* [PATCH] 2.6.14-rc4: wider use of for_each_*cpu() in net
@ 2005-10-13 19:01 hawkes
2005-10-28 16:02 ` Arnaldo Carvalho de Melo
0 siblings, 1 reply; 2+ messages in thread
From: hawkes @ 2005-10-13 19:01 UTC (permalink / raw)
To: netdev, linux-kernel; +Cc: hawkes
In 'net' change the explicit use of for-loops and NR_CPUS into the
general for_each_cpu() or for_each_online_cpu() constructs, as
appropriate. This widens the scope of potential future optimizations
of the general constructs, as well as takes advantage of the existing
optimizations of first_cpu() and next_cpu(), which is advantageous
when the true CPU count is much smaller than NR_CPUS.
Signed-off-by: John Hawkes <hawkes@sgi.com>
Index: linux/net/core/neighbour.c
===================================================================
--- linux.orig/net/core/neighbour.c 2005-10-13 09:29:54.000000000 -0700
+++ linux/net/core/neighbour.c 2005-10-13 09:30:31.000000000 -0700
@@ -1641,12 +1641,9 @@
memset(&ndst, 0, sizeof(ndst));
- for (cpu = 0; cpu < NR_CPUS; cpu++) {
+ for_each_cpu(cpu) {
struct neigh_statistics *st;
- if (!cpu_possible(cpu))
- continue;
-
st = per_cpu_ptr(tbl->stats, cpu);
ndst.ndts_allocs += st->allocs;
ndst.ndts_destroys += st->destroys;
Index: linux/net/core/pktgen.c
===================================================================
--- linux.orig/net/core/pktgen.c 2005-10-13 09:29:54.000000000 -0700
+++ linux/net/core/pktgen.c 2005-10-13 09:30:31.000000000 -0700
@@ -3065,12 +3065,9 @@
/* Register us to receive netdevice events */
register_netdevice_notifier(&pktgen_notifier_block);
- for (cpu = 0; cpu < NR_CPUS ; cpu++) {
+ for_each_online_cpu(cpu) {
char buf[30];
- if (!cpu_online(cpu))
- continue;
-
sprintf(buf, "kpktgend_%i", cpu);
pktgen_create_thread(buf, cpu);
}
Index: linux/net/ipv4/icmp.c
===================================================================
--- linux.orig/net/ipv4/icmp.c 2005-10-13 09:29:54.000000000 -0700
+++ linux/net/ipv4/icmp.c 2005-10-13 09:30:31.000000000 -0700
@@ -1108,12 +1108,9 @@
struct inet_sock *inet;
int i;
- for (i = 0; i < NR_CPUS; i++) {
+ for_each_cpu(i) {
int err;
- if (!cpu_possible(i))
- continue;
-
err = sock_create_kern(PF_INET, SOCK_RAW, IPPROTO_ICMP,
&per_cpu(__icmp_socket, i));
Index: linux/net/ipv4/proc.c
===================================================================
--- linux.orig/net/ipv4/proc.c 2005-10-13 09:29:54.000000000 -0700
+++ linux/net/ipv4/proc.c 2005-10-13 09:30:31.000000000 -0700
@@ -90,9 +90,7 @@
unsigned long res = 0;
int i;
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_possible(i))
- continue;
+ for_each_cpu(i) {
res += *(((unsigned long *) per_cpu_ptr(mib[0], i)) + offt);
res += *(((unsigned long *) per_cpu_ptr(mib[1], i)) + offt);
}
Index: linux/net/ipv6/proc.c
===================================================================
--- linux.orig/net/ipv6/proc.c 2005-10-13 09:29:54.000000000 -0700
+++ linux/net/ipv6/proc.c 2005-10-13 09:30:31.000000000 -0700
@@ -140,9 +140,7 @@
unsigned long res = 0;
int i;
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_possible(i))
- continue;
+ for_each_cpu(i) {
res += *(((unsigned long *)per_cpu_ptr(mib[0], i)) + offt);
res += *(((unsigned long *)per_cpu_ptr(mib[1], i)) + offt);
}
Index: linux/net/ipv6/icmp.c
===================================================================
--- linux.orig/net/ipv6/icmp.c 2005-10-13 09:29:54.000000000 -0700
+++ linux/net/ipv6/icmp.c 2005-10-13 09:30:31.000000000 -0700
@@ -700,10 +700,7 @@
struct sock *sk;
int err, i, j;
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_possible(i))
- continue;
-
+ for_each_cpu(i) {
err = sock_create_kern(PF_INET6, SOCK_RAW, IPPROTO_ICMPV6,
&per_cpu(__icmpv6_socket, i));
if (err < 0) {
@@ -749,9 +746,7 @@
{
int i;
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_possible(i))
- continue;
+ for_each_cpu(i) {
sock_release(per_cpu(__icmpv6_socket, i));
}
inet6_del_protocol(&icmpv6_protocol, IPPROTO_ICMPV6);
Index: linux/net/sctp/proc.c
===================================================================
--- linux.orig/net/sctp/proc.c 2005-10-13 09:29:54.000000000 -0700
+++ linux/net/sctp/proc.c 2005-10-13 09:30:31.000000000 -0700
@@ -69,9 +69,7 @@
unsigned long res = 0;
int i;
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_possible(i))
- continue;
+ for_each_cpu(i) {
res +=
*((unsigned long *) (((void *) per_cpu_ptr(mib[0], i)) +
sizeof (unsigned long) * nr));
^ permalink raw reply [flat|nested] 2+ messages in thread* Re: [PATCH] 2.6.14-rc4: wider use of for_each_*cpu() in net
2005-10-13 19:01 [PATCH] 2.6.14-rc4: wider use of for_each_*cpu() in net hawkes
@ 2005-10-28 16:02 ` Arnaldo Carvalho de Melo
0 siblings, 0 replies; 2+ messages in thread
From: Arnaldo Carvalho de Melo @ 2005-10-28 16:02 UTC (permalink / raw)
To: hawkes@sgi.com; +Cc: netdev, linux-kernel
On 10/13/05, hawkes@sgi.com <hawkes@sgi.com> wrote:
> In 'net' change the explicit use of for-loops and NR_CPUS into the
> general for_each_cpu() or for_each_online_cpu() constructs, as
> appropriate. This widens the scope of potential future optimizations
> of the general constructs, as well as takes advantage of the existing
> optimizations of first_cpu() and next_cpu(), which is advantageous
> when the true CPU count is much smaller than NR_CPUS.
>
> Signed-off-by: John Hawkes <hawkes@sgi.com>
Thanks, applied and pushed to Linus,
- Arnaldo
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2005-10-28 16:02 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-10-13 19:01 [PATCH] 2.6.14-rc4: wider use of for_each_*cpu() in net hawkes
2005-10-28 16:02 ` Arnaldo Carvalho de Melo
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox