[PATCH] powerpc/mm: Drop dump_numa_memory_topology()

[PATCH] powerpc/mm: Drop dump_numa_memory_topology()

[Michael Ellerman]

At boot we dump the NUMA memory topology in dump_numa_memory_topology(),
at KERN_DEBUG level, resulting in output like:

  Node 0 Memory: 0x0-0x100000000
  Node 1 Memory: 0x100000000-0x200000000

Which is nice enough, but immediately after that we iterate over each
node and call setup_node_data(), which also prints out the node ranges,
at KERN_INFO, giving eg:

  numa: Initmem setup node 0 [mem 0x00000000-0xffffffff]
  numa: Initmem setup node 1 [mem 0x100000000-0x1ffffffff]

So drop dump_numa_memory_topology() as superfluous chatter.

Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
---
 arch/powerpc/mm/numa.c | 36 ------------------------------------
 1 file changed, 36 deletions(-)

diff --git a/arch/powerpc/mm/numa.c b/arch/powerpc/mm/numa.c
index 75b9cd6150cc..db5fc2b54c5a 100644
--- a/arch/powerpc/mm/numa.c
+++ b/arch/powerpc/mm/numa.c
@@ -871,40 +871,6 @@ void __init dump_numa_cpu_topology(void)
 	}
 }
 
-static void __init dump_numa_memory_topology(void)
-{
-	unsigned int node;
-	unsigned int count;
-
-	if (min_common_depth == -1 || !numa_enabled)
-		return;
-
-	for_each_online_node(node) {
-		unsigned long i;
-
-		printk(KERN_DEBUG "Node %d Memory:", node);
-
-		count = 0;
-
-		for (i = 0; i < memblock_end_of_DRAM();
-		     i += (1 << SECTION_SIZE_BITS)) {
-			if (early_pfn_to_nid(i >> PAGE_SHIFT) == node) {
-				if (count == 0)
-					printk(" 0x%lx", i);
-				++count;
-			} else {
-				if (count > 0)
-					printk("-0x%lx", i);
-				count = 0;
-			}
-		}
-
-		if (count > 0)
-			printk("-0x%lx", i);
-		printk("\n");
-	}
-}
-
 /* Initialize NODE_DATA for a node on the local memory */
 static void __init setup_node_data(int nid, u64 start_pfn, u64 end_pfn)
 {
@@ -947,8 +913,6 @@ void __init initmem_init(void)
 
 	if (parse_numa_properties())
 		setup_nonnuma();
-	else
-		dump_numa_memory_topology();
 
 	memblock_dump_all();
 
-- 
2.7.4

Re: [PATCH] powerpc/mm: Drop dump_numa_memory_topology()

[Balbir Singh]

On 13/10/16 16:27, Michael Ellerman wrote:
> At boot we dump the NUMA memory topology in dump_numa_memory_topology(),
> at KERN_DEBUG level, resulting in output like:
> 
>   Node 0 Memory: 0x0-0x100000000
>   Node 1 Memory: 0x100000000-0x200000000
> 
> Which is nice enough, but immediately after that we iterate over each
> node and call setup_node_data(), which also prints out the node ranges,
> at KERN_INFO, giving eg:
> 
>   numa: Initmem setup node 0 [mem 0x00000000-0xffffffff]
>   numa: Initmem setup node 1 [mem 0x100000000-0x1ffffffff]
> 
> So drop dump_numa_memory_topology() as superfluous chatter.
> 
> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
> ---


Actually the ranges for the lower output seem a little better
(inclusive), the output on top is not clear

Acked-by: Balbir Singh <bsingharora@gmail.com>

Re: [PATCH] powerpc/mm: Drop dump_numa_memory_topology()

[Michael Ellerman]

Balbir Singh <bsingharora@gmail.com> writes:

> On 13/10/16 16:27, Michael Ellerman wrote:
>> At boot we dump the NUMA memory topology in dump_numa_memory_topology(),
>> at KERN_DEBUG level, resulting in output like:
>> 
>>   Node 0 Memory: 0x0-0x100000000
>>   Node 1 Memory: 0x100000000-0x200000000
>> 
>> Which is nice enough, but immediately after that we iterate over each
>> node and call setup_node_data(), which also prints out the node ranges,
>> at KERN_INFO, giving eg:
>> 
>>   numa: Initmem setup node 0 [mem 0x00000000-0xffffffff]
>>   numa: Initmem setup node 1 [mem 0x100000000-0x1ffffffff]
>> 
>> So drop dump_numa_memory_topology() as superfluous chatter.
>> 
>> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
>> ---
>
> Actually the ranges for the lower output seem a little better
> (inclusive), the output on top is not clear

I think everyone knows what it means but it's not entirely precise, so
yeah the inclusive range is better.

cheers

Re: powerpc/mm: Drop dump_numa_memory_topology()

[Michael Ellerman]

On Thu, 2016-13-10 at 05:27:30 UTC, Michael Ellerman wrote:
> At boot we dump the NUMA memory topology in dump_numa_memory_topology(),
> at KERN_DEBUG level, resulting in output like:
> 
>   Node 0 Memory: 0x0-0x100000000
>   Node 1 Memory: 0x100000000-0x200000000
> 
> Which is nice enough, but immediately after that we iterate over each
> node and call setup_node_data(), which also prints out the node ranges,
> at KERN_INFO, giving eg:
> 
>   numa: Initmem setup node 0 [mem 0x00000000-0xffffffff]
>   numa: Initmem setup node 1 [mem 0x100000000-0x1ffffffff]
> 
> So drop dump_numa_memory_topology() as superfluous chatter.
> 
> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
> Acked-by: Balbir Singh <bsingharora@gmail.com>

Applied to powerpc fixes.

https://git.kernel.org/powerpc/c/08b5e79ebdb58868cbb6976ba0e389

cheers