* [PATCH] sched/nohz: fix affine unpinned timers mess
@ 2016-05-04 6:45 Wanpeng Li
2016-05-12 10:31 ` [tip:sched/core] sched/nohz: Fix " tip-bot for Wanpeng Li
0 siblings, 1 reply; 2+ messages in thread
From: Wanpeng Li @ 2016-05-04 6:45 UTC (permalink / raw)
To: linux-kernel
Cc: Wanpeng Li, Frederic Weisbecker, Peter Zijlstra, Thomas Gleixner,
Ingo Molnar
From: Wanpeng Li <wanpeng.li@hotmail.com>
'commit 9642d18eee2c ("nohz: Affine unpinned timers to housekeepers")'
intended to affine unpinned timers to housekeepers:
unpinned timers(full dynaticks, idle) => nearest busy housekeepers(otherwise, fallback to any housekeepers)
unpinned timers(full dynaticks, busy) => nearest busy housekeepers(otherwise, fallback to any housekeepers)
unpinned timers(houserkeepers, idle) => nearest busy housekeepers(otherwise, fallback to itself)
However, the !idle_cpu(i) && is_housekeeping_cpu(cpu) check modify the
intention to:
unpinned timers(full dynaticks, idle) => any housekeepers(no mattter cpu topology)
unpinned timers(full dynaticks, busy) => any housekeepers(no mattter cpu topology)
unpinned timers(housekeepers, idle) => any busy cpus(otherwise, fallback to any housekeepers)
This patch fix it by checking if there is busy housekeepers nearby,
otherwise fallback to any housekeepers/itself, after the patch:
unpinned timers(full dynaticks, idle) => nearest busy housekeepers(otherwise, fallback to any housekeepers)
unpinned timers(full dynaticks, busy) => nearest busy housekeepers(otherwise, fallback to any housekeepers)
unpinned timers(housekeepers, idle) => nearest busy housekeepers(otherwise, fallback to itself)
Fixes: 'commit 9642d18eee2c ("nohz: Affine unpinned timers to housekeepers")'
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
---
kernel/sched/core.c | 5 ++++-
1 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index adcafda..5538761 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -499,7 +499,10 @@ int get_nohz_timer_target(void)
rcu_read_lock();
for_each_domain(cpu, sd) {
for_each_cpu(i, sched_domain_span(sd)) {
- if (!idle_cpu(i) && is_housekeeping_cpu(cpu)) {
+ if (cpu == i)
+ continue;
+
+ if (!idle_cpu(i) && is_housekeeping_cpu(i)) {
cpu = i;
goto unlock;
}
--
1.7.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
* [tip:sched/core] sched/nohz: Fix affine unpinned timers mess
2016-05-04 6:45 [PATCH] sched/nohz: fix affine unpinned timers mess Wanpeng Li
@ 2016-05-12 10:31 ` tip-bot for Wanpeng Li
0 siblings, 0 replies; 2+ messages in thread
From: tip-bot for Wanpeng Li @ 2016-05-12 10:31 UTC (permalink / raw)
To: linux-tip-commits
Cc: fweisbec, peterz, efault, tglx, torvalds, linux-kernel, mingo,
hpa, wanpeng.li
Commit-ID: 444969223c81c7d0a95136b7b4cfdcfbc96ac5bd
Gitweb: http://git.kernel.org/tip/444969223c81c7d0a95136b7b4cfdcfbc96ac5bd
Author: Wanpeng Li <wanpeng.li@hotmail.com>
AuthorDate: Wed, 4 May 2016 14:45:34 +0800
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Thu, 12 May 2016 09:55:32 +0200
sched/nohz: Fix affine unpinned timers mess
The following commit:
9642d18eee2c ("nohz: Affine unpinned timers to housekeepers")'
intended to affine unpinned timers to housekeepers:
unpinned timers(full dynaticks, idle) => nearest busy housekeepers(otherwise, fallback to any housekeepers)
unpinned timers(full dynaticks, busy) => nearest busy housekeepers(otherwise, fallback to any housekeepers)
unpinned timers(houserkeepers, idle) => nearest busy housekeepers(otherwise, fallback to itself)
However, the !idle_cpu(i) && is_housekeeping_cpu(cpu) check modified the
intention to:
unpinned timers(full dynaticks, idle) => any housekeepers(no mattter cpu topology)
unpinned timers(full dynaticks, busy) => any housekeepers(no mattter cpu topology)
unpinned timers(housekeepers, idle) => any busy cpus(otherwise, fallback to any housekeepers)
This patch fixes it by checking if there are busy housekeepers nearby,
otherwise falls to any housekeepers/itself. After the patch:
unpinned timers(full dynaticks, idle) => nearest busy housekeepers(otherwise, fallback to any housekeepers)
unpinned timers(full dynaticks, busy) => nearest busy housekeepers(otherwise, fallback to any housekeepers)
unpinned timers(housekeepers, idle) => nearest busy housekeepers(otherwise, fallback to itself)
Signed-off-by: Wanpeng Li <wanpeng.li@hotmail.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
[ Fixed the changelog. ]
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org
Fixes: 'commit 9642d18eee2c ("nohz: Affine unpinned timers to housekeepers")'
Link: http://lkml.kernel.org/r/1462344334-8303-1-git-send-email-wanpeng.li@hotmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
kernel/sched/core.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 636c4b9..6f6962a 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -533,7 +533,10 @@ int get_nohz_timer_target(void)
rcu_read_lock();
for_each_domain(cpu, sd) {
for_each_cpu(i, sched_domain_span(sd)) {
- if (!idle_cpu(i) && is_housekeeping_cpu(cpu)) {
+ if (cpu == i)
+ continue;
+
+ if (!idle_cpu(i) && is_housekeeping_cpu(i)) {
cpu = i;
goto unlock;
}
^ permalink raw reply related [flat|nested] 2+ messages in thread
end of thread, other threads:[~2016-05-12 10:31 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-05-04 6:45 [PATCH] sched/nohz: fix affine unpinned timers mess Wanpeng Li
2016-05-12 10:31 ` [tip:sched/core] sched/nohz: Fix " tip-bot for Wanpeng Li
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).