numa.c

Go to the documentation of this file.

/* Common code for 32 and 64-bit NUMA */
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/string.h>
#include <linux/init.h>
#include <linux/bootmem.h>
#include <linux/memblock.h>
#include <linux/mmzone.h>
#include <linux/ctype.h>
#include <linux/module.h>
#include <linux/nodemask.h>
#include <linux/sched.h>
#include <linux/topology.h>

#include <asm/e820.h>
#include <asm/proto.h>
#include <asm/dma.h>
#include <asm/acpi.h>
#include <asm/amd_nb.h>

#include "numa_internal.h"

int __initdata numa_off;
nodemask_t numa_nodes_parsed __initdata;

struct pglist_data *node_data[MAX_NUMNODES] __read_mostly;
EXPORT_SYMBOL(node_data);

static struct numa_meminfo numa_meminfo
#ifndef CONFIG_MEMORY_HOTPLUG
__initdata
#endif
;

static int numa_distance_cnt;
static u8 *numa_distance;

static __init int numa_setup(char *opt)
{
    if (!opt)
        return -EINVAL;
    if (!strncmp(opt, "off", 3))
        numa_off = 1;
#ifdef CONFIG_NUMA_EMU
    if (!strncmp(opt, "fake=", 5))
        numa_emu_cmdline(opt + 5);
#endif
#ifdef CONFIG_ACPI_NUMA
    if (!strncmp(opt, "noacpi", 6))
        acpi_numa = -1;
#endif
    return 0;
}
early_param("numa", numa_setup);

/*
 * apicid, cpu, node mappings
 */
s16 __apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = {
    [0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
};

int __cpuinit numa_cpu_node(int cpu)
{
    int apicid = early_per_cpu(x86_cpu_to_apicid, cpu);

    if (apicid != BAD_APICID)
        return __apicid_to_node[apicid];
    return NUMA_NO_NODE;
}

cpumask_var_t node_to_cpumask_map[MAX_NUMNODES];
EXPORT_SYMBOL(node_to_cpumask_map);

/*
 * Map cpu index to node index
 */
DEFINE_EARLY_PER_CPU(int, x86_cpu_to_node_map, NUMA_NO_NODE);
EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_node_map);

void __cpuinit numa_set_node(int cpu, int node)
{
    int *cpu_to_node_map = early_per_cpu_ptr(x86_cpu_to_node_map);

    /* early setting, no percpu area yet */
    if (cpu_to_node_map) {
        cpu_to_node_map[cpu] = node;
        return;
    }

#ifdef CONFIG_DEBUG_PER_CPU_MAPS
    if (cpu >= nr_cpu_ids || !cpu_possible(cpu)) {
        printk(KERN_ERR "numa_set_node: invalid cpu# (%d)\n", cpu);
        dump_stack();
        return;
    }
#endif
    per_cpu(x86_cpu_to_node_map, cpu) = node;

    if (node != NUMA_NO_NODE)
        set_cpu_numa_node(cpu, node);
}

void __cpuinit numa_clear_node(int cpu)
{
    numa_set_node(cpu, NUMA_NO_NODE);
}

/*
 * 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.
 * (Use CONFIG_DEBUG_PER_CPU_MAPS to check this.)
 */
void __init setup_node_to_cpumask_map(void)
{
    unsigned int node, num = 0;

    /* setup nr_node_ids if not done yet */
    if (nr_node_ids == MAX_NUMNODES) {
        for_each_node_mask(node, node_possible_map)
            num = node;
        nr_node_ids = num + 1;
    }

    /* allocate the map */
    for (node = 0; node < nr_node_ids; node++)
        alloc_bootmem_cpumask_var(&node_to_cpumask_map[node]);

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

static int __init numa_add_memblk_to(int nid, u64 start, u64 end,
                     struct numa_meminfo *mi)
{
    /* ignore zero length blks */
    if (start == end)
        return 0;

    /* whine about and ignore invalid blks */
    if (start > end || nid < 0 || nid >= MAX_NUMNODES) {
        pr_warning("NUMA: Warning: invalid memblk node %d [mem %#010Lx-%#010Lx]\n",
               nid, start, end - 1);
        return 0;
    }

    if (mi->nr_blks >= NR_NODE_MEMBLKS) {
        pr_err("NUMA: too many memblk ranges\n");
        return -EINVAL;
    }

    mi->blk[mi->nr_blks].start = start;
    mi->blk[mi->nr_blks].end = end;
    mi->blk[mi->nr_blks].nid = nid;
    mi->nr_blks++;
    return 0;
}

void __init numa_remove_memblk_from(int idx, struct numa_meminfo *mi)
{
    mi->nr_blks--;
    memmove(&mi->blk[idx], &mi->blk[idx + 1],
        (mi->nr_blks - idx) * sizeof(mi->blk[0]));
}

int __init numa_add_memblk(int nid, u64 start, u64 end)
{
    return numa_add_memblk_to(nid, start, end, &numa_meminfo);
}

/* Initialize NODE_DATA for a node on the local memory */
static void __init setup_node_data(int nid, u64 start, u64 end)
{
    const size_t nd_size = roundup(sizeof(pg_data_t), PAGE_SIZE);
    bool remapped = false;
    u64 nd_pa;
    void *nd;
    int tnid;

    /*
     * Don't confuse VM with a node that doesn't have the
     * minimum amount of memory:
     */
    if (end && (end - start) < NODE_MIN_SIZE)
        return;

    /* initialize remap allocator before aligning to ZONE_ALIGN */
    init_alloc_remap(nid, start, end);

    start = roundup(start, ZONE_ALIGN);

    printk(KERN_INFO "Initmem setup node %d [mem %#010Lx-%#010Lx]\n",
           nid, start, end - 1);

    /*
     * Allocate node data.  Try remap allocator first, node-local
     * memory and then any node.  Never allocate in DMA zone.
     */
    nd = alloc_remap(nid, nd_size);
    if (nd) {
        nd_pa = __pa(nd);
        remapped = true;
    } else {
        nd_pa = memblock_alloc_nid(nd_size, SMP_CACHE_BYTES, nid);
        if (!nd_pa) {
            pr_err("Cannot find %zu bytes in node %d\n",
                   nd_size, nid);
            return;
        }
        nd = __va(nd_pa);
    }

    /* report and initialize */
    printk(KERN_INFO "  NODE_DATA [mem %#010Lx-%#010Lx]%s\n",
           nd_pa, nd_pa + nd_size - 1, remapped ? " (remapped)" : "");
    tnid = early_pfn_to_nid(nd_pa >> PAGE_SHIFT);
    if (!remapped && tnid != nid)
        printk(KERN_INFO "    NODE_DATA(%d) on node %d\n", nid, tnid);

    node_data[nid] = nd;
    memset(NODE_DATA(nid), 0, sizeof(pg_data_t));
    NODE_DATA(nid)->node_id = nid;
    NODE_DATA(nid)->node_start_pfn = start >> PAGE_SHIFT;
    NODE_DATA(nid)->node_spanned_pages = (end - start) >> PAGE_SHIFT;

    node_set_online(nid);
}

int __init numa_cleanup_meminfo(struct numa_meminfo *mi)
{
    const u64 low = 0;
    const u64 high = PFN_PHYS(max_pfn);
    int i, j, k;

    /* first, trim all entries */
    for (i = 0; i < mi->nr_blks; i++) {
        struct numa_memblk *bi = &mi->blk[i];

        /* make sure all blocks are inside the limits */
        bi->start = max(bi->start, low);
        bi->end = min(bi->end, high);

        /* and there's no empty block */
        if (bi->start >= bi->end)
            numa_remove_memblk_from(i--, mi);
    }

    /* merge neighboring / overlapping entries */
    for (i = 0; i < mi->nr_blks; i++) {
        struct numa_memblk *bi = &mi->blk[i];

        for (j = i + 1; j < mi->nr_blks; j++) {
            struct numa_memblk *bj = &mi->blk[j];
            u64 start, end;

            /*
             * See whether there are overlapping blocks.  Whine
             * about but allow overlaps of the same nid.  They
             * will be merged below.
             */
            if (bi->end > bj->start && bi->start < bj->end) {
                if (bi->nid != bj->nid) {
                    pr_err("NUMA: node %d [mem %#010Lx-%#010Lx] overlaps with node %d [mem %#010Lx-%#010Lx]\n",
                           bi->nid, bi->start, bi->end - 1,
                           bj->nid, bj->start, bj->end - 1);
                    return -EINVAL;
                }
                pr_warning("NUMA: Warning: node %d [mem %#010Lx-%#010Lx] overlaps with itself [mem %#010Lx-%#010Lx]\n",
                       bi->nid, bi->start, bi->end - 1,
                       bj->start, bj->end - 1);
            }

            /*
             * Join together blocks on the same node, holes
             * between which don't overlap with memory on other
             * nodes.
             */
            if (bi->nid != bj->nid)
                continue;
            start = min(bi->start, bj->start);
            end = max(bi->end, bj->end);
            for (k = 0; k < mi->nr_blks; k++) {
                struct numa_memblk *bk = &mi->blk[k];

                if (bi->nid == bk->nid)
                    continue;
                if (start < bk->end && end > bk->start)
                    break;
            }
            if (k < mi->nr_blks)
                continue;
            printk(KERN_INFO "NUMA: Node %d [mem %#010Lx-%#010Lx] + [mem %#010Lx-%#010Lx] -> [mem %#010Lx-%#010Lx]\n",
                   bi->nid, bi->start, bi->end - 1, bj->start,
                   bj->end - 1, start, end - 1);
            bi->start = start;
            bi->end = end;
            numa_remove_memblk_from(j--, mi);
        }
    }

    /* clear unused ones */
    for (i = mi->nr_blks; i < ARRAY_SIZE(mi->blk); i++) {
        mi->blk[i].start = mi->blk[i].end = 0;
        mi->blk[i].nid = NUMA_NO_NODE;
    }

    return 0;
}

/*
 * Set nodes, which have memory in @mi, in *@nodemask.
 */
static void __init numa_nodemask_from_meminfo(nodemask_t *nodemask,
                          const struct numa_meminfo *mi)
{
    int i;

    for (i = 0; i < ARRAY_SIZE(mi->blk); i++)
        if (mi->blk[i].start != mi->blk[i].end &&
            mi->blk[i].nid != NUMA_NO_NODE)
            node_set(mi->blk[i].nid, *nodemask);
}

void __init numa_reset_distance(void)
{
    size_t size = numa_distance_cnt * numa_distance_cnt * sizeof(numa_distance[0]);

    /* numa_distance could be 1LU marking allocation failure, test cnt */
    if (numa_distance_cnt)
        memblock_free(__pa(numa_distance), size);
    numa_distance_cnt = 0;
    numa_distance = NULL;   /* enable table creation */
}

static int __init numa_alloc_distance(void)
{
    nodemask_t nodes_parsed;
    size_t size;
    int i, j, cnt = 0;
    u64 phys;

    /* size the new table and allocate it */
    nodes_parsed = numa_nodes_parsed;
    numa_nodemask_from_meminfo(&nodes_parsed, &numa_meminfo);

    for_each_node_mask(i, nodes_parsed)
        cnt = i;
    cnt++;
    size = cnt * cnt * sizeof(numa_distance[0]);

    phys = memblock_find_in_range(0, PFN_PHYS(max_pfn_mapped),
                      size, PAGE_SIZE);
    if (!phys) {
        pr_warning("NUMA: Warning: can't allocate distance table!\n");
        /* don't retry until explicitly reset */
        numa_distance = (void *)1LU;
        return -ENOMEM;
    }
    memblock_reserve(phys, size);

    numa_distance = __va(phys);
    numa_distance_cnt = cnt;

    /* fill with the default distances */
    for (i = 0; i < cnt; i++)
        for (j = 0; j < cnt; j++)
            numa_distance[i * cnt + j] = i == j ?
                LOCAL_DISTANCE : REMOTE_DISTANCE;
    printk(KERN_DEBUG "NUMA: Initialized distance table, cnt=%d\n", cnt);

    return 0;
}

void __init numa_set_distance(int from, int to, int distance)
{
    if (!numa_distance && numa_alloc_distance() < 0)
        return;

    if (from >= numa_distance_cnt || to >= numa_distance_cnt ||
            from < 0 || to < 0) {
        pr_warn_once("NUMA: Warning: node ids are out of bound, from=%d to=%d distance=%d\n",
                from, to, distance);
        return;
    }

    if ((u8)distance != distance ||
        (from == to && distance != LOCAL_DISTANCE)) {
        pr_warn_once("NUMA: Warning: invalid distance parameter, from=%d to=%d distance=%d\n",
                 from, to, distance);
        return;
    }

    numa_distance[from * numa_distance_cnt + to] = distance;
}

int __node_distance(int from, int to)
{
    if (from >= numa_distance_cnt || to >= numa_distance_cnt)
        return from == to ? LOCAL_DISTANCE : REMOTE_DISTANCE;
    return numa_distance[from * numa_distance_cnt + to];
}
EXPORT_SYMBOL(__node_distance);

/*
 * Sanity check to catch more bad NUMA configurations (they are amazingly
 * common).  Make sure the nodes cover all memory.
 */
static bool __init numa_meminfo_cover_memory(const struct numa_meminfo *mi)
{
    u64 numaram, e820ram;
    int i;

    numaram = 0;
    for (i = 0; i < mi->nr_blks; i++) {
        u64 s = mi->blk[i].start >> PAGE_SHIFT;
        u64 e = mi->blk[i].end >> PAGE_SHIFT;
        numaram += e - s;
        numaram -= __absent_pages_in_range(mi->blk[i].nid, s, e);
        if ((s64)numaram < 0)
            numaram = 0;
    }

    e820ram = max_pfn - absent_pages_in_range(0, max_pfn);

    /* We seem to lose 3 pages somewhere. Allow 1M of slack. */
    if ((s64)(e820ram - numaram) >= (1 << (20 - PAGE_SHIFT))) {
        printk(KERN_ERR "NUMA: nodes only cover %LuMB of your %LuMB e820 RAM. Not used.\n",
               (numaram << PAGE_SHIFT) >> 20,
               (e820ram << PAGE_SHIFT) >> 20);
        return false;
    }
    return true;
}

static int __init numa_register_memblks(struct numa_meminfo *mi)
{
    unsigned long uninitialized_var(pfn_align);
    int i, nid;

    /* Account for nodes with cpus and no memory */
    node_possible_map = numa_nodes_parsed;
    numa_nodemask_from_meminfo(&node_possible_map, mi);
    if (WARN_ON(nodes_empty(node_possible_map)))
        return -EINVAL;

    for (i = 0; i < mi->nr_blks; i++) {
        struct numa_memblk *mb = &mi->blk[i];
        memblock_set_node(mb->start, mb->end - mb->start, mb->nid);
    }

    /*
     * If sections array is gonna be used for pfn -> nid mapping, check
     * whether its granularity is fine enough.
     */
#ifdef NODE_NOT_IN_PAGE_FLAGS
    pfn_align = node_map_pfn_alignment();
    if (pfn_align && pfn_align < PAGES_PER_SECTION) {
        printk(KERN_WARNING "Node alignment %LuMB < min %LuMB, rejecting NUMA config\n",
               PFN_PHYS(pfn_align) >> 20,
               PFN_PHYS(PAGES_PER_SECTION) >> 20);
        return -EINVAL;
    }
#endif
    if (!numa_meminfo_cover_memory(mi))
        return -EINVAL;

    /* Finally register nodes. */
    for_each_node_mask(nid, node_possible_map) {
        u64 start = PFN_PHYS(max_pfn);
        u64 end = 0;

        for (i = 0; i < mi->nr_blks; i++) {
            if (nid != mi->blk[i].nid)
                continue;
            start = min(mi->blk[i].start, start);
            end = max(mi->blk[i].end, end);
        }

        if (start < end)
            setup_node_data(nid, start, end);
    }

    /* Dump memblock with node info and return. */
    memblock_dump_all();
    return 0;
}

/*
 * There are unfortunately some poorly designed mainboards around that
 * only connect memory to a single CPU. This breaks the 1:1 cpu->node
 * mapping. To avoid this fill in the mapping for all possible CPUs,
 * as the number of CPUs is not known yet. We round robin the existing
 * nodes.
 */
static void __init numa_init_array(void)
{
    int rr, i;

    rr = first_node(node_online_map);
    for (i = 0; i < nr_cpu_ids; i++) {
        if (early_cpu_to_node(i) != NUMA_NO_NODE)
            continue;
        numa_set_node(i, rr);
        rr = next_node(rr, node_online_map);
        if (rr == MAX_NUMNODES)
            rr = first_node(node_online_map);
    }
}

static int __init numa_init(int (*init_func)(void))
{
    int i;
    int ret;

    for (i = 0; i < MAX_LOCAL_APIC; i++)
        set_apicid_to_node(i, NUMA_NO_NODE);

    nodes_clear(numa_nodes_parsed);
    nodes_clear(node_possible_map);
    nodes_clear(node_online_map);
    memset(&numa_meminfo, 0, sizeof(numa_meminfo));
    WARN_ON(memblock_set_node(0, ULLONG_MAX, MAX_NUMNODES));
    numa_reset_distance();

    ret = init_func();
    if (ret < 0)
        return ret;
    ret = numa_cleanup_meminfo(&numa_meminfo);
    if (ret < 0)
        return ret;

    numa_emulation(&numa_meminfo, numa_distance_cnt);

    ret = numa_register_memblks(&numa_meminfo);
    if (ret < 0)
        return ret;

    for (i = 0; i < nr_cpu_ids; i++) {
        int nid = early_cpu_to_node(i);

        if (nid == NUMA_NO_NODE)
            continue;
        if (!node_online(nid))
            numa_clear_node(i);
    }
    numa_init_array();
    return 0;
}

static int __init dummy_numa_init(void)
{
    printk(KERN_INFO "%s\n",
           numa_off ? "NUMA turned off" : "No NUMA configuration found");
    printk(KERN_INFO "Faking a node at [mem %#018Lx-%#018Lx]\n",
           0LLU, PFN_PHYS(max_pfn) - 1);

    node_set(0, numa_nodes_parsed);
    numa_add_memblk(0, 0, PFN_PHYS(max_pfn));

    return 0;
}

void __init x86_numa_init(void)
{
    if (!numa_off) {
#ifdef CONFIG_X86_NUMAQ
        if (!numa_init(numaq_numa_init))
            return;
#endif
#ifdef CONFIG_ACPI_NUMA
        if (!numa_init(x86_acpi_numa_init))
            return;
#endif
#ifdef CONFIG_AMD_NUMA
        if (!numa_init(amd_numa_init))
            return;
#endif
    }

    numa_init(dummy_numa_init);
}

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.
 *
 * Populate cpu_to_node[] only if x86_cpu_to_apicid[],
 * and apicid_to_node[] tables have valid entries for a CPU.
 * This means we skip cpu_to_node[] initialisation for NUMA
 * emulation and faking node case (when running a kernel compiled
 * for NUMA on a non NUMA box), which is OK as cpu_to_node[]
 * is already initialized in a round robin manner at numa_init_array,
 * prior to this call, and this initialization is good enough
 * for the fake NUMA cases.
 *
 * Called before the per_cpu areas are setup.
 */
void __init init_cpu_to_node(void)
{
    int cpu;
    u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid);

    BUG_ON(cpu_to_apicid == NULL);

    for_each_possible_cpu(cpu) {
        int node = numa_cpu_node(cpu);

        if (node == NUMA_NO_NODE)
            continue;
        if (!node_online(node))
            node = find_near_online_node(node);
        numa_set_node(cpu, node);
    }
}

#ifndef CONFIG_DEBUG_PER_CPU_MAPS

# ifndef CONFIG_NUMA_EMU
void __cpuinit numa_add_cpu(int cpu)
{
    cpumask_set_cpu(cpu, node_to_cpumask_map[early_cpu_to_node(cpu)]);
}

void __cpuinit numa_remove_cpu(int cpu)
{
    cpumask_clear_cpu(cpu, node_to_cpumask_map[early_cpu_to_node(cpu)]);
}
# endif /* !CONFIG_NUMA_EMU */

#else   /* !CONFIG_DEBUG_PER_CPU_MAPS */

int __cpu_to_node(int cpu)
{
    if (early_per_cpu_ptr(x86_cpu_to_node_map)) {
        printk(KERN_WARNING
            "cpu_to_node(%d): usage too early!\n", cpu);
        dump_stack();
        return early_per_cpu_ptr(x86_cpu_to_node_map)[cpu];
    }
    return per_cpu(x86_cpu_to_node_map, cpu);
}
EXPORT_SYMBOL(__cpu_to_node);

/*
 * Same function as cpu_to_node() but used if called before the
 * per_cpu areas are setup.
 */
int early_cpu_to_node(int cpu)
{
    if (early_per_cpu_ptr(x86_cpu_to_node_map))
        return early_per_cpu_ptr(x86_cpu_to_node_map)[cpu];

    if (!cpu_possible(cpu)) {
        printk(KERN_WARNING
            "early_cpu_to_node(%d): no per_cpu area!\n", cpu);
        dump_stack();
        return NUMA_NO_NODE;
    }
    return per_cpu(x86_cpu_to_node_map, cpu);
}

void debug_cpumask_set_cpu(int cpu, int node, bool enable)
{
    struct cpumask *mask;
    char buf[64];

    if (node == NUMA_NO_NODE) {
        /* early_cpu_to_node() already emits a warning and trace */
        return;
    }
    mask = node_to_cpumask_map[node];
    if (!mask) {
        pr_err("node_to_cpumask_map[%i] NULL\n", node);
        dump_stack();
        return;
    }

    if (enable)
        cpumask_set_cpu(cpu, mask);
    else
        cpumask_clear_cpu(cpu, mask);

    cpulist_scnprintf(buf, sizeof(buf), mask);
    printk(KERN_DEBUG "%s cpu %d node %d: mask now %s\n",
        enable ? "numa_add_cpu" : "numa_remove_cpu",
        cpu, node, buf);
    return;
}

# ifndef CONFIG_NUMA_EMU
static void __cpuinit numa_set_cpumask(int cpu, bool enable)
{
    debug_cpumask_set_cpu(cpu, early_cpu_to_node(cpu), enable);
}

void __cpuinit numa_add_cpu(int cpu)
{
    numa_set_cpumask(cpu, true);
}

void __cpuinit numa_remove_cpu(int cpu)
{
    numa_set_cpumask(cpu, false);
}
# endif /* !CONFIG_NUMA_EMU */

/*
 * Returns a pointer to the bitmask of CPUs on Node 'node'.
 */
const struct cpumask *cpumask_of_node(int node)
{
    if (node >= nr_node_ids) {
        printk(KERN_WARNING
            "cpumask_of_node(%d): node > nr_node_ids(%d)\n",
            node, nr_node_ids);
        dump_stack();
        return cpu_none_mask;
    }
    if (node_to_cpumask_map[node] == NULL) {
        printk(KERN_WARNING
            "cpumask_of_node(%d): no node_to_cpumask_map!\n",
            node);
        dump_stack();
        return cpu_online_mask;
    }
    return node_to_cpumask_map[node];
}
EXPORT_SYMBOL(cpumask_of_node);

#endif  /* !CONFIG_DEBUG_PER_CPU_MAPS */

#ifdef CONFIG_MEMORY_HOTPLUG
int memory_add_physaddr_to_nid(u64 start)
{
    struct numa_meminfo *mi = &numa_meminfo;
    int nid = mi->blk[0].nid;
    int i;

    for (i = 0; i < mi->nr_blks; i++)
        if (mi->blk[i].start <= start && mi->blk[i].end > start)
            nid = mi->blk[i].nid;
    return nid;
}
EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid);
#endif