public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Rusty Russell <rusty@rustcorp.com.au>
To: James Bottomley <James.Bottomley@steeleye.com>
Cc: linux-kernel@vger.kernel.org, mingo@redhat.com
Subject: Re: Optimisation for smp_num_cpus loop in hotplug
Date: Sat, 22 Jun 2002 01:14:46 +1000	[thread overview]
Message-ID: <E17LQ7b-0000Iz-00@wagner.rustcorp.com.au> (raw)
In-Reply-To: Your message of "Thu, 20 Jun 2002 19:41:52 -0400." <200206202341.g5KNfqU07377@localhost.localdomain>

In message <200206202341.g5KNfqU07377@localhost.localdomain> you write:
> The hotplug CPU patch introduces this to replace the loop over all active CPUs 
> abstraction:
> 
> 	for (i = 0; i < NR_CPUS; i++) {
> 		if (!cpu_online(i))
> 			continue;

Yeah, it's simple, and none of the current ones are really critical.
But I think we're better off with:
	for (i = first_cpu(); i < NR_CPUS; i = next_cpu(i)) {

Which is simple enough not to need an iterator macro, and also has the
bonus of giving irq-balancing et al. an efficient, portable way of
looking for the "next" cpu.

Cheers!
Rusty.
--
  Anyone who quotes me in their sig is an idiot. -- Rusty Russell.

Name: CPU iterator patch
Author: Rusty Russell
Status: Trivial

D: This patch adds first_cpu() & next_cpu(cpu) for more efficient cpu
D: iteration.

diff -urN -I \$.*\$ --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal linux-2.5.24/include/asm-i386/smp.h working-2.5.24-cpu-iter/include/asm-i386/smp.h
--- linux-2.5.24/include/asm-i386/smp.h	Thu Jun 20 01:28:51 2002
+++ working-2.5.24-cpu-iter/include/asm-i386/smp.h	Sat Jun 22 00:55:46 2002
@@ -107,6 +107,18 @@
 	return -1;
 }
 
+/* x86 is always linear, ie. cpu_online(0) always true at the moment. */
+static inline unsigned int first_cpu(void)
+{
+	return 0;
+}
+
+/* Return "next" online cpu.  Returns NR_CPUS on end. */
+static inline unsigned int next_cpu(unsigned int cpu)
+{
+	return find_next_bit(&cpu_online_map, NR_CPUS, cpu+1);
+}
+
 static __inline int hard_smp_processor_id(void)
 {
 	/* we don't want to mark this access volatile - bad code generation */
diff -urN -I \$.*\$ --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal linux-2.5.24/include/asm-ia64/smp.h working-2.5.24-cpu-iter/include/asm-ia64/smp.h
--- linux-2.5.24/include/asm-ia64/smp.h	Thu Jun 20 01:28:51 2002
+++ working-2.5.24-cpu-iter/include/asm-ia64/smp.h	Sat Jun 22 00:59:34 2002
@@ -45,6 +45,17 @@
 
 extern unsigned long ap_wakeup_vector;
 
+static inline unsigned int first_cpu(void)
+{
+	return __ffs(cpu_online_map);
+}
+
+/* Return "next" online cpu.  Returns NR_CPUS on end. */
+static inline unsigned int next_cpu(unsigned int cpu)
+{
+	return find_next_bit(&cpu_online_map, NR_CPUS, cpu+1);
+}
+
 #define cpu_online(cpu) (cpu_online_map & (1<<(cpu)))
 extern inline unsigned int num_online_cpus(void)
 {
diff -urN -I \$.*\$ --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal linux-2.5.24/include/asm-ppc/smp.h working-2.5.24-cpu-iter/include/asm-ppc/smp.h
--- linux-2.5.24/include/asm-ppc/smp.h	Thu Jun 20 01:28:51 2002
+++ working-2.5.24-cpu-iter/include/asm-ppc/smp.h	Sat Jun 22 00:58:37 2002
@@ -47,6 +47,17 @@
 
 #define smp_processor_id() (current_thread_info()->cpu)
 
+static inline unsigned int first_cpu(void)
+{
+	return __ffs(cpu_online_map);
+}
+
+/* Return "next" online cpu.  Returns NR_CPUS on end. */
+static inline unsigned int next_cpu(unsigned int cpu)
+{
+	return find_next_bit(&cpu_online_map, NR_CPUS, cpu+1);
+}
+
 #define cpu_online(cpu) (cpu_online_map & (1<<(cpu)))
 
 extern inline unsigned int num_online_cpus(void)
diff -urN -I \$.*\$ --exclude TAGS -X /home/rusty/devel/kernel/kernel-patches/current-dontdiff --minimal linux-2.5.24/include/asm-sparc64/smp.h working-2.5.24-cpu-iter/include/asm-sparc64/smp.h
--- linux-2.5.24/include/asm-sparc64/smp.h	Fri Jun 21 09:41:55 2002
+++ working-2.5.24-cpu-iter/include/asm-sparc64/smp.h	Sat Jun 22 01:06:58 2002
@@ -77,6 +77,17 @@
 	return -1;
 }
 
+static inline unsigned int first_cpu(void)
+{
+	return __ffs(cpu_online_map);
+}
+
+/* Return "next" online cpu.  Returns NR_CPUS on end. */
+static inline unsigned int next_cpu(unsigned int cpu)
+{
+	return find_next_bit(&cpu_online_map, NR_CPUS, cpu+1);
+}
+
 /*
  *	General functions that each host system must provide.
  */

  reply	other threads:[~2002-06-21 15:10 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2002-06-20 23:41 Optimisation for smp_num_cpus loop in hotplug James Bottomley
2002-06-21 15:14 ` Rusty Russell [this message]
2002-06-21 15:31   ` James Bottomley
2002-06-21 19:17     ` Rusty Russell
     [not found] <mailman.1024617156.25656.linux-kernel2news@redhat.com>
2002-06-21  4:16 ` Pete Zaitcev
2002-06-21 13:10   ` James Bottomley

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=E17LQ7b-0000Iz-00@wagner.rustcorp.com.au \
    --to=rusty@rustcorp.com.au \
    --cc=James.Bottomley@steeleye.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox