Re: [PATCH 0/3] sched: newidle and RT wake-buddy fixes

["Gregory Haskins" <ghaskins@novell.com>]

>>> On Mon, Jun 30, 2008 at  9:15 AM, in message <20080630131511.GA7506@elte.hu>,
Ingo Molnar <mingo@elte.hu> wrote: 

> * Gregory Haskins <ghaskins@novell.com> wrote:
> 
>> Hi Ingo,
>>   The following patches apply to linux-tip/sched/devel and enhance the
>> performance of the kernel (specifically in PREEMPT_RT, though they do
>> not regress mainline performance as far as I can tell).  They offer
>> somewhere between 50-100% speedups in netperf performance, depending
>> on the test.
> 
> -tip testing found this boot hang:

I may have found the issue:  It looks like the hunk that initially disables interrupts in load_balance_newidle() was inadvertently applied to load_balance() instead during the
merge to linux-tip.  If you fold the following patch into my original patch, it should set
things right again.

-----

sched: fix merge problem with newidle enhancement patch

From: Gregory Haskins <ghaskins@novell.com>

commit cc8160c56843201891766660e3816d2e546c1b17 introduces a locking
enhancement for newidle.  However, one hunk misapplied to load_balance
instead of load_balance_newidle.  This fixes the issue.

Signed-off-by: Gregory Haskins <ghaskins@novell.com>
---

 kernel/sched.c |   18 +++++++++---------
 1 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/kernel/sched.c b/kernel/sched.c
index f35d73c..f36406f 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3459,15 +3459,6 @@ static int load_balance(int this_cpu, struct rq *this_rq,
 
 	cpus_setall(*cpus);
 
-	schedstat_inc(sd, lb_count[CPU_NEWLY_IDLE]);
-
-	/*
-	 * We are in a preempt-disabled section, so dropping the lock/irq
-	 * here simply means that other cores may acquire the lock,
-	 * and interrupts may occur.
-	 */
-	spin_unlock_irq(&this_rq->lock);
-
 	/*
 	 * When power savings policy is enabled for the parent domain, idle
 	 * sibling can pick up load irrespective of busy siblings. In this case,
@@ -3630,6 +3621,15 @@ load_balance_newidle(int this_cpu, struct rq *this_rq, struct sched_domain *sd,
 
 	cpus_setall(*cpus);
 
+	schedstat_inc(sd, lb_count[CPU_NEWLY_IDLE]);
+
+	/*
+	 * We are in a preempt-disabled section, so dropping the lock/irq
+	 * here simply means that other cores may acquire the lock,
+	 * and interrupts may occur.
+	 */
+	spin_unlock_irq(&this_rq->lock);
+
 	/*
 	 * When power savings policy is enabled for the parent domain, idle
 	 * sibling can pick up load irrespective of busy siblings. In this case,