[PATCH v2] perfcounters: clean up rotate_percpu

[PATCH v2] perfcounters: clean up rotate_percpu

[MinChan Kim]

Ingo implemented rotate function at 235c7fc7c500e4fd1700c4ad01b5612bcdc1b449.
I think he want to implement rotate_percpu but he don't implement it yet.

If we need not it immediately,
It would be better to remove it until we will need it again.

Signed-off-by: MinChan Kim <minchan.kim@gmail.com>
---
 kernel/perf_counter.c |   10 ----------
 1 files changed, 0 insertions(+), 10 deletions(-)

diff --git a/kernel/perf_counter.c b/kernel/perf_counter.c
index ad62965..211b13b 100644
--- a/kernel/perf_counter.c
+++ b/kernel/perf_counter.c
@@ -970,20 +970,10 @@ static void rotate_ctx(struct perf_counter_context *ctx)
 
 void perf_counter_task_tick(struct task_struct *curr, int cpu)
 {
-	struct perf_cpu_context *cpuctx = &per_cpu(perf_cpu_context, cpu);
 	struct perf_counter_context *ctx = &curr->perf_counter_ctx;
-	const int rotate_percpu = 0;
 
-	if (rotate_percpu)
-		perf_counter_cpu_sched_out(cpuctx);
 	perf_counter_task_sched_out(curr, cpu);
-
-	if (rotate_percpu)
-		rotate_ctx(&cpuctx->ctx);
 	rotate_ctx(ctx);
-
-	if (rotate_percpu)
-		perf_counter_cpu_sched_in(cpuctx, cpu);
 	perf_counter_task_sched_in(curr, cpu);
 }
 
-- 
1.5.4.3


-- 
Kinds Regards
MinChan Kim

Re: [PATCH v2] perfcounters: clean up rotate_percpu

[Peter Zijlstra]

On Wed, 2009-02-18 at 21:27 +0900, MinChan Kim wrote:
> Ingo implemented rotate function at 235c7fc7c500e4fd1700c4ad01b5612bcdc1b449.
> I think he want to implement rotate_percpu but he don't implement it yet.
> 
> If we need not it immediately,
> It would be better to remove it until we will need it again.

It seems to me, a more constructive patch would be to implement it for
him.

Re: [PATCH v2] perfcounters: clean up rotate_percpu

[MinChan Kim]

On Wed, Feb 18, 2009 at 9:31 PM, Peter Zijlstra <peterz@infradead.org> wrote:
> On Wed, 2009-02-18 at 21:27 +0900, MinChan Kim wrote:
>> Ingo implemented rotate function at 235c7fc7c500e4fd1700c4ad01b5612bcdc1b449.
>> I think he want to implement rotate_percpu but he don't implement it yet.
>>
>> If we need not it immediately,
>> It would be better to remove it until we will need it again.
>
> It seems to me, a more constructive patch would be to implement it for
> him.

If it is his intention, I don't care about this patch commit. :)

-- 
Kinds regards,
MinChan Kim