[PATCH] sched: rt: document the risk of small values in the bandwidth settings

[PATCH] sched: rt: document the risk of small values in the bandwidth settings

[Peter Zijlstra]

Thomas noted that we should disallow sysctl_sched_rt_runtime == 0 for (!
RT_GROUP) since the root group always has some RT tasks in it.

Further, update the documentation to inspire clue.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
---
 Documentation/scheduler/sched-rt-group.txt |   18 ++++++++++++++++++
 kernel/sched.c                             |    7 +++++++
 2 files changed, 25 insertions(+), 0 deletions(-)

diff --git a/Documentation/scheduler/sched-rt-group.txt b/Documentation/scheduler/sched-rt-group.txt
index 5ba4d3f..cfd51c0 100644
--- a/Documentation/scheduler/sched-rt-group.txt
+++ b/Documentation/scheduler/sched-rt-group.txt
@@ -4,6 +4,7 @@
 CONTENTS
 ========
 
+0. WARNING
 1. Overview
   1.1 The problem
   1.2 The solution
@@ -14,6 +15,23 @@ CONTENTS
 3. Future plans
 
 
+0. WARNING
+==========
+
+Fiddling with these settings can result in an unusable system, the knobs are
+root only and assume root knows what he's on about.
+
+Most notable:
+ 
+ * very small values in sched_rt_period_us can result in an unstable
+   system when the period is smaller than either the available hrtimer
+   resolution, or the time it takes to handle the budget refresh itself.
+
+ * very small values in sched_rt_runtime_us can result in an unstable
+   system when the runtime is so small the system has difficulty making
+   forward progress (NOTE: the migration thread and kstopmachine both
+   are real-time processes).
+
 1. Overview
 ===========
 
diff --git a/kernel/sched.c b/kernel/sched.c
index 11f6da4..c5ef114 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -10021,6 +10021,13 @@ static int sched_rt_global_constraints(void)
 	if (sysctl_sched_rt_period <= 0)
 		return -EINVAL;
 
+	/*
+	 * There's always some RT tasks in the root group
+	 * -- migration, kstopmachine etc..
+	 */
+	if (sysctl_sched_rt_runtime == 0)
+		return -EBUSY;
+
 	spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags);
 	for_each_possible_cpu(i) {
 		struct rt_rq *rt_rq = &cpu_rq(i)->rt;

Re: [PATCH] sched: rt: document the risk of small values in the bandwidth settings

[Randy Dunlap]

Peter Zijlstra wrote:
> Thomas noted that we should disallow sysctl_sched_rt_runtime == 0 for (!
> RT_GROUP) since the root group always has some RT tasks in it.
> 
> Further, update the documentation to inspire clue.
> 
> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
> Acked-by: Thomas Gleixner <tglx@linutronix.de>
> ---
>  Documentation/scheduler/sched-rt-group.txt |   18 ++++++++++++++++++
>  kernel/sched.c                             |    7 +++++++
>  2 files changed, 25 insertions(+), 0 deletions(-)
> 
> diff --git a/Documentation/scheduler/sched-rt-group.txt b/Documentation/scheduler/sched-rt-group.txt
> index 5ba4d3f..cfd51c0 100644
> --- a/Documentation/scheduler/sched-rt-group.txt
> +++ b/Documentation/scheduler/sched-rt-group.txt
> @@ -4,6 +4,7 @@
>  CONTENTS
>  ========
>  
> +0. WARNING
>  1. Overview
>    1.1 The problem
>    1.2 The solution
> @@ -14,6 +15,23 @@ CONTENTS
>  3. Future plans
>  
>  
> +0. WARNING
> +==========
> +
> +Fiddling with these settings can result in an unusable system, the knobs are
> +root only and assume root knows what he's on about.

I know that my English is limited, but I don't know what this means:
  "what he's on about."


> +
> +Most notable:
> + 
> + * very small values in sched_rt_period_us can result in an unstable
> +   system when the period is smaller than either the available hrtimer
> +   resolution, or the time it takes to handle the budget refresh itself.
> +
> + * very small values in sched_rt_runtime_us can result in an unstable
> +   system when the runtime is so small the system has difficulty making
> +   forward progress (NOTE: the migration thread and kstopmachine both
> +   are real-time processes).
> +
>  1. Overview
>  ===========
>  
> diff --git a/kernel/sched.c b/kernel/sched.c
> index 11f6da4..c5ef114 100644
> --- a/kernel/sched.c
> +++ b/kernel/sched.c
> @@ -10021,6 +10021,13 @@ static int sched_rt_global_constraints(void)
>  	if (sysctl_sched_rt_period <= 0)
>  		return -EINVAL;
>  
> +	/*
> +	 * There's always some RT tasks in the root group
> +	 * -- migration, kstopmachine etc..
> +	 */
> +	if (sysctl_sched_rt_runtime == 0)
> +		return -EBUSY;
> +
>  	spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags);
>  	for_each_possible_cpu(i) {
>  		struct rt_rq *rt_rq = &cpu_rq(i)->rt;
> 
> --


-- 
~Randy

Re: [PATCH] sched: rt: document the risk of small values in the bandwidth settings

[Peter Zijlstra]

On Mon, 2009-05-04 at 08:34 -0700, Randy Dunlap wrote:
> > +Fiddling with these settings can result in an unusable system, the knobs are
> > +root only and assume root knows what he's on about.
> 
> I know that my English is limited, but I don't know what this means:
>   "what he's on about."

Hehe, OK, how about we write it like:

Fiddling with these settings can result in an unstable system, the knobs are
root only and assumes root knows what he is doing.

Does that work better for you?

Re: [PATCH] sched: rt: document the risk of small values in the bandwidth settings

[Randy Dunlap]

Peter Zijlstra wrote:
> On Mon, 2009-05-04 at 08:34 -0700, Randy Dunlap wrote:
>>> +Fiddling with these settings can result in an unusable system, the knobs are
>>> +root only and assume root knows what he's on about.
>> I know that my English is limited, but I don't know what this means:
>>   "what he's on about."
> 
> Hehe, OK, how about we write it like:
> 
> Fiddling with these settings can result in an unstable system, the knobs are
> root only and assumes root knows what he is doing.
> 
> Does that work better for you?

Yes, thanks.

-- 
~Randy