The patch titled
sched: few scheduler event counters
has been removed from the -mm tree. Its filename was
mm-only-sched-add-a-few-scheduler-event-counters.patch
This patch was dropped because Suresh's patches trashed it, and I'm not sure anyone is using it much
------------------------------------------------------
Subject: sched: few scheduler event counters
From: Christoph Lameter <clameter@xxxxxxx>
Add a few counters to be able to see how the softirq affects the scheduler.
This patch should not go into mainline and maybe also not into mm.
(akpm: heh. Bad boy)
New counters in /proc/vmstat
sched_tick_balance Ticks that required invokation of the softirq
sched_tick_no_balance Ticks that did not require the softirq
sched_load_balance # invocations of load_balance().
sched_balance_softirq # of softirq executions
sched_balance_running # of times that serialization prevented
load balancing of a sched_domain.
Signed-off-by: Christoph Lameter <clameter@xxxxxxx>
Cc: Peter Williams <pwil3058@xxxxxxxxxxxxxx>
Cc: Nick Piggin <nickpiggin@xxxxxxxxxxxx>
Cc: Christoph Lameter <clameter@xxxxxxx>
Cc: "Siddha, Suresh B" <suresh.b.siddha@xxxxxxxxx>
Cc: "Chen, Kenneth W" <kenneth.w.chen@xxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxx>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/vmstat.h | 2 ++
kernel/sched.c | 12 ++++++++++--
mm/vmstat.c | 5 +++++
3 files changed, 17 insertions(+), 2 deletions(-)
diff -puN include/linux/vmstat.h~mm-only-sched-add-a-few-scheduler-event-counters include/linux/vmstat.h
--- a/include/linux/vmstat.h~mm-only-sched-add-a-few-scheduler-event-counters
+++ a/include/linux/vmstat.h
@@ -37,6 +37,8 @@ enum vm_event_item { PGPGIN, PGPGOUT, PS
FOR_ALL_ZONES(PGSCAN_DIRECT),
PGINODESTEAL, SLABS_SCANNED, KSWAPD_STEAL, KSWAPD_INODESTEAL,
PAGEOUTRUN, ALLOCSTALL, PGROTATED,
+ SCHED_TICK_NO_BALANCE, SCHED_TICK_BALANCE, SCHED_BALANCE_SOFTIRQ,
+ SCHED_LOAD_BALANCE, SCHED_BALANCE_RUNNING,
NR_VM_EVENT_ITEMS
};
diff -puN kernel/sched.c~mm-only-sched-add-a-few-scheduler-event-counters kernel/sched.c
--- a/kernel/sched.c~mm-only-sched-add-a-few-scheduler-event-counters
+++ a/kernel/sched.c
@@ -2953,6 +2953,8 @@ static void run_rebalance_domains(struct
/* Earliest time when we have to call run_rebalance_domains again */
unsigned long next_balance = jiffies + 60*HZ;
+ __count_vm_event(SCHED_BALANCE_SOFTIRQ);
+
for_each_domain(this_cpu, sd) {
if (!(sd->flags & SD_LOAD_BALANCE))
continue;
@@ -2967,11 +2969,14 @@ static void run_rebalance_domains(struct
interval = 1;
if (sd->flags & SD_SERIALIZE) {
- if (!spin_trylock(&balancing))
+ if (!spin_trylock(&balancing)) {
+ __count_vm_event(SCHED_BALANCE_RUNNING);
goto out;
+ }
}
if (time_after_eq(jiffies, sd->last_balance + interval)) {
+ __count_vm_event(SCHED_LOAD_BALANCE);
if (load_balance(this_cpu, this_rq, sd, idle, &balance)) {
/*
* We've pulled tasks over so either we're no
@@ -3247,8 +3252,11 @@ void scheduler_tick(void)
task_running_tick(rq, p);
#ifdef CONFIG_SMP
update_load(rq);
- if (time_after_eq(jiffies, rq->next_balance))
+ if (time_after_eq(jiffies, rq->next_balance)) {
+ __count_vm_event(SCHED_TICK_BALANCE);
raise_softirq(SCHED_SOFTIRQ);
+ }
+ __count_vm_event(SCHED_TICK_NO_BALANCE);
#endif
}
diff -puN mm/vmstat.c~mm-only-sched-add-a-few-scheduler-event-counters mm/vmstat.c
--- a/mm/vmstat.c~mm-only-sched-add-a-few-scheduler-event-counters
+++ a/mm/vmstat.c
@@ -484,6 +484,11 @@ static const char * const vmstat_text[]
"allocstall",
"pgrotated",
+ "sched_tick_no_balance",
+ "sched_tick_balance",
+ "sched_balance_softirq",
+ "sched_load_balance",
+ "sched_balance_running",
#endif
};
_
Patches currently in -mm which might be from clameter@xxxxxxx are
use-zvc-counters-to-establish-exact-size-of-dirtyable-pages.patch
use-zvc-counters-to-establish-exact-size-of-dirtyable-pages-fix.patch
make-try_to_unmap-return-a-special-exit-code.patch
add-nr_mlock-zvc.patch
add-pagemlocked-page-state-bit-and-lru-infrastructure.patch
add-pagemlocked-page-state-bit-and-lru-infrastructure-fix.patch
logic-to-move-mlocked-pages.patch
consolidate-new-anonymous-page-code-paths.patch
avoid-putting-new-mlocked-anonymous-pages-on-lru.patch
opportunistically-move-mlocked-pages-off-the-lru.patch
smaps-extract-pmd-walker-from-smaps-code.patch
smaps-add-pages-referenced-count-to-smaps.patch
smaps-add-clear_refs-file-to-clear-reference.patch
smaps-add-clear_refs-file-to-clear-reference-fix.patch
smaps-add-clear_refs-file-to-clear-reference-fix-fix.patch
slab-shutdown-cache_reaper-when-cpu-goes-down.patch
mm-only-sched-add-a-few-scheduler-event-counters.patch
mm-implement-swap-prefetching-vs-zvc-stuff.patch
mm-implement-swap-prefetching-vs-zvc-stuff-2.patch
zvc-support-nr_slab_reclaimable--nr_slab_unreclaimable-swap_prefetch.patch
reduce-max_nr_zones-swap_prefetch-remove-incorrect-use-of-zone_highmem.patch
numa-add-zone_to_nid-function-swap_prefetch.patch
remove-uses-of-kmem_cache_t-from-mm-and-include-linux-slabh-prefetch.patch
readahead-state-based-method-aging-accounting.patch
readahead-state-based-method-aging-accounting-vs-zvc-changes.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
