Release 4.14 arch/powerpc/mm/numa.c

/*
 * pSeries NUMA support
 *
 * Copyright (C) 2002 Anton Blanchard <anton@au.ibm.com>, IBM
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License
 * as published by the Free Software Foundation; either version
 * 2 of the License, or (at your option) any later version.
 */

#define pr_fmt(fmt) "numa: " fmt

#include <linux/threads.h>
#include <linux/bootmem.h>
#include <linux/init.h>
#include <linux/mm.h>
#include <linux/mmzone.h>
#include <linux/export.h>
#include <linux/nodemask.h>
#include <linux/cpu.h>
#include <linux/notifier.h>
#include <linux/memblock.h>
#include <linux/of.h>
#include <linux/pfn.h>
#include <linux/cpuset.h>
#include <linux/node.h>
#include <linux/stop_machine.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/uaccess.h>
#include <linux/slab.h>
#include <asm/cputhreads.h>
#include <asm/sparsemem.h>
#include <asm/prom.h>
#include <asm/smp.h>
#include <asm/cputhreads.h>
#include <asm/topology.h>
#include <asm/firmware.h>
#include <asm/paca.h>
#include <asm/hvcall.h>
#include <asm/setup.h>
#include <asm/vdso.h>


static int numa_enabled = 1;


static char *cmdline __initdata;


static int numa_debug;

#define dbg(args...) if (numa_debug) { printk(KERN_INFO args); }


int numa_cpu_lookup_table[NR_CPUS];

cpumask_var_t node_to_cpumask_map[MAX_NUMNODES];

struct pglist_data *node_data[MAX_NUMNODES];


EXPORT_SYMBOL(numa_cpu_lookup_table);

EXPORT_SYMBOL(node_to_cpumask_map);

EXPORT_SYMBOL(node_data);


static int min_common_depth;


static int n_mem_addr_cells, n_mem_size_cells;

static int form1_affinity;


#define MAX_DISTANCE_REF_POINTS 4

static int distance_ref_points_depth;

static const __be32 *distance_ref_points;

static int distance_lookup_table[MAX_NUMNODES][MAX_DISTANCE_REF_POINTS];

/*
 * Allocate node_to_cpumask_map based on number of available nodes
 * Requires node_possible_map to be valid.
 *
 * Note: cpumask_of_node() is not valid until after this is done.
 */


static void __init setup_node_to_cpumask_map(void)
{
	unsigned int node;

	/* setup nr_node_ids if not done yet */
	if (nr_node_ids == MAX_NUMNODES)
		setup_nr_node_ids();

	/* allocate the map */
	for_each_node(node)
		alloc_bootmem_cpumask_var(&node_to_cpumask_map[node]);

	/* cpumask_of_node() will now work */
	dbg("Node to cpumask map for %d nodes\n", nr_node_ids);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Anton Blanchard
39
86.67%
1
33.33%
Raghavendra K T
4
8.89%
1
33.33%
Cody P Schafer
2
4.44%
1
33.33%
Total
45
100.00%
3
100.00%



static int __init fake_numa_create_new_node(unsigned long end_pfn,
						unsigned int *nid)
{
	unsigned long long mem;
	char *p = cmdline;
	static unsigned int fake_nid;
	static unsigned long long curr_boundary;

	/*
         * Modify node id, iff we started creating NUMA nodes
         * We want to continue from where we left of the last time
         */
	if (fake_nid)
		*nid = fake_nid;
	/*
         * In case there are no more arguments to parse, the
         * node_id should be the same as the last fake node id
         * (we've handled this above).
         */
	if (!p)
		return 0;

	mem = memparse(p, &p);
	if (!mem)
		return 0;

	if (mem < curr_boundary)
		return 0;

	curr_boundary = mem;

	if ((end_pfn << PAGE_SHIFT) > mem) {
		/*
                 * Skip commas and spaces
                 */
		while (*p == ',' || *p == ' ' || *p == '\t')
			p++;

		cmdline = p;
		fake_nid++;
		*nid = fake_nid;
		dbg("created new fake_node with id %d\n", fake_nid);
		return 1;
	}
	return 0;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Balbir Singh
145
99.32%
1
50.00%
Stephen Rothwell
1
0.68%
1
50.00%
Total
146
100.00%
2
100.00%



static void reset_numa_cpu_lookup_table(void)
{
	unsigned int cpu;

	for_each_possible_cpu(cpu)
		numa_cpu_lookup_table[cpu] = -1;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Srivatsa S. Bhat
21
87.50%
1
50.00%
Anton Blanchard
3
12.50%
1
50.00%
Total
24
100.00%
2
100.00%



static void update_numa_cpu_lookup_table(unsigned int cpu, int node)
{
	numa_cpu_lookup_table[cpu] = node;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Anton Blanchard
15
75.00%
1
50.00%
Srivatsa S. Bhat
5
25.00%
1
50.00%
Total
20
100.00%
2
100.00%



static void map_cpu_to_node(int cpu, int node)
{
	update_numa_cpu_lookup_table(cpu, node);

	dbg("adding cpu %d to node %d\n", cpu, node);

	if (!(cpumask_test_cpu(cpu, node_to_cpumask_map[node])))
		cpumask_set_cpu(cpu, node_to_cpumask_map[node]);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Srivatsa S. Bhat
18
33.96%
1
16.67%
Anton Blanchard
16
30.19%
3
50.00%
Andrew Morton
10
18.87%
1
16.67%
Nathan Lynch
9
16.98%
1
16.67%
Total
53
100.00%
6
100.00%

#if defined(CONFIG_HOTPLUG_CPU) || defined(CONFIG_PPC_SPLPAR)


static void unmap_cpu_from_node(unsigned long cpu)
{
	int node = numa_cpu_lookup_table[cpu];

	dbg("removing cpu %lu from node %d\n", cpu, node);

	if (cpumask_test_cpu(cpu, node_to_cpumask_map[node])) {
		cpumask_clear_cpu(cpu, node_to_cpumask_map[node]);
	} else {
		printk(KERN_ERR "WARNING: cpu %lu not found in node %d\n",
		       cpu, node);
	}
}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Paul Mackerras
60
93.75%
1
33.33%
Anton Blanchard
4
6.25%
2
66.67%
Total
64
100.00%
3
100.00%

#endif /* CONFIG_HOTPLUG_CPU || CONFIG_PPC_SPLPAR */

/* must hold reference to node during call */


static const __be32 *of_get_associativity(struct device_node *dev)
{
	return of_get_property(dev, "ibm,associativity", NULL);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Andrew Morton
18
78.26%
1
20.00%
Paul Mackerras
2
8.70%
1
20.00%
Alistair Popple
1
4.35%
1
20.00%
Stephen Rothwell
1
4.35%
1
20.00%
Jeremy Kerr
1
4.35%
1
20.00%
Total
23
100.00%
5
100.00%

/*
 * Returns the property linux,drconf-usable-memory if
 * it exists (the property exists only in kexec/kdump kernels,
 * added by kexec-tools)
 */


static const __be32 *of_get_usable_memory(struct device_node *memory)
{
	const __be32 *prop;
	u32 len;
	prop = of_get_property(memory, "linux,drconf-usable-memory", &len);
	if (!prop || len < sizeof(unsigned int))
		return NULL;
	return prop;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Chandru
49
94.23%
1
33.33%
Alistair Popple
2
3.85%
1
33.33%
Robert Jennings
1
1.92%
1
33.33%
Total
52
100.00%
3
100.00%



int __node_distance(int a, int b)
{
	int i;
	int distance = LOCAL_DISTANCE;

	if (!form1_affinity)
		return ((a == b) ? LOCAL_DISTANCE : REMOTE_DISTANCE);

	for (i = 0; i < distance_ref_points_depth; i++) {
		if (distance_lookup_table[a][i] == distance_lookup_table[b][i])
			break;

		/* Double the distance for each NUMA level */
		distance *= 2;
	}

	return distance;
}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Anton Blanchard
68
86.08%
1
50.00%
Vaidyanathan Srinivasan
11
13.92%
1
50.00%
Total
79
100.00%
2
100.00%


EXPORT_SYMBOL(__node_distance);



static void initialize_distance_lookup_table(int nid,
		const __be32 *associativity)
{
	int i;

	if (!form1_affinity)
		return;

	for (i = 0; i < distance_ref_points_depth; i++) {
		const __be32 *entry;

		entry = &associativity[be32_to_cpu(distance_ref_points[i]) - 1];
		distance_lookup_table[nid][i] = of_read_number(entry, 1);
	}
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Anton Blanchard
46
62.16%
1
33.33%
Alistair Popple
26
35.14%
1
33.33%
Nikunj A. Dadhania
2
2.70%
1
33.33%
Total
74
100.00%
3
100.00%

/* Returns nid in the range [0..MAX_NUMNODES-1], or -1 if no useful numa
 * info is found.
 */


static int associativity_to_nid(const __be32 *associativity)
{
	int nid = -1;

	if (min_common_depth == -1)
		goto out;

	if (of_read_number(associativity, 1) >= min_common_depth)
		nid = of_read_number(&associativity[min_common_depth], 1);

	/* POWER4 LPAR uses 0xffff as invalid node */
	if (nid == 0xffff || nid >= MAX_NUMNODES)
		nid = -1;

	if (nid > 0 &&
		of_read_number(associativity, 1) >= distance_ref_points_depth) {
		/*
                 * Skip the length field and send start of associativity array
                 */
		initialize_distance_lookup_table(nid, associativity + 1);
	}

out:
	return nid;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Nathan Lynch
29
29.29%
3
33.33%
Andrew Morton
17
17.17%
1
11.11%
Alistair Popple
17
17.17%
1
11.11%
Anton Blanchard
15
15.15%
1
11.11%
Paul Mackerras
9
9.09%
1
11.11%
Jesse Larrew
7
7.07%
1
11.11%
Nikunj A. Dadhania
5
5.05%
1
11.11%
Total
99
100.00%
9
100.00%

/* Returns the nid associated with the given device tree node,
 * or -1 if not found.
 */


static int of_node_to_nid_single(struct device_node *device)
{
	int nid = -1;
	const __be32 *tmp;

	tmp = of_get_associativity(device);
	if (tmp)
		nid = associativity_to_nid(tmp);
	return nid;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Jesse Larrew
42
97.67%
1
50.00%
Alistair Popple
1
2.33%
1
50.00%
Total
43
100.00%
2
100.00%

/* Walk the device tree upwards, looking for an associativity id */


int of_node_to_nid(struct device_node *device)
{
	int nid = -1;

	of_node_get(device);
	while (device) {
		nid = of_node_to_nid_single(device);
		if (nid != -1)
			break;

		device = of_get_next_parent(device);
	}
	of_node_put(device);

	return nid;
}
Contributors
Person
Tokens
Prop
Commits
CommitProp
Jeremy Kerr
55
96.49%
1
50.00%
Christophe Jaillet
2
3.51%
1
50.00%
Total
57
100.00%
2
100.00%


EXPORT_SYMBOL(of_node_to_nid);



static int __init find_min_common_depth(void)
{
	int depth;
	struct device_node *root;

	if (firmware_has_feature(FW_FEATURE_OPAL))
		root = of_find_node_by_path("/ibm,opal");
	else
		root = of_find_node_by_path("/rtas");
	if (!root)
		root = of_find_node_by_path("/");

	/*
         * This property is a set of 32-bit integers, each representing
         * an index into the ibm,associativity nodes.
         *
         * With form 0 affinity the first integer is for an SMP configuration
         * (should be all 0's) and the second is for a normal NUMA
         * configuration. We have only one level of NUMA.
         *
         * With form 1 affinity the first integer is the most significant
         * NUMA boundary and the following are progressively less significant
         * boundaries. There can be more than one level of NUMA.
         */
	distance_ref_points = of_get_property(root,
					"ibm,associativity-reference-points",
					&distance_ref_points_depth);

	if (!distance_ref_points) {
		dbg("NUMA: ibm,associativity-reference-points not found.\n");
		goto err;
	}

	distance_ref_points_depth /= sizeof(int);

	if (firmware_has_feature(FW_FEATURE_OPAL) ||
	    firmware_has_feature(FW_FEATURE_TYPE1_AFFINITY)) {
		dbg("Using form 1 affinity\n");
		form1_affinity = 1;
	}

	if (form1_affinity) {
		depth = of_read_number(distance_ref_points, 1);
	} else {
		if (distance_ref_points_depth < 2) {
			printk(KERN_WARNING "NUMA: "
				"short ibm,associativity-reference-points\n");
			goto err;
		}

		depth = of_read_number(&distance_ref_points[1], 1);
	}

	/*
         * Warn and cap if the hardware supports more than
         * MAX_DISTANCE_REF_POINTS domains.
         */
	if (distance_ref_points_depth > MAX_DISTANCE_REF_POINTS) {
		printk(KERN_WARNING "NUMA: distance array capped at "
			"%d entries\n", MAX_DISTANCE_REF_POINTS);
		distance_ref_points_depth = MAX_DISTANCE_REF_POINTS;
	}

	of_node_put(root);
	return depth;

err:
	of_node_put(root);
	return -1;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Anton Blanchard
92
46.23%
3
30.00%
Andrew Morton
57
28.64%
1
10.00%
Dipankar Sarma
21
10.55%
1
10.00%
Michael Ellerman
13
6.53%
1
10.00%
Alistair Popple
11
5.53%
1
10.00%
Nathan Fontenot
3
1.51%
1
10.00%
Paul Mackerras
1
0.50%
1
10.00%
Stephen Rothwell
1
0.50%
1
10.00%
Total
199
100.00%
10
100.00%



static void __init get_n_mem_cells(int *n_addr_cells, int *n_size_cells)
{
	struct device_node *memory = NULL;

	memory = of_find_node_by_type(memory, "memory");
	if (!memory)
		panic("numa.c: No memory nodes found!");

	*n_addr_cells = of_n_addr_cells(memory);
	*n_size_cells = of_n_size_cells(memory);
	of_node_put(memory);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Mike Kravetz
58
93.55%
3
42.86%
Stephen Rothwell
2
3.23%
2
28.57%
Paul Mackerras
1
1.61%
1
14.29%
Andrew Morton
1
1.61%
1
14.29%
Total
62
100.00%
7
100.00%



static unsigned long read_n_cells(int n, const __be32 **buf)
{
	unsigned long result = 0;

	while (n--) {
		result = (result << 32) | of_read_number(*buf, 1);
		(*buf)++;
	}
	return result;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Andrew Morton
39
72.22%
1
25.00%
Mike Kravetz
8
14.81%
1
25.00%
Alistair Popple
6
11.11%
1
25.00%
Jeremy Kerr
1
1.85%
1
25.00%
Total
54
100.00%
4
100.00%

/*
 * Read the next memblock list entry from the ibm,dynamic-memory property
 * and return the information in the provided of_drconf_cell structure.
 */


static void read_drconf_cell(struct of_drconf_cell *drmem, const __be32 **cellp)
{
	const __be32 *cp;

	drmem->base_addr = read_n_cells(n_mem_addr_cells, cellp);

	cp = *cellp;
	drmem->drc_index = of_read_number(cp, 1);
	drmem->reserved = of_read_number(&cp[1], 1);
	drmem->aa_index = of_read_number(&cp[2], 1);
	drmem->flags = of_read_number(&cp[3], 1);

	*cellp = cp + 4;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Nathan Fontenot
76
75.25%
1
50.00%
Alistair Popple
25
24.75%
1
50.00%
Total
101
100.00%
2
100.00%

/*
 * Retrieve and validate the ibm,dynamic-memory property of the device tree.
 *
 * The layout of the ibm,dynamic-memory property is a number N of memblock
 * list entries followed by N memblock list entries.  Each memblock list entry
 * contains information as laid out in the of_drconf_cell struct above.
 */


static int of_get_drconf_memory(struct device_node *memory, const __be32 **dm)
{
	const __be32 *prop;
	u32 len, entries;

	prop = of_get_property(memory, "ibm,dynamic-memory", &len);
	if (!prop || len < sizeof(unsigned int))
		return 0;

	entries = of_read_number(prop++, 1);

	/* Now that we know the number of entries, revalidate the size
         * of the property read in to ensure we have everything
         */
	if (len < (entries * (n_mem_addr_cells + 4) + 1) * sizeof(unsigned int))
		return 0;

	*dm = prop;
	return entries;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Nathan Fontenot
92
92.93%
1
50.00%
Alistair Popple
7
7.07%
1
50.00%
Total
99
100.00%
2
100.00%

/*
 * Retrieve and validate the ibm,lmb-size property for drconf memory
 * from the device tree.
 */


static u64 of_get_lmb_size(struct device_node *memory)
{
	const __be32 *prop;
	u32 len;

	prop = of_get_property(memory, "ibm,lmb-size", &len);
	if (!prop || len < sizeof(unsigned int))
		return 0;

	return read_n_cells(n_mem_size_cells, &prop);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Nathan Fontenot
53
94.64%
1
33.33%
Benjamin Herrenschmidt
2
3.57%
1
33.33%
Alistair Popple
1
1.79%
1
33.33%
Total
56
100.00%
3
100.00%


struct assoc_arrays {
	
u32	n_arrays;
	
u32	array_sz;
	
const __be32 *arrays;
};

/*
 * Retrieve and validate the list of associativity arrays for drconf
 * memory from the ibm,associativity-lookup-arrays property of the
 * device tree..
 *
 * The layout of the ibm,associativity-lookup-arrays property is a number N
 * indicating the number of associativity arrays, followed by a number M
 * indicating the size of each associativity array, followed by a list
 * of N associativity arrays.
 */


static int of_get_assoc_arrays(struct device_node *memory,
			       struct assoc_arrays *aa)
{
	const __be32 *prop;
	u32 len;

	prop = of_get_property(memory, "ibm,associativity-lookup-arrays", &len);
	if (!prop || len < 2 * sizeof(unsigned int))
		return -1;

	aa->n_arrays = of_read_number(prop++, 1);
	aa->array_sz = of_read_number(prop++, 1);

	/* Now that we know the number of arrays and size of each array,
         * revalidate the size of the property read in.
         */
	if (len < (aa->n_arrays * aa->array_sz + 2) * sizeof(unsigned int))
		return -1;

	aa->arrays = prop;
	return 0;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Nathan Fontenot
103
89.57%
1
33.33%
Alistair Popple
11
9.57%
1
33.33%
Justin P. Mattock
1
0.87%
1
33.33%
Total
115
100.00%
3
100.00%

/*
 * This is like of_node_to_nid_single() for memory represented in the
 * ibm,dynamic-reconfiguration-memory node.
 */


static int of_drconf_to_nid_single(struct of_drconf_cell *drmem,
				   struct assoc_arrays *aa)
{
	int default_nid = 0;
	int nid = default_nid;
	int index;

	if (min_common_depth > 0 && min_common_depth <= aa->array_sz &&
	    !(drmem->flags & DRCONF_MEM_AI_INVALID) &&
	    drmem->aa_index < aa->n_arrays) {
		index = drmem->aa_index * aa->array_sz + min_common_depth - 1;
		nid = of_read_number(&aa->arrays[index], 1);

		if (nid == 0xffff || nid >= MAX_NUMNODES)
			nid = default_nid;

		if (nid > 0) {
			index = drmem->aa_index * aa->array_sz;
			initialize_distance_lookup_table(nid,
							&aa->arrays[index]);
		}
	}

	return nid;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Nathan Fontenot
100
72.99%
1
33.33%
Nikunj A. Dadhania
31
22.63%
1
33.33%
Alistair Popple
6
4.38%
1
33.33%
Total
137
100.00%
3
100.00%

/*
 * Figure out to which domain a cpu belongs and stick it there.
 * Return the id of the domain used.
 */


static int numa_setup_cpu(unsigned long lcpu)
{
	int nid = -1;
	struct device_node *cpu;

	/*
         * If a valid cpu-to-node mapping is already available, use it
         * directly instead of querying the firmware, since it represents
         * the most recent mapping notified to us by the platform (eg: VPHN).
         */
	if ((nid = numa_cpu_lookup_table[lcpu]) >= 0) {
		map_cpu_to_node(lcpu, nid);
		return nid;
	}

	cpu = of_get_cpu_node(lcpu, NULL);

	if (!cpu) {
		WARN_ON(1);
		if (cpu_present(lcpu))
			goto out_present;
		else
			goto out;
	}

	nid = of_node_to_nid_single(cpu);

out_present:
	if (nid < 0 || !node_online(nid))
		nid = first_online_node;

	map_cpu_to_node(lcpu, nid);
	of_node_put(cpu);
out:
	return nid;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Paul Mackerras
58
46.77%
1
12.50%
Srivatsa S. Bhat
30
24.19%
1
12.50%
Li Zhong
18
14.52%
1
12.50%
Nathan Lynch
13
10.48%
2
25.00%
Milton D. Miller II
3
2.42%
1
12.50%
H Hartley Sweeten
1
0.81%
1
12.50%
Jeremy Kerr
1
0.81%
1
12.50%
Total
124
100.00%
8
100.00%



static void verify_cpu_node_mapping(int cpu, int node)
{
	int base, sibling, i;

	/* Verify that all the threads in the core belong to the same node */
	base = cpu_first_thread_sibling(cpu);

	for (i = 0; i < threads_per_core; i++) {
		sibling = base + i;

		if (sibling == cpu || cpu_is_offline(sibling))
			continue;

		if (cpu_to_node(sibling) != node) {
			WARN(1, "CPU thread siblings %d and %d don't belong"
				" to the same node!\n", cpu, sibling);
			break;
		}
	}
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Srivatsa S. Bhat
84
100.00%
1
100.00%
Total
84
100.00%
1
100.00%

/* Must run before sched domains notifier. */


static int ppc_numa_cpu_prepare(unsigned int cpu)
{
	int nid;

	nid = numa_setup_cpu(cpu);
	verify_cpu_node_mapping(cpu, nid);
	return 0;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Paul Mackerras
12
40.00%
1
33.33%
Srivatsa S. Bhat
9
30.00%
1
33.33%
Sebastian Andrzej Siewior
9
30.00%
1
33.33%
Total
30
100.00%
3
100.00%



static int ppc_numa_cpu_dead(unsigned int cpu)
{
#ifdef CONFIG_HOTPLUG_CPU
	unmap_cpu_from_node(cpu);
#endif
	return 0;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Paul Mackerras
12
52.17%
1
50.00%
Sebastian Andrzej Siewior
11
47.83%
1
50.00%
Total
23
100.00%
2
100.00%

/*
 * Check and possibly modify a memory region to enforce the memory limit.
 *
 * Returns the size the region should have to enforce the memory limit.
 * This will either be the original value of size, a truncated value,
 * or zero. If the returned value of size is 0 the region should be
 * discarded as it lies wholly above the memory limit.
 */


static unsigned long __init numa_enforce_memory_limit(unsigned long start,
						      unsigned long size)
{
	/*
         * We use memblock_end_of_DRAM() in here instead of memory_limit because
         * we've already adjusted it for the limit and it takes care of
         * having memory holes below the limit.  Also, in the case of
         * iommu_is_off, memory_limit is not set but is implicitly enforced.
         */

	if (start + size <= memblock_end_of_DRAM())
		return size;

	if (start >= memblock_end_of_DRAM())
		return 0;

	return memblock_end_of_DRAM() - start;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Michael Ellerman
41
91.11%
1
50.00%
Yinghai Lu
4
8.89%
1
50.00%
Total
45
100.00%
2
100.00%

/*
 * Reads the counter for a given entry in
 * linux,drconf-usable-memory property
 */


static inline int __init read_usm_ranges(const __be32 **usm)
{
	/*
         * For each lmb in ibm,dynamic-memory a corresponding
         * entry in linux,drconf-usable-memory property contains
         * a counter followed by that many (base, size) duple.
         * read the counter from linux,drconf-usable-memory
         */
	return read_n_cells(n_mem_size_cells, usm);
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Chandru
21
91.30%
1
33.33%
Benjamin Herrenschmidt
1
4.35%
1
33.33%
Alistair Popple
1
4.35%
1
33.33%
Total
23
100.00%
3
100.00%

/*
 * Extract NUMA information from the ibm,dynamic-reconfiguration-memory
 * node.  This assumes n_mem_{addr,size}_cells have been set.
 */


static void __init parse_drconf_memory(struct device_node *memory)
{
	const __be32 *uninitialized_var(dm), *usm;
	unsigned int n, rc, ranges, is_kexec_kdump = 0;
	unsigned long lmb_size, base, size, sz;
	int nid;
	struct assoc_arrays aa = { .arrays = NULL };

	n = of_get_drconf_memory(memory, &dm);
	if (!n)
		return;

	lmb_size = of_get_lmb_size(memory);
	if (!lmb_size)
		return;

	rc = of_get_assoc_arrays(memory, &aa);
	if (rc)
		return;

	/* check if this is a kexec/kdump kernel */
	usm = of_get_usable_memory(memory);
	if (usm != NULL)
		is_kexec_kdump = 1;

	for (; n != 0; --n) {
		struct of_drconf_cell drmem;

		read_drconf_cell(&drmem, &dm);

		/* skip this block if the reserved bit is set in flags (0x80)
                   or if the block is not assigned to this partition (0x8) */
		if ((drmem.flags & DRCONF_MEM_RESERVED)
		    || !(drmem.flags & DRCONF_MEM_ASSIGNED))
			continue;

		base = drmem.base_addr;
		size = lmb_size;
		ranges = 1;

		if (is_kexec_kdump) {
			ranges = read_usm_ranges(&usm);
			if (!ranges) /* there are no (base, size) duple */
				continue;
		}
		do {
			if (is_kexec_kdump) {
				base = read_n_cells(n_mem_addr_cells, &usm);
				size = read_n_cells(n_mem_size_cells, &usm);
			}
			nid = of_drconf_to_nid_single(&drmem, &aa);
			fake_numa_create_new_node(
				((base + size) >> PAGE_SHIFT),
					   &nid);
			node_set_online(nid);
			sz = numa_enforce_memory_limit(base, size);
			if (sz)
				memblock_set_node(base, sz,
						  &memblock.memory, nid);
		} while (--ranges);
	}
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Paul Mackerras
113
38.18%
1
10.00%
Chandru
108
36.49%
1
10.00%
Nathan Fontenot
37
12.50%
1
10.00%
Balbir Singh
14
4.73%
1
10.00%
Benjamin Herrenschmidt
12
4.05%
2
20.00%
Tang Chen
5
1.69%
1
10.00%
Michael Neuling
3
1.01%
1
10.00%
Tejun Heo
3
1.01%
1
10.00%
Alistair Popple
1
0.34%
1
10.00%
Total
296
100.00%
10
100.00%



static int __init parse_numa_properties(void)
{
	struct device_node *memory;
	int default_nid = 0;
	unsigned long i;

	if (numa_enabled == 0) {
		printk(KERN_WARNING "NUMA disabled by user\n");
		return -1;
	}

	min_common_depth = find_min_common_depth();

	if (min_common_depth < 0)
		return min_common_depth;

	dbg("NUMA associativity depth for CPU/Memory: %d\n", min_common_depth);

	/*
         * Even though we connect cpus to numa domains later in SMP
         * init, we need to know the node ids now. This is because
         * each node to be onlined must have NODE_DATA etc backing it.
         */
	for_each_present_cpu(i) {
		struct device_node *cpu;
		int nid;

		cpu = of_get_cpu_node(i, NULL);
		BUG_ON(!cpu);
		nid = of_node_to_nid_single(cpu);
		of_node_put(cpu);

		/*
                 * Don't fall back to default_nid yet -- we will plug
                 * cpus into nodes once the memory scan has discovered
                 * the topology.
                 */
		if (nid < 0)
			continue;
		node_set_online(nid);
	}

	get_n_mem_cells(&n_mem_addr_cells, &n_mem_size_cells);

	for_each_node_by_type(memory, "memory") {
		unsigned long start;
		unsigned long size;
		int nid;
		int ranges;
		const __be32 *memcell_buf;
		unsigned int len;

		memcell_buf = of_get_property(memory,
			"linux,usable-memory", &len);
		if (!memcell_buf || len <= 0)
			memcell_buf = of_get_property(memory, "reg", &len);
		if (!memcell_buf || len <= 0)
			continue;

		/* ranges in cell */
		ranges = (len >> 2) / (n_mem_addr_cells + n_mem_size_cells);
new_range:
		/* these are order-sensitive, and modify the buffer pointer */
		start = read_n_cells(n_mem_addr_cells, &memcell_buf);
		size = read_n_cells(n_mem_size_cells, &memcell_buf);

		/*
                 * Assumption: either all memory nodes or none will
                 * have associativity properties.  If none, then
                 * everything goes to default_nid.
                 */
		nid = of_node_to_nid_single(memory);
		if (nid < 0)
			nid = default_nid;

		fake_numa_create_new_node(((start + size) >> PAGE_SHIFT), &nid);
		node_set_online(nid);

		size = numa_enforce_memory_limit(start, size);
		if (size)
			memblock_set_node(start, size, &memblock.memory, nid);

		if (--ranges)
			goto new_range;
	}

	/*
         * Now do the same thing for each MEMBLOCK listed in the
         * ibm,dynamic-memory property in the
         * ibm,dynamic-reconfiguration-memory node.
         */
	memory = of_find_node_by_path("/ibm,dynamic-reconfiguration-memory");
	if (memory)
		parse_drconf_memory(memory);

	return 0;
}

Contributors
Person
Tokens
Prop
Commits
CommitProp
Anton Blanchard
139
41.87%
8
23.53%
Andrew Morton
54
16.27%
5
14.71%
Nathan Lynch
39
11.75%
4
11.76%
Michael Ellerman
21
6.33%
2
5.88%
Paul Mackerras
19
5.72%
2
5.88%
Balbir Singh
16
4.82%
1
2.94%
Mike Kravetz
13
3.92%
3
8.82%
Benjamin Herrenschmidt
12
3.61%
1
2.94%
Tang Chen
5
1.51%
1
2.94%
Milton D. Miller II
3
0.90%
1
2.94%
Tejun Heo
3
0.90%
1
2.94%
Jeremy Kerr
3
0.90%
2
5.88%
Stephen Rothwell
2
0.60%
1
2.94%
Reza Arbab
2
0.60%
1
2.94%
Alistair Popple
1
0.30%
1
2.94%
Total
332
100.00%
34
100.00%



static void __init setup_nonnuma(void)
{
	unsigned long top_of_ram = memblock_end_of_DRAM();
	unsigned long total_ram = memblock_phys_mem_size();
	unsigned long start_pfn, end_pfn;
	unsigned int nid = 0;
	struct memblock_region *reg;