[PATCH v2 01/10] powerpc/smp: Cache node for reuse

[Michael Ellerman]

Srikar Dronamraju <srikar at linux.vnet.ibm.com> writes:
> While cpu_to_node is inline function with access to per_cpu variable.
> However when using repeatedly, it may be cleaner to cache it in a local
> variable.

It's not clear what "cleaner" is supposed to mean. Are you saying it
makes the source clearer, or the generated code?

I'm not sure it will make any difference to the latter.

> Also fix a build error in a some weird config.
> "error: _numa_cpu_lookup_table_ undeclared"

Separate patch please.

> No functional change

The ifdef change means that's not true.

> Cc: linuxppc-dev <linuxppc-dev at lists.ozlabs.org>
> Cc: LKML <linux-kernel at vger.kernel.org>
> Cc: Michael Ellerman <michaele at au1.ibm.com>
> Cc: Ingo Molnar <mingo at kernel.org>
> Cc: Peter Zijlstra <peterz at infradead.org>
> Cc: Valentin Schneider <valentin.schneider at arm.com>
> Cc: Nick Piggin <npiggin at au1.ibm.com>
> Cc: Oliver OHalloran <oliveroh at au1.ibm.com>
> Cc: Nathan Lynch <nathanl at linux.ibm.com>
> Cc: Michael Neuling <mikey at linux.ibm.com>
> Cc: Anton Blanchard <anton at au1.ibm.com>
> Cc: Gautham R Shenoy <ego at linux.vnet.ibm.com>
> Cc: Vaidyanathan Srinivasan <svaidy at linux.ibm.com>
> Cc: Jordan Niethe <jniethe5 at gmail.com>
> Reviewed-by: Gautham R. Shenoy <ego at linux.vnet.ibm.com>
> Signed-off-by: Srikar Dronamraju <srikar at linux.vnet.ibm.com>
> ---
>  arch/powerpc/kernel/smp.c | 18 +++++++++++-------
>  1 file changed, 11 insertions(+), 7 deletions(-)
>
> diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
> index 73199470c265..680c0edcc59d 100644
> --- a/arch/powerpc/kernel/smp.c
> +++ b/arch/powerpc/kernel/smp.c
> @@ -843,7 +843,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
>  
>  	DBG("smp_prepare_cpus\n");
>  
> -	/* 
> +	/*
>  	 * setup_cpu may need to be called on the boot cpu. We havent
>  	 * spun any cpus up but lets be paranoid.
>  	 */
> @@ -854,20 +854,24 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
>  	cpu_callin_map[boot_cpuid] = 1;
>  
>  	for_each_possible_cpu(cpu) {
> +		int node = cpu_to_node(cpu);
> +

Does cpu_to_node() even work here?

Doesn't look like it to me.

More fallout from 8c272261194d ("powerpc/numa: Enable USE_PERCPU_NUMA_NODE_ID") ?

>  		zalloc_cpumask_var_node(&per_cpu(cpu_sibling_map, cpu),
> -					GFP_KERNEL, cpu_to_node(cpu));
> +					GFP_KERNEL, node);
>  		zalloc_cpumask_var_node(&per_cpu(cpu_l2_cache_map, cpu),
> -					GFP_KERNEL, cpu_to_node(cpu));
> +					GFP_KERNEL, node);
>  		zalloc_cpumask_var_node(&per_cpu(cpu_core_map, cpu),
> -					GFP_KERNEL, cpu_to_node(cpu));
> +					GFP_KERNEL, node);
> +#ifdef CONFIG_NEED_MULTIPLE_NODES
>  		/*
>  		 * numa_node_id() works after this.
>  		 */
>  		if (cpu_present(cpu)) {
> -			set_cpu_numa_node(cpu, numa_cpu_lookup_table[cpu]);
> -			set_cpu_numa_mem(cpu,
> -				local_memory_node(numa_cpu_lookup_table[cpu]));
> +			node = numa_cpu_lookup_table[cpu];
> +			set_cpu_numa_node(cpu, node);
> +			set_cpu_numa_mem(cpu, local_memory_node(node));
>  		}
> +#endif
>  	}
>  
>  	/* Init the cpumasks so the boot CPU is related to itself */
> -- 
> 2.17.1

cheers