From: Andi Kleen
cpusets has settings per cpu sets to enable NUMA node
interleaving for the slab cache or for the file cache.
These are quite useful, especially the setting for interleaving
the file page cache. This avoids the problem that some
program doing IO fills up a node completely and
From: Andi Kleen a...@linux.intel.com
cpusets has settings per cpu sets to enable NUMA node
interleaving for the slab cache or for the file cache.
These are quite useful, especially the setting for interleaving
the file page cache. This avoids the problem that some
program doing IO fills up a
On Tue, Nov 19, 2013 at 4:49 PM, Andi Kleen wrote:
> Michal Hocko writes:
>>
>> Another option would be to use sysctl values for the top cpuset as a
>> default. But then why not just do it manually without sysctl?
>
> I want to provide an alternative to having to use cpusets to use this,
> that
Michal Hocko writes:
>
> Another option would be to use sysctl values for the top cpuset as a
> default. But then why not just do it manually without sysctl?
I want to provide an alternative to having to use cpusets to use this,
that is actually usable for normal people.
Also this is really a
On Tue 19-11-13 12:13:33, Andi Kleen wrote:
> On Tue, Nov 19, 2013 at 08:11:35PM +0100, Michal Hocko wrote:
> > On Tue 19-11-13 19:42:00, Andi Kleen wrote:
> > > On Tue, Nov 19, 2013 at 11:42:03AM +0100, Michal Hocko wrote:
> > > > On Mon 18-11-13 16:50:22, Andi Kleen wrote:
> > > > [...]
> > > >
On Tue, Nov 19, 2013 at 08:11:35PM +0100, Michal Hocko wrote:
> On Tue 19-11-13 19:42:00, Andi Kleen wrote:
> > On Tue, Nov 19, 2013 at 11:42:03AM +0100, Michal Hocko wrote:
> > > On Mon 18-11-13 16:50:22, Andi Kleen wrote:
> > > [...]
> > > > diff --git a/include/linux/cpuset.h
On Tue 19-11-13 19:42:00, Andi Kleen wrote:
> On Tue, Nov 19, 2013 at 11:42:03AM +0100, Michal Hocko wrote:
> > On Mon 18-11-13 16:50:22, Andi Kleen wrote:
> > [...]
> > > diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
> > > index cc1b01c..10966f5 100644
> > > ---
On Tue, Nov 19, 2013 at 11:42:03AM +0100, Michal Hocko wrote:
> On Mon 18-11-13 16:50:22, Andi Kleen wrote:
> [...]
> > diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
> > index cc1b01c..10966f5 100644
> > --- a/include/linux/cpuset.h
> > +++ b/include/linux/cpuset.h
> > @@ -72,12
On Mon 18-11-13 16:50:22, Andi Kleen wrote:
[...]
> diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
> index cc1b01c..10966f5 100644
> --- a/include/linux/cpuset.h
> +++ b/include/linux/cpuset.h
> @@ -72,12 +72,14 @@ extern int cpuset_slab_spread_node(void);
>
> static inline int
On Mon 18-11-13 16:50:22, Andi Kleen wrote:
[...]
diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index cc1b01c..10966f5 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -72,12 +72,14 @@ extern int cpuset_slab_spread_node(void);
static inline int
On Tue, Nov 19, 2013 at 11:42:03AM +0100, Michal Hocko wrote:
On Mon 18-11-13 16:50:22, Andi Kleen wrote:
[...]
diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index cc1b01c..10966f5 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -72,12 +72,14 @@
On Tue 19-11-13 19:42:00, Andi Kleen wrote:
On Tue, Nov 19, 2013 at 11:42:03AM +0100, Michal Hocko wrote:
On Mon 18-11-13 16:50:22, Andi Kleen wrote:
[...]
diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index cc1b01c..10966f5 100644
--- a/include/linux/cpuset.h
+++
On Tue, Nov 19, 2013 at 08:11:35PM +0100, Michal Hocko wrote:
On Tue 19-11-13 19:42:00, Andi Kleen wrote:
On Tue, Nov 19, 2013 at 11:42:03AM +0100, Michal Hocko wrote:
On Mon 18-11-13 16:50:22, Andi Kleen wrote:
[...]
diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
On Tue 19-11-13 12:13:33, Andi Kleen wrote:
On Tue, Nov 19, 2013 at 08:11:35PM +0100, Michal Hocko wrote:
On Tue 19-11-13 19:42:00, Andi Kleen wrote:
On Tue, Nov 19, 2013 at 11:42:03AM +0100, Michal Hocko wrote:
On Mon 18-11-13 16:50:22, Andi Kleen wrote:
[...]
diff --git
Michal Hocko mho...@suse.cz writes:
Another option would be to use sysctl values for the top cpuset as a
default. But then why not just do it manually without sysctl?
I want to provide an alternative to having to use cpusets to use this,
that is actually usable for normal people.
Also this is
On Tue, Nov 19, 2013 at 4:49 PM, Andi Kleen a...@firstfloor.org wrote:
Michal Hocko mho...@suse.cz writes:
Another option would be to use sysctl values for the top cpuset as a
default. But then why not just do it manually without sysctl?
I want to provide an alternative to having to use
From: Andi Kleen
cpusets has settings per cpu sets to enable NUMA node
interleaving for the slab cache or for the file cache.
These are quite useful, especially the setting for interleaving
the file page cache. This avoids the problem that some
program doing IO fills up a node completely and
From: Andi Kleen a...@linux.intel.com
cpusets has settings per cpu sets to enable NUMA node
interleaving for the slab cache or for the file cache.
These are quite useful, especially the setting for interleaving
the file page cache. This avoids the problem that some
program doing IO fills up a
18 matches
Mail list logo
