[PATCH 1/4] cpumask: cpu_coregroup_mask(): x86

[PATCH 1/4] cpumask: cpu_coregroup_mask(): x86

[Rusty Russell]

Like cpu_coregroup_map, but returns a (const) pointer.

(This will go to Ingo separately as part of the x86 series, just
airing it here for thoroughness).

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Mike Travis <travis@sgi.com>
Cc: Ingo Molnar <mingo@redhat.com>
---
 arch/x86/include/asm/topology.h |    1 +
 arch/x86/kernel/smpboot.c       |   11 ++++++++---
 2 files changed, 9 insertions(+), 3 deletions(-)

diff -r 11a884ebd99b arch/x86/include/asm/topology.h
--- a/arch/x86/include/asm/topology.h	Thu Nov 20 13:54:24 2008 +1030
+++ b/arch/x86/include/asm/topology.h	Thu Nov 20 13:55:44 2008 +1030
@@ -231,6 +231,7 @@
 #endif
 
 extern cpumask_t cpu_coregroup_map(int cpu);
+extern const struct cpumask *cpu_coregroup_mask(int cpu);
 
 #ifdef ENABLE_TOPO_DEFINES
 #define topology_physical_package_id(cpu)	(cpu_data(cpu).phys_proc_id)
diff -r 11a884ebd99b arch/x86/kernel/smpboot.c
--- a/arch/x86/kernel/smpboot.c	Thu Nov 20 13:54:24 2008 +1030
+++ b/arch/x86/kernel/smpboot.c	Thu Nov 20 13:55:44 2008 +1030
@@ -497,7 +497,7 @@
 }
 
 /* maps the cpu to the sched domain representing multi-core */
-cpumask_t cpu_coregroup_map(int cpu)
+const struct cpumask *cpu_coregroup_mask(int cpu)
 {
 	struct cpuinfo_x86 *c = &cpu_data(cpu);
 	/*
@@ -505,9 +505,14 @@
 	 * And for power savings, we return cpu_core_map
 	 */
 	if (sched_mc_power_savings || sched_smt_power_savings)
-		return per_cpu(cpu_core_map, cpu);
+		return &per_cpu(cpu_core_map, cpu);
 	else
-		return c->llc_shared_map;
+		return &c->llc_shared_map;
+}
+
+cpumask_t cpu_coregroup_map(int cpu)
+{
+	return *cpu_coregroup_mask(cpu);
 }
 
 static void impress_friends(void)

Re: [PATCH 1/4] cpumask: cpu_coregroup_mask(): x86

[Mike Travis]

Rusty Russell wrote:
> Like cpu_coregroup_map, but returns a (const) pointer.
> 
> (This will go to Ingo separately as part of the x86 series, just
> airing it here for thoroughness).
> 
> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
> Signed-off-by: Mike Travis <travis@sgi.com>
> Cc: Ingo Molnar <mingo@redhat.com>

I've pulled this patch into the queue for my cpus4096-for-ingo tree.

	[PATCH 1/4] cpumask: cpu_coregroup_mask(): x86

I can also line up a queue for sched related changes:

	[PATCH 4/4] cpumask: Replace cpu_coregroup_map with cpu_coregroup_map()

Is there any status on the corresponding changes for sparc, s390?  (I assume
that they'll need to be merged into linux-next?)

	[PATCH 2/4] cpumask: cpu_coregroup_mask(): sparc
	[PATCH 3/4] cpumask: cpu_coregroup_mask(): s390

Thanks,
Mike

> ---
>  arch/x86/include/asm/topology.h |    1 +
>  arch/x86/kernel/smpboot.c       |   11 ++++++++---
>  2 files changed, 9 insertions(+), 3 deletions(-)
> 
> diff -r 11a884ebd99b arch/x86/include/asm/topology.h
> --- a/arch/x86/include/asm/topology.h	Thu Nov 20 13:54:24 2008 +1030
> +++ b/arch/x86/include/asm/topology.h	Thu Nov 20 13:55:44 2008 +1030
> @@ -231,6 +231,7 @@
>  #endif
>  
>  extern cpumask_t cpu_coregroup_map(int cpu);
> +extern const struct cpumask *cpu_coregroup_mask(int cpu);
>  
>  #ifdef ENABLE_TOPO_DEFINES
>  #define topology_physical_package_id(cpu)	(cpu_data(cpu).phys_proc_id)
> diff -r 11a884ebd99b arch/x86/kernel/smpboot.c
> --- a/arch/x86/kernel/smpboot.c	Thu Nov 20 13:54:24 2008 +1030
> +++ b/arch/x86/kernel/smpboot.c	Thu Nov 20 13:55:44 2008 +1030
> @@ -497,7 +497,7 @@
>  }
>  
>  /* maps the cpu to the sched domain representing multi-core */
> -cpumask_t cpu_coregroup_map(int cpu)
> +const struct cpumask *cpu_coregroup_mask(int cpu)
>  {
>  	struct cpuinfo_x86 *c = &cpu_data(cpu);
>  	/*
> @@ -505,9 +505,14 @@
>  	 * And for power savings, we return cpu_core_map
>  	 */
>  	if (sched_mc_power_savings || sched_smt_power_savings)
> -		return per_cpu(cpu_core_map, cpu);
> +		return &per_cpu(cpu_core_map, cpu);
>  	else
> -		return c->llc_shared_map;
> +		return &c->llc_shared_map;
> +}
> +
> +cpumask_t cpu_coregroup_map(int cpu)
> +{
> +	return *cpu_coregroup_mask(cpu);
>  }
>  
>  static void impress_friends(void)
> 

Re: [PATCH 1/4] cpumask: cpu_coregroup_mask(): x86

[Mike Travis]

Mike Travis wrote:
> Rusty Russell wrote:
>> Like cpu_coregroup_map, but returns a (const) pointer.
>>
>> (This will go to Ingo separately as part of the x86 series, just
>> airing it here for thoroughness).
>>
>> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
>> Signed-off-by: Mike Travis <travis@sgi.com>
>> Cc: Ingo Molnar <mingo@redhat.com>
> 
> I've pulled this patch into the queue for my cpus4096-for-ingo tree.
> 
> 	[PATCH 1/4] cpumask: cpu_coregroup_mask(): x86
> 
> I can also line up a queue for sched related changes:
> 
> 	[PATCH 4/4] cpumask: Replace cpu_coregroup_map with cpu_coregroup_map()
> 
> Is there any status on the corresponding changes for sparc, s390?  (I assume
> that they'll need to be merged into linux-next?)
> 
> 	[PATCH 2/4] cpumask: cpu_coregroup_mask(): sparc
> 	[PATCH 3/4] cpumask: cpu_coregroup_mask(): s390

Oops, never mind.  I just noticed that all 4 have already been pushed via
linux-next.

Ingo - how do we get these back into -tip for testing with other cpus4096
changes?  Do I need to do anything?

Thanks,
Mike

Re: [PATCH 1/4] cpumask: cpu_coregroup_mask(): x86

[Rusty Russell]

On Thursday 18 December 2008 09:24:19 Mike Travis wrote:
> Rusty Russell wrote:
> > Like cpu_coregroup_map, but returns a (const) pointer.
> > 
> > (This will go to Ingo separately as part of the x86 series, just
> > airing it here for thoroughness).
> > 
> > Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
> > Signed-off-by: Mike Travis <travis@sgi.com>
> > Cc: Ingo Molnar <mingo@redhat.com>
> 
> I've pulled this patch into the queue for my cpus4096-for-ingo tree.
> 
> 	[PATCH 1/4] cpumask: cpu_coregroup_mask(): x86
> 
> I can also line up a queue for sched related changes:
> 
> 	[PATCH 4/4] cpumask: Replace cpu_coregroup_map with cpu_coregroup_map()
> 
> Is there any status on the corresponding changes for sparc, s390?  (I assume
> that they'll need to be merged into linux-next?)
> 
> 	[PATCH 2/4] cpumask: cpu_coregroup_mask(): sparc
> 	[PATCH 3/4] cpumask: cpu_coregroup_mask(): s390

You can't take 4/4 until 2/4 and 3/4 are merged, and since it's not
x86-specific as well, we'll leave it in my tree.

Thanks,
Rusty.