[PATCH] Documentation/sysctl: Add timer_migration to kernel.rst

[PATCH] Documentation/sysctl: Add timer_migration to kernel.rst

[Phil Auld]

There is no mention of timer_migration in the docs. Add
a short description.

Signed-off-by: Phil Auld <pauld@redhat.com>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: linux-doc@vger.kernel.org
---
 Documentation/admin-guide/sysctl/kernel.rst | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/admin-guide/sysctl/kernel.rst
index b2b36d0c3094..e03691e2cf4a 100644
--- a/Documentation/admin-guide/sysctl/kernel.rst
+++ b/Documentation/admin-guide/sysctl/kernel.rst
@@ -1544,6 +1544,13 @@ constant ``FUTEX_TID_MASK`` (0x3fffffff).
 If a value outside of this range is written to ``threads-max`` an
 ``EINVAL`` error occurs.
 
+timer_migration
+===============
+
+When set, attempt to migrate timers away from idle cpus to allow them to remain
+in low power states longer.
+
+Default is set.
 
 traceoff_on_warning
 ===================
-- 
2.47.1

Re: [PATCH] Documentation/sysctl: Add timer_migration to kernel.rst

[Jonathan Corbet]

Phil Auld <pauld@redhat.com> writes:

> There is no mention of timer_migration in the docs. Add
> a short description.
>
> Signed-off-by: Phil Auld <pauld@redhat.com>
> Cc: Jonathan Corbet <corbet@lwn.net>
> Cc: linux-doc@vger.kernel.org
> ---
>  Documentation/admin-guide/sysctl/kernel.rst | 7 +++++++
>  1 file changed, 7 insertions(+)
>
> diff --git a/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/admin-guide/sysctl/kernel.rst
> index b2b36d0c3094..e03691e2cf4a 100644
> --- a/Documentation/admin-guide/sysctl/kernel.rst
> +++ b/Documentation/admin-guide/sysctl/kernel.rst
> @@ -1544,6 +1544,13 @@ constant ``FUTEX_TID_MASK`` (0x3fffffff).
>  If a value outside of this range is written to ``threads-max`` an
>  ``EINVAL`` error occurs.
>  
> +timer_migration
> +===============
> +
> +When set, attempt to migrate timers away from idle cpus to allow them to remain
> +in low power states longer.
> +
> +Default is set.

When set to *what*?  It will always be set to *some* value, right?  So
we should really say what the specific values mean.

Thanks,

jon

Re: [PATCH] Documentation/sysctl: Add timer_migration to kernel.rst

[Phil Auld]

On Tue, Jan 14, 2025 at 10:08:13AM -0700 Jonathan Corbet wrote:
> Phil Auld <pauld@redhat.com> writes:
> 
> > There is no mention of timer_migration in the docs. Add
> > a short description.
> >
> > Signed-off-by: Phil Auld <pauld@redhat.com>
> > Cc: Jonathan Corbet <corbet@lwn.net>
> > Cc: linux-doc@vger.kernel.org
> > ---
> >  Documentation/admin-guide/sysctl/kernel.rst | 7 +++++++
> >  1 file changed, 7 insertions(+)
> >
> > diff --git a/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/admin-guide/sysctl/kernel.rst
> > index b2b36d0c3094..e03691e2cf4a 100644
> > --- a/Documentation/admin-guide/sysctl/kernel.rst
> > +++ b/Documentation/admin-guide/sysctl/kernel.rst
> > @@ -1544,6 +1544,13 @@ constant ``FUTEX_TID_MASK`` (0x3fffffff).
> >  If a value outside of this range is written to ``threads-max`` an
> >  ``EINVAL`` error occurs.
> >  
> > +timer_migration
> > +===============
> > +
> > +When set, attempt to migrate timers away from idle cpus to allow them to remain
> > +in low power states longer.
> > +
> > +Default is set.
> 
> When set to *what*?  It will always be set to *some* value, right?  So
> we should really say what the specific values mean.
>


Well, it's boolean so set means 1 and unset is 0. I was using the same
style as the one below it:

"
traceoff_on_warning
===================

When set, disables tracing (see Documentation/trace/ftrace.rst) when a
``WARN()`` is hit.

"


But I can change it to "enabled (1)" or something if you prefer.


Cheers,
Phil


> Thanks,
> 
> jon
> 

-- 

Re: [PATCH] Documentation/sysctl: Add timer_migration to kernel.rst

[Jonathan Corbet]

Phil Auld <pauld@redhat.com> writes:

> On Tue, Jan 14, 2025 at 10:08:13AM -0700 Jonathan Corbet wrote:
>> Phil Auld <pauld@redhat.com> writes:
>> 
>> > There is no mention of timer_migration in the docs. Add
>> > a short description.
>> >
>> > Signed-off-by: Phil Auld <pauld@redhat.com>
>> > Cc: Jonathan Corbet <corbet@lwn.net>
>> > Cc: linux-doc@vger.kernel.org
>> > ---
>> >  Documentation/admin-guide/sysctl/kernel.rst | 7 +++++++
>> >  1 file changed, 7 insertions(+)
>> >
>> > diff --git a/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/admin-guide/sysctl/kernel.rst
>> > index b2b36d0c3094..e03691e2cf4a 100644
>> > --- a/Documentation/admin-guide/sysctl/kernel.rst
>> > +++ b/Documentation/admin-guide/sysctl/kernel.rst
>> > @@ -1544,6 +1544,13 @@ constant ``FUTEX_TID_MASK`` (0x3fffffff).
>> >  If a value outside of this range is written to ``threads-max`` an
>> >  ``EINVAL`` error occurs.
>> >  
>> > +timer_migration
>> > +===============
>> > +
>> > +When set, attempt to migrate timers away from idle cpus to allow them to remain
>> > +in low power states longer.
>> > +
>> > +Default is set.
>> 
>> When set to *what*?  It will always be set to *some* value, right?  So
>> we should really say what the specific values mean.
>>
>
>
> Well, it's boolean so set means 1 and unset is 0. I was using the same
> style as the one below it:
>
> "
> traceoff_on_warning
> ===================
>
> When set, disables tracing (see Documentation/trace/ftrace.rst) when a
> ``WARN()`` is hit.
>
> "

Just saying it's a boolean would help.  And our existing documentation
should rarely be taken as an example of what *good* documentation would
be... :)

> But I can change it to "enabled (1)" or something if you prefer.

"When set to a non-zero value" or some such would, I think, be a lot
more clear and make a better example for the next person to come along.

Get me an updated version and I can still queue it for 6.14.

Thanks,

jon

Re: [PATCH] Documentation/sysctl: Add timer_migration to kernel.rst

[Phil Auld]

On Tue, Jan 14, 2025 at 11:40:10AM -0700 Jonathan Corbet wrote:
> Phil Auld <pauld@redhat.com> writes:
> 
> > On Tue, Jan 14, 2025 at 10:08:13AM -0700 Jonathan Corbet wrote:
> >> Phil Auld <pauld@redhat.com> writes:
> >> 
> >> > There is no mention of timer_migration in the docs. Add
> >> > a short description.
> >> >
> >> > Signed-off-by: Phil Auld <pauld@redhat.com>
> >> > Cc: Jonathan Corbet <corbet@lwn.net>
> >> > Cc: linux-doc@vger.kernel.org
> >> > ---
> >> >  Documentation/admin-guide/sysctl/kernel.rst | 7 +++++++
> >> >  1 file changed, 7 insertions(+)
> >> >
> >> > diff --git a/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/admin-guide/sysctl/kernel.rst
> >> > index b2b36d0c3094..e03691e2cf4a 100644
> >> > --- a/Documentation/admin-guide/sysctl/kernel.rst
> >> > +++ b/Documentation/admin-guide/sysctl/kernel.rst
> >> > @@ -1544,6 +1544,13 @@ constant ``FUTEX_TID_MASK`` (0x3fffffff).
> >> >  If a value outside of this range is written to ``threads-max`` an
> >> >  ``EINVAL`` error occurs.
> >> >  
> >> > +timer_migration
> >> > +===============
> >> > +
> >> > +When set, attempt to migrate timers away from idle cpus to allow them to remain
> >> > +in low power states longer.
> >> > +
> >> > +Default is set.
> >> 
> >> When set to *what*?  It will always be set to *some* value, right?  So
> >> we should really say what the specific values mean.
> >>
> >
> >
> > Well, it's boolean so set means 1 and unset is 0. I was using the same
> > style as the one below it:
> >
> > "
> > traceoff_on_warning
> > ===================
> >
> > When set, disables tracing (see Documentation/trace/ftrace.rst) when a
> > ``WARN()`` is hit.
> >
> > "
> 
> Just saying it's a boolean would help.  And our existing documentation
> should rarely be taken as an example of what *good* documentation would
> be... :)

Fair enough :)

> 
> > But I can change it to "enabled (1)" or something if you prefer.
> 
> "When set to a non-zero value" or some such would, I think, be a lot
> more clear and make a better example for the next person to come along.
> 
> Get me an updated version and I can still queue it for 6.14.

Shortly.

Thanks,
Phil


> 
> Thanks,
> 
> jon
> 

-- 

Re: [PATCH] Documentation/sysctl: Add timer_migration to kernel.rst

[Phil Auld]

Hi Jon,

On Tue, Jan 14, 2025 at 11:40:10AM -0700 Jonathan Corbet wrote:
> Phil Auld <pauld@redhat.com> writes:
> 
> > On Tue, Jan 14, 2025 at 10:08:13AM -0700 Jonathan Corbet wrote:
> >> Phil Auld <pauld@redhat.com> writes:
> >> 
> >> > There is no mention of timer_migration in the docs. Add
> >> > a short description.
> >> >
> >> > Signed-off-by: Phil Auld <pauld@redhat.com>
> >> > Cc: Jonathan Corbet <corbet@lwn.net>
> >> > Cc: linux-doc@vger.kernel.org
> >> > ---
> >> >  Documentation/admin-guide/sysctl/kernel.rst | 7 +++++++
> >> >  1 file changed, 7 insertions(+)
> >> >
> >> > diff --git a/Documentation/admin-guide/sysctl/kernel.rst b/Documentation/admin-guide/sysctl/kernel.rst
> >> > index b2b36d0c3094..e03691e2cf4a 100644
> >> > --- a/Documentation/admin-guide/sysctl/kernel.rst
> >> > +++ b/Documentation/admin-guide/sysctl/kernel.rst
> >> > @@ -1544,6 +1544,13 @@ constant ``FUTEX_TID_MASK`` (0x3fffffff).
> >> >  If a value outside of this range is written to ``threads-max`` an
> >> >  ``EINVAL`` error occurs.
> >> >  
> >> > +timer_migration
> >> > +===============
> >> > +
> >> > +When set, attempt to migrate timers away from idle cpus to allow them to remain
> >> > +in low power states longer.
> >> > +
> >> > +Default is set.
> >> 
> >> When set to *what*?  It will always be set to *some* value, right?  So
> >> we should really say what the specific values mean.
> >>
> >
> >
> > Well, it's boolean so set means 1 and unset is 0. I was using the same
> > style as the one below it:
> >
> > "
> > traceoff_on_warning
> > ===================
> >
> > When set, disables tracing (see Documentation/trace/ftrace.rst) when a
> > ``WARN()`` is hit.
> >
> > "
> 
> Just saying it's a boolean would help.  And our existing documentation
> should rarely be taken as an example of what *good* documentation would
> be... :)
> 
> > But I can change it to "enabled (1)" or something if you prefer.
> 
> "When set to a non-zero value" or some such would, I think, be a lot
> more clear and make a better example for the next person to come along.
> 
> Get me an updated version and I can still queue it for 6.14.


It's sent but I failed to do it as a reply to this message.
Sorry about that.


Cheers,
Phil


> 
> Thanks,
> 
> jon
> 

-- 

Re: [PATCH] Documentation/sysctl: Add timer_migration to kernel.rst

[Jonathan Corbet]

Phil Auld <pauld@redhat.com> writes:

>> "When set to a non-zero value" or some such would, I think, be a lot
>> more clear and make a better example for the next person to come along.
>> 
>> Get me an updated version and I can still queue it for 6.14.
>
>
> It's sent but I failed to do it as a reply to this message.
> Sorry about that.

Doing a resend as a separate thread is actually better - no need to
apologize.

Thanks,

jon