2012-05-25 06:51:02

by Hiroshi Shimamoto

[permalink] [raw]
Subject: [PATCH 1/3] sched: Make sched_feat_names const

From: Hiroshi Shimamoto <[email protected]>

The strings sched_feat_names are never changed.

Signed-off-by: Hiroshi Shimamoto <[email protected]>
---
kernel/sched/core.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index fdf8f98..e8e595a 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -142,7 +142,7 @@ const_debug unsigned int sysctl_sched_features =
#define SCHED_FEAT(name, enabled) \
#name ,

-static __read_mostly char *sched_feat_names[] = {
+static const char * const sched_feat_names[] = {
#include "features.h"
NULL
};
--
1.7.7.6


2012-05-25 06:51:15

by Hiroshi Shimamoto

[permalink] [raw]
Subject: [PATCH 2/3] sched: Remove the last NULL entry from sched_feat_names

From: Hiroshi Shimamoto <[email protected]>

No need to have the last NULL entry.

Signed-off-by: Hiroshi Shimamoto <[email protected]>
---
kernel/sched/core.c | 1 -
1 files changed, 0 insertions(+), 1 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index e8e595a..621b10c 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -144,7 +144,6 @@ const_debug unsigned int sysctl_sched_features =

static const char * const sched_feat_names[] = {
#include "features.h"
- NULL
};

#undef SCHED_FEAT
--
1.7.7.6

2012-05-25 06:51:27

by Hiroshi Shimamoto

[permalink] [raw]
Subject: [PATCH 3/3] sched/rt: Remove redundant check before push_rt_task()

From: Hiroshi Shimamoto <[email protected]>

Currently rq->rt.overloaded is checked twice in switched_to_rt().
Remove the first one which out of push_rt_task().

Signed-off-by: Hiroshi Shimamoto <[email protected]>
---
kernel/sched/rt.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index c5565c3..a601a70 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -1907,7 +1907,7 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p)
*/
if (p->on_rq && rq->curr != p) {
#ifdef CONFIG_SMP
- if (rq->rt.overloaded && push_rt_task(rq) &&
+ if (push_rt_task(rq) &&
/* Don't resched if we changed runqueues */
rq != task_rq(p))
check_resched = 0;
--
1.7.7.6

2012-05-25 09:40:13

by Peter Zijlstra

[permalink] [raw]
Subject: Re: [PATCH 3/3] sched/rt: Remove redundant check before push_rt_task()

On Fri, 2012-05-25 at 15:43 +0900, Hiroshi Shimamoto wrote:
> From: Hiroshi Shimamoto <[email protected]>
>
> Currently rq->rt.overloaded is checked twice in switched_to_rt().
> Remove the first one which out of push_rt_task().

Steven was this on purpose to avoid the call or can I apply this thing?

> Signed-off-by: Hiroshi Shimamoto <[email protected]>
> ---
> kernel/sched/rt.c | 2 +-
> 1 files changed, 1 insertions(+), 1 deletions(-)
>
> diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
> index c5565c3..a601a70 100644
> --- a/kernel/sched/rt.c
> +++ b/kernel/sched/rt.c
> @@ -1907,7 +1907,7 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p)
> */
> if (p->on_rq && rq->curr != p) {
> #ifdef CONFIG_SMP
> - if (rq->rt.overloaded && push_rt_task(rq) &&
> + if (push_rt_task(rq) &&
> /* Don't resched if we changed runqueues */
> rq != task_rq(p))
> check_resched = 0;


2012-05-25 13:08:54

by Steven Rostedt

[permalink] [raw]
Subject: Re: [PATCH 3/3] sched/rt: Remove redundant check before push_rt_task()

On Fri, 2012-05-25 at 11:39 +0200, Peter Zijlstra wrote:
> On Fri, 2012-05-25 at 15:43 +0900, Hiroshi Shimamoto wrote:
> > From: Hiroshi Shimamoto <[email protected]>
> >
> > Currently rq->rt.overloaded is checked twice in switched_to_rt().
> > Remove the first one which out of push_rt_task().
>
> Steven was this on purpose to avoid the call or can I apply this thing?
>

Yeah, I think I added that as a way to avoid the call, but I never did
any benchmarks to see if it was actually better (extra check to avoid
the call, or just do the call and remove the duplicate check).

It would be a good idea to see how ofter that check happens in the real
world. As two or more RT tasks on the same CPU is only something that
happens on a real RT system (as the common case that is).

It's a micro optimization, as a double duplicate check should be fast as
everything is nicely in the cache line. But an extra call may be
expensive. But it's not a big difference to me if you take the patch or
not.

> > Signed-off-by: Hiroshi Shimamoto <[email protected]>
> > ---
> > kernel/sched/rt.c | 2 +-
> > 1 files changed, 1 insertions(+), 1 deletions(-)
> >
> > diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
> > index c5565c3..a601a70 100644
> > --- a/kernel/sched/rt.c
> > +++ b/kernel/sched/rt.c
> > @@ -1907,7 +1907,7 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p)
> > */
> > if (p->on_rq && rq->curr != p) {
> > #ifdef CONFIG_SMP
> > - if (rq->rt.overloaded && push_rt_task(rq) &&
> > + if (push_rt_task(rq) &&
> > /* Don't resched if we changed runqueues */
> > rq != task_rq(p))

If you decide to take this patch, it may be better to switch it to:

/* Don't resched if we changed runqueues */
if (push_rt_task(rq) && rq != task_rq(p))

-- Steve



> > check_resched = 0;
>
>

2012-05-30 13:41:47

by Hiroshi Shimamoto

[permalink] [raw]
Subject: [tip:sched/urgent] sched: Make sched_feat_names const

Commit-ID: 1292531f6f27af909e713671dd9cc3bcab8114b7
Gitweb: http://git.kernel.org/tip/1292531f6f27af909e713671dd9cc3bcab8114b7
Author: Hiroshi Shimamoto <[email protected]>
AuthorDate: Fri, 25 May 2012 15:41:54 +0900
Committer: Ingo Molnar <[email protected]>
CommitDate: Wed, 30 May 2012 14:02:26 +0200

sched: Make sched_feat_names const

The strings sched_feat_names are never changed.

Signed-off-by: Hiroshi Shimamoto <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>
---
kernel/sched/core.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 70cc36a..c1679a0 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -142,7 +142,7 @@ const_debug unsigned int sysctl_sched_features =
#define SCHED_FEAT(name, enabled) \
#name ,

-static __read_mostly char *sched_feat_names[] = {
+static const char * const sched_feat_names[] = {
#include "features.h"
NULL
};

2012-05-30 13:42:37

by Hiroshi Shimamoto

[permalink] [raw]
Subject: [tip:sched/urgent] sched: Remove the last NULL entry from sched_feat_names

Commit-ID: 7997a456ef841bb78eb6f881d7cc2c17c2f9b35e
Gitweb: http://git.kernel.org/tip/7997a456ef841bb78eb6f881d7cc2c17c2f9b35e
Author: Hiroshi Shimamoto <[email protected]>
AuthorDate: Fri, 25 May 2012 15:42:47 +0900
Committer: Ingo Molnar <[email protected]>
CommitDate: Wed, 30 May 2012 14:02:27 +0200

sched: Remove the last NULL entry from sched_feat_names

No need to have the last NULL entry.

Signed-off-by: Hiroshi Shimamoto <[email protected]>
Signed-off-by: Peter Zijlstra <[email protected]>
Link: http://lkml.kernel.org/r/[email protected]
Signed-off-by: Ingo Molnar <[email protected]>
---
kernel/sched/core.c | 1 -
1 files changed, 0 insertions(+), 1 deletions(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index c1679a0..94d598a 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -144,7 +144,6 @@ const_debug unsigned int sysctl_sched_features =

static const char * const sched_feat_names[] = {
#include "features.h"
- NULL
};

#undef SCHED_FEAT