The patch titled
memcg: break out event counters from other stats
has been added to the -mm tree. Its filename is
memcg-break-out-event-counters-from-other-stats.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/SubmitChecklist when testing your code ***
See http://userweb.kernel.org/~akpm/stuff/added-to-mm.txt to find
out what to do about this
The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/
------------------------------------------------------
Subject: memcg: break out event counters from other stats
From: Johannes Weiner <hannes@xxxxxxxxxxx>
For increasing and decreasing per-cpu cgroup usage counters it makes sense
to use signed types, as single per-cpu values might go negative during
updates. But this is not the case for only-ever-increasing event
counters.
All the counters have been signed 64-bit so far, which was enough to count
events even with the sign bit wasted.
The next patch narrows the usage counters type (on 32-bit CPUs, that is),
though, so break out the event counters and make them unsigned words as
they should have been from the start.
Signed-off-by: Johannes Weiner <hannes@xxxxxxxxxxx>
Cc: Andrea Righi <arighi@xxxxxxxxxxx>
Cc: Balbir Singh <balbir@xxxxxxxxxxxxxxxxxx>
Cc: Greg Thelen <gthelen@xxxxxxxxxx>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Cc: Daisuke Nishimura <nishimura@xxxxxxxxxxxxxxxxx>
Reviewed-by: Minchan Kim <minchan.kim@xxxxxxxxx>
Cc: Wu Fengguang <fengguang.wu@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
mm/memcontrol.c | 49 ++++++++++++++++++++++++++++++++++------------
1 file changed, 37 insertions(+), 12 deletions(-)
diff -puN mm/memcontrol.c~memcg-break-out-event-counters-from-other-stats mm/memcontrol.c
--- a/mm/memcontrol.c~memcg-break-out-event-counters-from-other-stats
+++ a/mm/memcontrol.c
@@ -85,21 +85,23 @@ enum mem_cgroup_stat_index {
*/
MEM_CGROUP_STAT_CACHE, /* # of pages charged as cache */
MEM_CGROUP_STAT_RSS, /* # of pages charged as anon rss */
- MEM_CGROUP_STAT_PGPGIN_COUNT, /* # of pages paged in */
- MEM_CGROUP_STAT_PGPGOUT_COUNT, /* # of pages paged out */
MEM_CGROUP_STAT_SWAPOUT, /* # of pages, swapped out */
MEM_CGROUP_STAT_FILE_MAPPED, /* # of pages charged as file rss */
MEM_CGROUP_STAT_FILE_DIRTY, /* # of dirty pages in page cache */
MEM_CGROUP_STAT_FILE_WRITEBACK, /* # of pages under writeback */
MEM_CGROUP_STAT_FILE_UNSTABLE_NFS, /* # of NFS unstable pages */
MEM_CGROUP_STAT_DATA, /* end of data requires synchronization */
- /* incremented at every pagein/pageout */
- MEM_CGROUP_EVENTS = MEM_CGROUP_STAT_DATA,
MEM_CGROUP_ON_MOVE, /* someone is moving account between groups */
-
MEM_CGROUP_STAT_NSTATS,
};
+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_NSTATS,
+};
+
enum {
MEM_CGROUP_DIRTY_RATIO,
MEM_CGROUP_DIRTY_LIMIT_IN_BYTES,
@@ -109,6 +111,7 @@ enum {
struct mem_cgroup_stat_cpu {
s64 count[MEM_CGROUP_STAT_NSTATS];
+ unsigned long events[MEM_CGROUP_EVENTS_NSTATS];
};
/*
@@ -612,6 +615,22 @@ static void mem_cgroup_swap_statistics(s
this_cpu_add(mem->stat->count[MEM_CGROUP_STAT_SWAPOUT], val);
}
+static unsigned long mem_cgroup_read_events(struct mem_cgroup *mem,
+ enum mem_cgroup_events_index idx)
+{
+ unsigned long val = 0;
+ int cpu;
+
+ for_each_online_cpu(cpu)
+ val += per_cpu(mem->stat->events[idx], cpu);
+#ifdef CONFIG_HOTPLUG_CPU
+ spin_lock(&mem->pcp_counter_lock);
+ val += mem->nocpu_base.events[idx];
+ spin_unlock(&mem->pcp_counter_lock);
+#endif
+ return val;
+}
+
static void mem_cgroup_charge_statistics(struct mem_cgroup *mem,
struct page_cgroup *pc,
bool charge)
@@ -626,10 +645,10 @@ static void mem_cgroup_charge_statistics
__this_cpu_add(mem->stat->count[MEM_CGROUP_STAT_RSS], val);
if (charge)
- __this_cpu_inc(mem->stat->count[MEM_CGROUP_STAT_PGPGIN_COUNT]);
+ __this_cpu_inc(mem->stat->events[MEM_CGROUP_EVENTS_PGPGIN]);
else
- __this_cpu_inc(mem->stat->count[MEM_CGROUP_STAT_PGPGOUT_COUNT]);
- __this_cpu_inc(mem->stat->count[MEM_CGROUP_EVENTS]);
+ __this_cpu_inc(mem->stat->events[MEM_CGROUP_EVENTS_PGPGOUT]);
+ __this_cpu_inc(mem->stat->events[MEM_CGROUP_EVENTS_COUNT]);
preempt_enable();
}
@@ -651,9 +670,9 @@ static unsigned long mem_cgroup_get_loca
static bool __memcg_event_check(struct mem_cgroup *mem, int event_mask_shift)
{
- s64 val;
+ unsigned long val;
- val = this_cpu_read(mem->stat->count[MEM_CGROUP_EVENTS]);
+ val = this_cpu_read(mem->stat->events[MEM_CGROUP_EVENTS_COUNT]);
return !(val & ((1 << event_mask_shift) - 1));
}
@@ -2055,6 +2074,12 @@ static void mem_cgroup_drain_pcp_counter
per_cpu(mem->stat->count[i], cpu) = 0;
mem->nocpu_base.count[i] += x;
}
+ for (i = 0; i < MEM_CGROUP_EVENTS_NSTATS; i++) {
+ unsigned long x = per_cpu(mem->stat->events[i], cpu);
+
+ per_cpu(mem->stat->events[i], cpu) = 0;
+ mem->nocpu_base.events[i] += x;
+ }
/* need to clear ON_MOVE value, works as a kind of lock. */
per_cpu(mem->stat->count[MEM_CGROUP_ON_MOVE], cpu) = 0;
spin_unlock(&mem->pcp_counter_lock);
@@ -3892,9 +3917,9 @@ mem_cgroup_get_local_stat(struct mem_cgr
s->stat[MCS_RSS] += val * PAGE_SIZE;
val = mem_cgroup_read_stat(mem, MEM_CGROUP_STAT_FILE_MAPPED);
s->stat[MCS_FILE_MAPPED] += val * PAGE_SIZE;
- val = mem_cgroup_read_stat(mem, MEM_CGROUP_STAT_PGPGIN_COUNT);
+ val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_PGPGIN);
s->stat[MCS_PGPGIN] += val;
- val = mem_cgroup_read_stat(mem, MEM_CGROUP_STAT_PGPGOUT_COUNT);
+ val = mem_cgroup_read_events(mem, MEM_CGROUP_EVENTS_PGPGOUT);
s->stat[MCS_PGPGOUT] += val;
if (do_swap_account) {
val = mem_cgroup_read_stat(mem, MEM_CGROUP_STAT_SWAPOUT);
_
Patches currently in -mm which might be from hannes@xxxxxxxxxxx are
memcg-add-dirty-limits-to-mem_cgroup-use-native-word-to-represent-dirtyable-pages.patch
memcg-add-dirty-limits-to-mem_cgroup-catch-negative-per-cpu-sums-in-dirty-info.patch
memcg-break-out-event-counters-from-other-stats.patch
memcg-use-native-word-page-statistics-counters.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
