Optimize various per cpu area operations through these new percpu
operations. These operations avoid address calculations through the
use of segment prefixes and multiple memory references through RMW
instructions etc.
Reduces code size:
Before:
christoph@linux-2.6$ size fs/buffer.o
text data bss dec hex filename
19169 80 28 19277 4b4d fs/buffer.o
After:
christoph@linux-2.6$ size fs/buffer.o
text data bss dec hex filename
19138 80 28 19246 4b2e fs/buffer.o
V3->V4:
- Move the use of this_cpu_inc_return into a later patch so that
this one can go in without percpu infrastructure changes.
Cc: Wu Fengguang <fengguang.wu@intel.com>
Cc: Christoph Hellwig <hch@lst.de>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
static void bh_lru_install(struct buffer_head *bh)
{
struct buffer_head *evictee = NULL;
static void bh_lru_install(struct buffer_head *bh)
{
struct buffer_head *evictee = NULL;
check_irqs_on();
bh_lru_lock();
check_irqs_on();
bh_lru_lock();
- lru = &__get_cpu_var(bh_lrus);
- if (lru->bhs[0] != bh) {
+ if (__this_cpu_read(bh_lrus.bhs[0]) != bh) {
struct buffer_head *bhs[BH_LRU_SIZE];
int in;
int out = 0;
struct buffer_head *bhs[BH_LRU_SIZE];
int in;
int out = 0;
get_bh(bh);
bhs[out++] = bh;
for (in = 0; in < BH_LRU_SIZE; in++) {
get_bh(bh);
bhs[out++] = bh;
for (in = 0; in < BH_LRU_SIZE; in++) {
- struct buffer_head *bh2 = lru->bhs[in];
+ struct buffer_head *bh2 =
+ __this_cpu_read(bh_lrus.bhs[in]);
if (bh2 == bh) {
__brelse(bh2);
if (bh2 == bh) {
__brelse(bh2);
}
while (out < BH_LRU_SIZE)
bhs[out++] = NULL;
}
while (out < BH_LRU_SIZE)
bhs[out++] = NULL;
- memcpy(lru->bhs, bhs, sizeof(bhs));
+ memcpy(__this_cpu_ptr(&bh_lrus.bhs), bhs, sizeof(bhs));
lookup_bh_lru(struct block_device *bdev, sector_t block, unsigned size)
{
struct buffer_head *ret = NULL;
lookup_bh_lru(struct block_device *bdev, sector_t block, unsigned size)
{
struct buffer_head *ret = NULL;
unsigned int i;
check_irqs_on();
bh_lru_lock();
unsigned int i;
check_irqs_on();
bh_lru_lock();
- lru = &__get_cpu_var(bh_lrus);
for (i = 0; i < BH_LRU_SIZE; i++) {
for (i = 0; i < BH_LRU_SIZE; i++) {
- struct buffer_head *bh = lru->bhs[i];
+ struct buffer_head *bh = __this_cpu_read(bh_lrus.bhs[i]);
if (bh && bh->b_bdev == bdev &&
bh->b_blocknr == block && bh->b_size == size) {
if (i) {
while (i) {
if (bh && bh->b_bdev == bdev &&
bh->b_blocknr == block && bh->b_size == size) {
if (i) {
while (i) {
- lru->bhs[i] = lru->bhs[i - 1];
+ __this_cpu_write(bh_lrus.bhs[i],
+ __this_cpu_read(bh_lrus.bhs[i - 1]));
+ __this_cpu_write(bh_lrus.bhs[0], bh);
if (__get_cpu_var(bh_accounting).ratelimit++ < 4096)
return;
if (__get_cpu_var(bh_accounting).ratelimit++ < 4096)
return;
- __get_cpu_var(bh_accounting).ratelimit = 0;
+ __this_cpu_write(bh_accounting.ratelimit, 0);
for_each_online_cpu(i)
tot += per_cpu(bh_accounting, i).nr;
buffer_heads_over_limit = (tot > max_buffer_heads);
}
for_each_online_cpu(i)
tot += per_cpu(bh_accounting, i).nr;
buffer_heads_over_limit = (tot > max_buffer_heads);
}
struct buffer_head *alloc_buffer_head(gfp_t gfp_flags)
{
struct buffer_head *ret = kmem_cache_zalloc(bh_cachep, gfp_flags);
if (ret) {
INIT_LIST_HEAD(&ret->b_assoc_buffers);
struct buffer_head *alloc_buffer_head(gfp_t gfp_flags)
{
struct buffer_head *ret = kmem_cache_zalloc(bh_cachep, gfp_flags);
if (ret) {
INIT_LIST_HEAD(&ret->b_assoc_buffers);
- get_cpu_var(bh_accounting).nr++;
+ preempt_disable();
+ __this_cpu_inc(bh_accounting.nr);
- put_cpu_var(bh_accounting);
{
BUG_ON(!list_empty(&bh->b_assoc_buffers));
kmem_cache_free(bh_cachep, bh);
{
BUG_ON(!list_empty(&bh->b_assoc_buffers));
kmem_cache_free(bh_cachep, bh);
- get_cpu_var(bh_accounting).nr--;
+ preempt_disable();
+ __this_cpu_dec(bh_accounting.nr);
- put_cpu_var(bh_accounting);
}
EXPORT_SYMBOL(free_buffer_head);
}
EXPORT_SYMBOL(free_buffer_head);
brelse(b->bhs[i]);
b->bhs[i] = NULL;
}
brelse(b->bhs[i]);
b->bhs[i] = NULL;
}
- get_cpu_var(bh_accounting).nr += per_cpu(bh_accounting, cpu).nr;
+ this_cpu_add(bh_accounting.nr, per_cpu(bh_accounting, cpu).nr);
per_cpu(bh_accounting, cpu).nr = 0;
per_cpu(bh_accounting, cpu).nr = 0;
- put_cpu_var(bh_accounting);
}
static int buffer_cpu_notify(struct notifier_block *self,
}
static int buffer_cpu_notify(struct notifier_block *self,