[PATCH, take2] percpu_counter: FBC_BATCH might be too big

From: Eric Dumazet
Date: Sun Dec 07 2008 - 04:36:24 EST

In this second version I guarded hotcpu_notifier() call by

I wonder why hotcpu_notifier() is not a null op if !CONFIG_HOTPLUG_CPU

Thank you

[PATCH] percpu_counter: FBC_BATCH might be too big

For NR_CPUS >= 16 values, FBC_BATCH is 2*NR_CPUS

Considering more and more distros are using high NR_CPUS values,
it makes sense to use a more sensible value for FBC_BATCH.

A sensible value is 2*num_online_cpus(), with a minimum value of 32
(This minimum value helps branch prediction in __percpu_counter_add())

We already have a hotcpu notifier, so we can adjust FBC_BATCH dynamically.

Signed-off-by: Eric Dumazet <dada1@xxxxxxxxxxxxx>
Acked-by: David S. Miller <davem@xxxxxxxxxxxxx>
Acked-by: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
include/linux/percpu_counter.h | 6 +-----
lib/percpu_counter.c | 16 +++++++++++++++-
2 files changed, 16 insertions(+), 6 deletions(-)

diff --git a/include/linux/percpu_counter.h b/include/linux/percpu_counter.h
index 9007ccd..c42a184 100644
--- a/include/linux/percpu_counter.h
+++ b/include/linux/percpu_counter.h
@@ -24,11 +24,7 @@ struct percpu_counter {
s32 *counters;

-#if NR_CPUS >= 16
-#define FBC_BATCH (NR_CPUS*2)
-#define FBC_BATCH (NR_CPUS*4)
+extern int FBC_BATCH;

int percpu_counter_init(struct percpu_counter *fbc, s64 amount);
int percpu_counter_init_irq(struct percpu_counter *fbc, s64 amount);
diff --git a/lib/percpu_counter.c b/lib/percpu_counter.c
index a866389..e21ce7c 100644
--- a/lib/percpu_counter.c
+++ b/lib/percpu_counter.c
@@ -114,6 +114,16 @@ void percpu_counter_destroy(struct percpu_counter *fbc)

+int FBC_BATCH __read_mostly = 32;
+static void compute_batch_value(void)
+ int nr = num_online_cpus();
+ FBC_BATCH = max(32, nr*2);
static int __cpuinit percpu_counter_hotcpu_callback(struct notifier_block *nb,
unsigned long action, void *hcpu)
@@ -121,6 +131,7 @@ static int __cpuinit percpu_counter_hotcpu_callback(struct notifier_block *nb,
unsigned int cpu;
struct percpu_counter *fbc;

+ compute_batch_value();
if (action != CPU_DEAD)
return NOTIFY_OK;

@@ -139,11 +150,14 @@ static int __cpuinit percpu_counter_hotcpu_callback(struct notifier_block *nb,
return NOTIFY_OK;

static int __init percpu_counter_startup(void)
+ compute_batch_value();
hotcpu_notifier(percpu_counter_hotcpu_callback, 0);
return 0;
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/