Skip to contentSkip to conversation

x86: use near online node instead of round bin for numa

PATCH In review Yinghai Lu 2 participants 6 messages Oct 3, 2009 – Oct 5, 2009

Subsystems (2)
lore
Share & tools
lore

Yinghai Lu

author
#0 lore raw

PATCH 1/1 x86: use near online node instead of round bin for numa

cpu to node mapping is set in following sequence:
1. numa_init_array: set up roundbin from cpu to online node
2. init_cpu_to_node: set that according to apicid_to_node[] according to srat only handle that node is online, and leave other cpu on node without ram (aka not online) to still round-bin
3. later srat_detect_node for intel/amd, will use first_online node or near by node.

problem is that setup_per_cpu_areas() is called between 2 and 3. the per_cpu for cpu on node with ram is on different node. and could put that on node with two hops away.

so try add find_near_online_node() and call int init_cpu_to_node()

---
 arch/x86/kernel/cpu/intel.c |    6 +++++-
 arch/x86/mm/numa_64.c       |   21 ++++++++++++++++++++-
 2 files changed, 25 insertions(+), 2 deletions(-)
arch/x86/mm/numa_64.c +20 -1
Index: linux-2.6/arch/x86/mm/numa_64.c
===================================================================
@@ -601,6 +601,25 @@ static __init int numa_setup(char *opt)
 early_param("numa", numa_setup);
 
 #ifdef CONFIG_NUMA
+
+static __init int find_near_online_node(int node)
+{
+	int n, val;
+	int min_val = INT_MAX;
+	int best_node = -1;
+
+	for_each_online_node(n) {
+		val = node_distance(node, n);
+
+		if (val < min_val) {
+			min_val = val;
+			best_node = n;
+		}
+	}
+
+	return best_node;
+}
+
 /*
  * Setup early cpu_to_node.
  *
@@ -632,7 +651,7 @@ void __init init_cpu_to_node(void)
 		if (node == NUMA_NO_NODE)
 			continue;
 		if (!node_online(node))
-			continue;
+			node = find_near_online_node(node);
 		numa_set_node(cpu, node);
 	}
 }
arch/x86/kernel/cpu/intel.c +5 -1
Index: linux-2.6/arch/x86/kernel/cpu/intel.c
===================================================================
@@ -263,8 +263,12 @@ static void __cpuinit srat_detect_node(s
 	/* Don't do the funky fallback heuristics the AMD version employs
 	   for now. */
 	node = apicid_to_node[apicid];
-	if (node == NUMA_NO_NODE || !node_online(node))
+	if (node == NUMA_NO_NODE)
 		node = first_node(node_online_map);
+	else if (!node_online(node)) {
+		/* reuse the value from init_cpu_to_node() */
+		node = cpu_to_node(cpu);
+	}
 	numa_set_node(cpu, node);
 
 	printk(KERN_INFO "CPU %d/0x%x -> Node %d\n", cpu, apicid, node);
#1 lore raw
Quoted 8 lines · Yinghai Lu

cpu to node mapping is set in following sequence:
1. numa_init_array: set up roundbin from cpu to online node
2. init_cpu_to_node: set that according to apicid_to_node[] according to srat only handle that node is online, and leave other cpu on node without ram (aka not online) to still round-bin
3. later srat_detect_node for intel/amd, will use first_online node or near by node.

Quoted 5 lines

problem is that setup_per_cpu_areas() is called between 2 and 3. the per_cpu for cpu on node with ram is on different node. and could put that on node with two hops away.

so try add find_near_online_node() and call int init_cpu_to_node()

This fallback case should not really happen anyways, unless the BIOS is buggy (in this case it might better to completely reject the SRAT because more might be wrong).

Do you have a system where this is needed?

-Andi

Signature · 1 line
[email protected] -- Speaking for myself only.
author
#2 lore raw
Quoted 14 lines · Andi Kleen
Quoted 12 lines · Yinghai Lu

cpu to node mapping is set in following sequence:
1. numa_init_array: set up roundbin from cpu to online node
2. init_cpu_to_node: set that according to apicid_to_node[] according to srat only handle that node is online, and leave other cpu on node without ram (aka not online) to still round-bin
3. later srat_detect_node for intel/amd, will use first_online node or near by node.

problem is that setup_per_cpu_areas() is called between 2 and 3. the per_cpu for cpu on node with ram is on different node. and could put that on node with two hops away.

Quoted 5 lines
Quoted 1 line

so try add find_near_online_node() and call int init_cpu_to_node()

This fallback case should not really happen anyways, unless the BIOS is buggy (in this case it might better to completely reject the SRAT because more might be wrong).

SRAT is right, and some node has no ram installed.

Quoted 2 lines

Do you have a system where this is needed?

sure.

YH

#3 lore raw
Quoted 15 lines · Yinghai Lu
Quoted 14 lines · Andi Kleen
Quoted 12 lines · Yinghai Lu

cpu to node mapping is set in following sequence:
1. numa_init_array: set up roundbin from cpu to online node
2. init_cpu_to_node: set that according to apicid_to_node[] according to srat only handle that node is online, and leave other cpu on node without ram (aka not online) to still round-bin
3. later srat_detect_node for intel/amd, will use first_online node or near by node.

problem is that setup_per_cpu_areas() is called between 2 and 3. the per_cpu for cpu on node with ram is on different node. and could put that on node with two hops away.

Quoted 6 lines
Quoted 5 lines
Quoted 1 line

so try add find_near_online_node() and call int init_cpu_to_node()

This fallback case should not really happen anyways, unless the BIOS is buggy (in this case it might better to completely reject the SRAT because more might be wrong).

SRAT is right, and some node has no ram installed.

In this case there should be still a PXM to define the CPU locality -- your BIOS is broken. Please fix it there.

-Andi

Signature · 1 line
[email protected] -- Speaking for myself only.
author
#4 lore raw
Quoted 16 lines · Andi Kleen
Quoted 15 lines · Yinghai Lu
Quoted 14 lines · Andi Kleen
Quoted 12 lines · Yinghai Lu

cpu to node mapping is set in following sequence:
1. numa_init_array: set up roundbin from cpu to online node
2. init_cpu_to_node: set that according to apicid_to_node[] according to srat only handle that node is online, and leave other cpu on node without ram (aka not online) to still round-bin
3. later srat_detect_node for intel/amd, will use first_online node or near by node.

problem is that setup_per_cpu_areas() is called between 2 and 3. the per_cpu for cpu on node with ram is on different node. and could put that on node with two hops away.

Quoted 8 lines
Quoted 5 lines
Quoted 4 lines
Quoted 1 line

so try add find_near_online_node() and call int init_cpu_to_node()

This fallback case should not really happen anyways, unless the BIOS is buggy (in this case it might better to completely reject the SRAT because more might be wrong).

SRAT is right, and some node has no ram installed.

In this case there should be still a PXM to define the CPU locality -- your BIOS is broken. Please fix it there.

I don't think so.

YH

#5 lore raw
Quoted 21 lines · Yinghai Lu
Quoted 20 lines · Andi Kleen
Quoted 19 lines · Yinghai Lu
Quoted 18 lines · Andi Kleen
Quoted 8 lines · Yinghai Lu

cpu to node mapping is set in following sequence:
1. numa_init_array: set up roundbin from cpu to online node
2. init_cpu_to_node: set that according to apicid_to_node[] according to srat only handle that node is online, and leave other cpu on node without ram (aka not online) to still round-bin
3. later srat_detect_node for intel/amd, will use first_online node or near by node.

Quoted 5 lines

problem is that setup_per_cpu_areas() is called between 2 and 3. the per_cpu for cpu on node with ram is on different node. and could put that on node with two hops away.

so try add find_near_online_node() and call int init_cpu_to_node()

This fallback case should not really happen anyways, unless the BIOS is buggy (in this case it might better to completely reject the SRAT because more might be wrong).

Quoted 6 lines
Quoted 4 lines
Quoted 1 line

SRAT is right, and some node has no ram installed.

In this case there should be still a PXM to define the CPU locality -- your BIOS is broken. Please fix it there.

I don't think so.

Let's put it like this: your BIOS does not describe the full system topology which is a severe BIOS bug. Putting hacks into Linux to work around that is not the right solution.

-Andi

Signature · 1 line
[email protected] -- Speaking for myself only.