numa.c

linux 内核源代码

		cpu = find_cpu_node(i);
		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);
	memory = NULL;
	while ((memory = of_find_node_by_type(memory, "memory")) != NULL) {
		unsigned long start;
		unsigned long size;
		int nid;
		int ranges;
		const unsigned int *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;
		node_set_online(nid);

		if (!(size = numa_enforce_memory_limit(start, size))) {
			if (--ranges)
				goto new_range;
			else
				continue;
		}

		add_active_range(nid, start >> PAGE_SHIFT,
				(start >> PAGE_SHIFT) + (size >> PAGE_SHIFT));

		if (--ranges)
			goto new_range;
	}

	/*
	 * Now do the same thing for each LMB 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;
}

static void __init setup_nonnuma(void)
{
	unsigned long top_of_ram = lmb_end_of_DRAM();
	unsigned long total_ram = lmb_phys_mem_size();
	unsigned long start_pfn, end_pfn;
	unsigned int i;

	printk(KERN_DEBUG "Top of RAM: 0x%lx, Total RAM: 0x%lx\n",
	       top_of_ram, total_ram);
	printk(KERN_DEBUG "Memory hole size: %ldMB\n",
	       (top_of_ram - total_ram) >> 20);

	for (i = 0; i < lmb.memory.cnt; ++i) {
		start_pfn = lmb.memory.region[i].base >> PAGE_SHIFT;
		end_pfn = start_pfn + lmb_size_pages(&lmb.memory, i);
		add_active_range(0, start_pfn, end_pfn);
	}
	node_set_online(0);
}

void __init dump_numa_cpu_topology(void)
{
	unsigned int node;
	unsigned int cpu, count;

	if (min_common_depth == -1 || !numa_enabled)
		return;

	for_each_online_node(node) {
		printk(KERN_DEBUG "Node %d CPUs:", node);

		count = 0;
		/*
		 * If we used a CPU iterator here we would miss printing
		 * the holes in the cpumap.
		 */
		for (cpu = 0; cpu < NR_CPUS; cpu++) {
			if (cpu_isset(cpu, numa_cpumask_lookup_table[node])) {
				if (count == 0)
					printk(" %u", cpu);
				++count;
			} else {
				if (count > 1)
					printk("-%u", cpu - 1);
				count = 0;
			}
		}

		if (count > 1)
			printk("-%u", NR_CPUS - 1);
		printk("\n");
	}
}

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 < lmb_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");
	}
}

/*
 * Allocate some memory, satisfying the lmb or bootmem allocator where
 * required. nid is the preferred node and end is the physical address of
 * the highest address in the node.
 *
 * Returns the physical address of the memory.
 */
static void __init *careful_allocation(int nid, unsigned long size,
				       unsigned long align,
				       unsigned long end_pfn)
{
	int new_nid;
	unsigned long ret = __lmb_alloc_base(size, align, end_pfn << PAGE_SHIFT);

	/* retry over all memory */
	if (!ret)
		ret = __lmb_alloc_base(size, align, lmb_end_of_DRAM());

	if (!ret)
		panic("numa.c: cannot allocate %lu bytes on node %d",
		      size, nid);

	/*
	 * If the memory came from a previously allocated node, we must
	 * retry with the bootmem allocator.
	 */
	new_nid = early_pfn_to_nid(ret >> PAGE_SHIFT);
	if (new_nid < nid) {
		ret = (unsigned long)__alloc_bootmem_node(NODE_DATA(new_nid),
				size, align, 0);

		if (!ret)
			panic("numa.c: cannot allocate %lu bytes on node %d",
			      size, new_nid);

		ret = __pa(ret);

		dbg("alloc_bootmem %lx %lx\n", ret, size);
	}

	return (void *)ret;
}

static struct notifier_block __cpuinitdata ppc64_numa_nb = {
	.notifier_call = cpu_numa_callback,
	.priority = 1 /* Must run before sched domains notifier. */
};

void __init do_init_bootmem(void)
{
	int nid;
	unsigned int i;

	min_low_pfn = 0;
	max_low_pfn = lmb_end_of_DRAM() >> PAGE_SHIFT;
	max_pfn = max_low_pfn;

	if (parse_numa_properties())
		setup_nonnuma();
	else
		dump_numa_memory_topology();

	register_cpu_notifier(&ppc64_numa_nb);
	cpu_numa_callback(&ppc64_numa_nb, CPU_UP_PREPARE,
			  (void *)(unsigned long)boot_cpuid);

	for_each_online_node(nid) {
		unsigned long start_pfn, end_pfn;
		unsigned long bootmem_paddr;
		unsigned long bootmap_pages;

		get_pfn_range_for_nid(nid, &start_pfn, &end_pfn);

		/* Allocate the node structure node local if possible */
		NODE_DATA(nid) = careful_allocation(nid,
					sizeof(struct pglist_data),
					SMP_CACHE_BYTES, end_pfn);
		NODE_DATA(nid) = __va(NODE_DATA(nid));
		memset(NODE_DATA(nid), 0, sizeof(struct pglist_data));

  		dbg("node %d\n", nid);
		dbg("NODE_DATA() = %p\n", NODE_DATA(nid));

		NODE_DATA(nid)->bdata = &plat_node_bdata[nid];
		NODE_DATA(nid)->node_start_pfn = start_pfn;
		NODE_DATA(nid)->node_spanned_pages = end_pfn - start_pfn;

		if (NODE_DATA(nid)->node_spanned_pages == 0)
  			continue;

  		dbg("start_paddr = %lx\n", start_pfn << PAGE_SHIFT);
  		dbg("end_paddr = %lx\n", end_pfn << PAGE_SHIFT);

		bootmap_pages = bootmem_bootmap_pages(end_pfn - start_pfn);
		bootmem_paddr = (unsigned long)careful_allocation(nid,
					bootmap_pages << PAGE_SHIFT,
					PAGE_SIZE, end_pfn);
		memset(__va(bootmem_paddr), 0, bootmap_pages << PAGE_SHIFT);

		dbg("bootmap_paddr = %lx\n", bootmem_paddr);

		init_bootmem_node(NODE_DATA(nid), bootmem_paddr >> PAGE_SHIFT,
				  start_pfn, end_pfn);

		free_bootmem_with_active_regions(nid, end_pfn);

		/* Mark reserved regions on this node */
		for (i = 0; i < lmb.reserved.cnt; i++) {
			unsigned long physbase = lmb.reserved.region[i].base;
			unsigned long size = lmb.reserved.region[i].size;
			unsigned long start_paddr = start_pfn << PAGE_SHIFT;
			unsigned long end_paddr = end_pfn << PAGE_SHIFT;

			if (early_pfn_to_nid(physbase >> PAGE_SHIFT) != nid &&
			    early_pfn_to_nid((physbase+size-1) >> PAGE_SHIFT) != nid)
				continue;

			if (physbase < end_paddr &&
			    (physbase+size) > start_paddr) {
				/* overlaps */
				if (physbase < start_paddr) {
					size -= start_paddr - physbase;
					physbase = start_paddr;
				}

				if (size > end_paddr - physbase)
					size = end_paddr - physbase;

				dbg("reserve_bootmem %lx %lx\n", physbase,
				    size);
				reserve_bootmem_node(NODE_DATA(nid), physbase,
						     size);
			}
		}

		sparse_memory_present_with_active_regions(nid);
	}
}

void __init paging_init(void)
{
	unsigned long max_zone_pfns[MAX_NR_ZONES];
	memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
	max_zone_pfns[ZONE_DMA] = lmb_end_of_DRAM() >> PAGE_SHIFT;
	free_area_init_nodes(max_zone_pfns);
}

static int __init early_numa(char *p)
{
	if (!p)
		return 0;

	if (strstr(p, "off"))
		numa_enabled = 0;

	if (strstr(p, "debug"))
		numa_debug = 1;

	return 0;
}
early_param("numa", early_numa);

#ifdef CONFIG_MEMORY_HOTPLUG
/*
 * Find the node associated with a hot added memory section.  Section
 * corresponds to a SPARSEMEM section, not an LMB.  It is assumed that
 * sections are fully contained within a single LMB.
 */
int hot_add_scn_to_nid(unsigned long scn_addr)
{
	struct device_node *memory = NULL;
	nodemask_t nodes;
	int default_nid = any_online_node(NODE_MASK_ALL);
	int nid;

	if (!numa_enabled || (min_common_depth < 0))
		return default_nid;

	while ((memory = of_find_node_by_type(memory, "memory")) != NULL) {
		unsigned long start, size;
		int ranges;
		const unsigned int *memcell_buf;
		unsigned int len;

		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);
ha_new_range:
		start = read_n_cells(n_mem_addr_cells, &memcell_buf);
		size = read_n_cells(n_mem_size_cells, &memcell_buf);
		nid = of_node_to_nid_single(memory);

		/* Domains not present at boot default to 0 */
		if (nid < 0 || !node_online(nid))
			nid = default_nid;

		if ((scn_addr >= start) && (scn_addr < (start + size))) {
			of_node_put(memory);
			goto got_nid;
		}

		if (--ranges)		/* process all ranges in cell */
			goto ha_new_range;
	}
	BUG();	/* section address should be found above */
	return 0;

	/* Temporary code to ensure that returned node is not empty */
got_nid:
	nodes_setall(nodes);
	while (NODE_DATA(nid)->node_spanned_pages == 0) {
		node_clear(nid, nodes);
		nid = any_online_node(nodes);
	}
	return nid;
}
#endif /* CONFIG_MEMORY_HOTPLUG */