !(*type & VM_FAULT_MAJOR)) {
- __count_vm_event(PGMAJFAULT);
+ if (type)
*type |= VM_FAULT_MAJOR;
- }
- spin_unlock(&info->lock);
swappage = shmem_swapin(swap, gfp, info, idx);
if (!swappage) {
spin_lock(&info->lock);
@@ -1548,7 +1546,10 @@ static int shmem_fault(struct vm_area_st
error = shmem_getpage(inode, vmf->pgoff, &vmf->page, SGP_CACHE, &ret);
if (error)
return ((error == -ENOMEM) ? VM_FAULT_OOM : VM_FAULT_SIGBUS);
-
+ if (ret & VM_FAULT_MAJOR) {
+ count_vm_event(PGMAJFAULT);
+ mem_cgroup_count_vm_event(vma->vm_mm, PGMAJFAULT);
+ }
return ret | VM_FAULT_LOCKED;
}
_
Patches currently in -mm which might be from yinghan@xxxxxxxxxx are
vmscan-change-shrink_slab-interfaces-by-passing-shrink_control.patch
vmscan-change-shrink_slab-interfaces-by-passing-shrink_control-fix.patch
vmscan-change-shrink_slab-interfaces-by-passing-shrink_control-fix-2.patch
vmscan-change-shrinker-api-by-passing-shrink_control-struct.patch
vmscan-change-shrinker-api-by-passing-shrink_control-struct-fix.patch
vmscan-change-shrinker-api-by-passing-shrink_control-struct-fix-2.patch
mm-move-enum-vm_event_item-into-a-standalone-header-file.patch
memcg-count-the-soft_limit-reclaim-in-global-background-reclaim.patch
memcg-add-the-soft_limit-reclaim-in-global-direct-reclaim.patch
memcg-reclaim-memory-from-nodes-in-round-robin-order.patch
memcg-reclaim-memory-from-nodes-in-round-robin-fix.patch
memcg-reclaim-memory-from-nodes-in-round-robin-fix-2.patch
memcg-fix-get_scan_count-for-small-targets.patch
memcg-remove-unused-retry-signal-from-reclaim.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
