[V6,7/7] docs: mm: numaperf.rst Add brief description for access class 1.

Message ID 20191216153809.105463-8-Jonathan.Cameron@huawei.com
State New
Headers show
Series
  • [V6,1/7] ACPI: Support Generic Initiator only domains
Related show

Commit Message

Jonathan Cameron Dec. 16, 2019, 3:38 p.m.
Try to make minimal changes to the document which already describes
access class 0 in a generic fashion (including IO initiatiors that
are not CPUs).

Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

---
 Documentation/admin-guide/mm/numaperf.rst | 8 ++++++++
 1 file changed, 8 insertions(+)

-- 
2.19.1

Comments

Brice Goglin Dec. 18, 2019, 11:34 a.m. | #1
Le 16/12/2019 à 16:38, Jonathan Cameron a écrit :
> Try to make minimal changes to the document which already describes

> access class 0 in a generic fashion (including IO initiatiors that

> are not CPUs).

>

> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

> ---

>  Documentation/admin-guide/mm/numaperf.rst | 8 ++++++++

>  1 file changed, 8 insertions(+)

>

> diff --git a/Documentation/admin-guide/mm/numaperf.rst b/Documentation/admin-guide/mm/numaperf.rst

> index a80c3c37226e..327c0d72692d 100644

> --- a/Documentation/admin-guide/mm/numaperf.rst

> +++ b/Documentation/admin-guide/mm/numaperf.rst

> @@ -56,6 +56,11 @@ nodes' access characteristics share the same performance relative to other

>  linked initiator nodes. Each target within an initiator's access class,

>  though, do not necessarily perform the same as each other.

>  

> +The access class "1" is used to allow differentiation between initiators

> +that are CPUs and hence suitable for generic task scheduling, and

> +IO initiators such as GPUs and CPUs.  Unlike access class 0, only

> +nodes containing CPUs are considered.

> +

>  ================

>  NUMA Performance

>  ================

> @@ -88,6 +93,9 @@ The latency attributes are provided in nanoseconds.

>  The values reported here correspond to the rated latency and bandwidth

>  for the platform.

>  

> +Access class 0, takes the same form, but only includes values for CPU to

> +memory activity.



Shouldn't this be "class 1" here?

Both hunks look contradictory to me.

Brice
Jonathan Cameron Dec. 18, 2019, 2:37 p.m. | #2
On Wed, 18 Dec 2019 12:34:34 +0100
Brice Goglin <brice.goglin@gmail.com> wrote:

> Le 16/12/2019 à 16:38, Jonathan Cameron a écrit :

> > Try to make minimal changes to the document which already describes

> > access class 0 in a generic fashion (including IO initiatiors that

> > are not CPUs).

> >

> > Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>

> > ---

> >  Documentation/admin-guide/mm/numaperf.rst | 8 ++++++++

> >  1 file changed, 8 insertions(+)

> >

> > diff --git a/Documentation/admin-guide/mm/numaperf.rst b/Documentation/admin-guide/mm/numaperf.rst

> > index a80c3c37226e..327c0d72692d 100644

> > --- a/Documentation/admin-guide/mm/numaperf.rst

> > +++ b/Documentation/admin-guide/mm/numaperf.rst

> > @@ -56,6 +56,11 @@ nodes' access characteristics share the same performance relative to other

> >  linked initiator nodes. Each target within an initiator's access class,

> >  though, do not necessarily perform the same as each other.

> >  

> > +The access class "1" is used to allow differentiation between initiators

> > +that are CPUs and hence suitable for generic task scheduling, and

> > +IO initiators such as GPUs and CPUs.  Unlike access class 0, only

> > +nodes containing CPUs are considered.

> > +

> >  ================

> >  NUMA Performance

> >  ================

> > @@ -88,6 +93,9 @@ The latency attributes are provided in nanoseconds.

> >  The values reported here correspond to the rated latency and bandwidth

> >  for the platform.

> >  

> > +Access class 0, takes the same form, but only includes values for CPU to

> > +memory activity.  

> 

> 

> Shouldn't this be "class 1" here?

> 

Good point.

Jonathan

> Both hunks look contradictory to me.

> 

> Brice

> 

>

Patch

diff --git a/Documentation/admin-guide/mm/numaperf.rst b/Documentation/admin-guide/mm/numaperf.rst
index a80c3c37226e..327c0d72692d 100644
--- a/Documentation/admin-guide/mm/numaperf.rst
+++ b/Documentation/admin-guide/mm/numaperf.rst
@@ -56,6 +56,11 @@  nodes' access characteristics share the same performance relative to other
 linked initiator nodes. Each target within an initiator's access class,
 though, do not necessarily perform the same as each other.
 
+The access class "1" is used to allow differentiation between initiators
+that are CPUs and hence suitable for generic task scheduling, and
+IO initiators such as GPUs and CPUs.  Unlike access class 0, only
+nodes containing CPUs are considered.
+
 ================
 NUMA Performance
 ================
@@ -88,6 +93,9 @@  The latency attributes are provided in nanoseconds.
 The values reported here correspond to the rated latency and bandwidth
 for the platform.
 
+Access class 0, takes the same form, but only includes values for CPU to
+memory activity.
+
 ==========
 NUMA Cache
 ==========