linux-kernel - [PATCH]sched: cpuidle: Do not fetch current cpu in loop in idle thread

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Date:	Wed, 27 Apr 2016 09:54:41 +0000
From:	"Gaurav Jindal (Gaurav Jindal)" <Gaurav.Jindal@...eadtrum.com>
To:	"mingo@...hat.com" <mingo@...hat.com>,
	"peterz@...radead.com" <peterz@...radead.com>
CC:	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: [PATCH]sched: cpuidle: Do not fetch current cpu in loop in idle
 thread

Currently in idle thread loop, smp_processor_id() fetches the cpu id
every time it runs thus taking more cpu cycles.

Instead, save the cpu id as local varaible while entering the idle
thread and use it in the loop. Since it is local variable, it will be
correct for each core. It will also save cycles consumed while fetching cpu
id using smp_proceesor_id() in loop.

Patch:
------------------------------------------------------------------------------

diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c
index 1214f0a..9696f9e 100644
--- a/kernel/sched/idle.c
+++ b/kernel/sched/idle.c
@@ -185,6 +185,9 @@ exit_idle:
*/
static void cpu_idle_loop(void)
{
+       int cpu_id;
+
+       cpu_id = smp_processor_id();
        while(1) {
            /*
            *If the arch has a polling bit, we maintain an invariant:
    @@ -202,7 +205,7 @@ static void cpu_idle_loop(void)
            check_pgt_cache();
            rmb();

-       if (cpu_is_offline(smp_processor_id()))
+       if(cpu_is_offline(cpu_id))
                arch_cpu_idle_dead();

                local_irq_disable();
------------------------------------------------------------------------------

Signed-off-by: Gaurav Jindal<gaurav.jindal@...eadtrum.com>

Regards
Gaurav Jindal