[net-next-2.6.git] / mm / page_cgroup.c

#include <linux/mm.h>
#include <linux/mmzone.h>
#include <linux/bootmem.h>
#include <linux/bit_spinlock.h>
#include <linux/page_cgroup.h>
#include <linux/hash.h>
#include <linux/slab.h>
#include <linux/memory.h>
#include <linux/vmalloc.h>
#include <linux/cgroup.h>

static void __meminit
__init_page_cgroup(struct page_cgroup *pc, unsigned long pfn)
{
	pc->flags = 0;
	pc->mem_cgroup = NULL;
	pc->page = pfn_to_page(pfn);
}
static unsigned long total_usage;

#if !defined(CONFIG_SPARSEMEM)


void __meminit pgdat_page_cgroup_init(struct pglist_data *pgdat)
{
	pgdat->node_page_cgroup = NULL;
}

struct page_cgroup *lookup_page_cgroup(struct page *page)
{
	unsigned long pfn = page_to_pfn(page);
	unsigned long offset;
	struct page_cgroup *base;

	base = NODE_DATA(page_to_nid(page))->node_page_cgroup;
	if (unlikely(!base))
		return NULL;

	offset = pfn - NODE_DATA(page_to_nid(page))->node_start_pfn;
	return base + offset;
}

static int __init alloc_node_page_cgroup(int nid)
{
	struct page_cgroup *base, *pc;
	unsigned long table_size;
	unsigned long start_pfn, nr_pages, index;

	start_pfn = NODE_DATA(nid)->node_start_pfn;
	nr_pages = NODE_DATA(nid)->node_spanned_pages;

	table_size = sizeof(struct page_cgroup) * nr_pages;

	base = __alloc_bootmem_node_nopanic(NODE_DATA(nid),
			table_size, PAGE_SIZE, __pa(MAX_DMA_ADDRESS));
	if (!base)
		return -ENOMEM;
	for (index = 0; index < nr_pages; index++) {
		pc = base + index;
		__init_page_cgroup(pc, start_pfn + index);
	}
	NODE_DATA(nid)->node_page_cgroup = base;
	total_usage += table_size;
	return 0;
}

void __init page_cgroup_init(void)
{

	int nid, fail;

	if (mem_cgroup_subsys.disabled)
		return;

	for_each_online_node(nid)  {
		fail = alloc_node_page_cgroup(nid);
		if (fail)
			goto fail;
	}
	printk(KERN_INFO "allocated %ld bytes of page_cgroup\n", total_usage);
	printk(KERN_INFO "please try cgroup_disable=memory option if you"
	" don't want\n");
	return;
fail:
	printk(KERN_CRIT "allocation of page_cgroup was failed.\n");
	printk(KERN_CRIT "please try cgroup_disable=memory boot option\n");
	panic("Out of memory");
}

#else /* CONFIG_FLAT_NODE_MEM_MAP */

struct page_cgroup *lookup_page_cgroup(struct page *page)
{
	unsigned long pfn = page_to_pfn(page);
	struct mem_section *section = __pfn_to_section(pfn);

	return section->page_cgroup + pfn;
}

/* __alloc_bootmem...() is protected by !slab_available() */
int __init_refok init_section_page_cgroup(unsigned long pfn)
{
	struct mem_section *section;
	struct page_cgroup *base, *pc;
	unsigned long table_size;
	int nid, index;

	section = __pfn_to_section(pfn);

	if (section->page_cgroup)
		return 0;

	nid = page_to_nid(pfn_to_page(pfn));

	table_size = sizeof(struct page_cgroup) * PAGES_PER_SECTION;
	if (slab_is_available()) {
		base = kmalloc_node(table_size, GFP_KERNEL, nid);
		if (!base)
			base = vmalloc_node(table_size, nid);
	} else {
		base = __alloc_bootmem_node_nopanic(NODE_DATA(nid), table_size,
				PAGE_SIZE, __pa(MAX_DMA_ADDRESS));
	}

	if (!base) {
		printk(KERN_ERR "page cgroup allocation failure\n");
		return -ENOMEM;
	}

	for (index = 0; index < PAGES_PER_SECTION; index++) {
		pc = base + index;
		__init_page_cgroup(pc, pfn + index);
	}

	section = __pfn_to_section(pfn);
	section->page_cgroup = base - pfn;
	total_usage += table_size;
	return 0;
}
#ifdef CONFIG_MEMORY_HOTPLUG
void __free_page_cgroup(unsigned long pfn)
{
	struct mem_section *ms;
	struct page_cgroup *base;

	ms = __pfn_to_section(pfn);
	if (!ms || !ms->page_cgroup)
		return;
	base = ms->page_cgroup + pfn;
	if (is_vmalloc_addr(base)) {
		vfree(base);
		ms->page_cgroup = NULL;
	} else {
		struct page *page = virt_to_page(base);
		if (!PageReserved(page)) { /* Is bootmem ? */
			kfree(base);
			ms->page_cgroup = NULL;
		}
	}
}

int __meminit online_page_cgroup(unsigned long start_pfn,
			unsigned long nr_pages,
			int nid)
{
	unsigned long start, end, pfn;
	int fail = 0;

	start = start_pfn & ~(PAGES_PER_SECTION - 1);
	end = ALIGN(start_pfn + nr_pages, PAGES_PER_SECTION);

	for (pfn = start; !fail && pfn < end; pfn += PAGES_PER_SECTION) {
		if (!pfn_present(pfn))
			continue;
		fail = init_section_page_cgroup(pfn);
	}
	if (!fail)
		return 0;

	/* rollback */
	for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION)
		__free_page_cgroup(pfn);

	return -ENOMEM;
}

int __meminit offline_page_cgroup(unsigned long start_pfn,
		unsigned long nr_pages, int nid)
{
	unsigned long start, end, pfn;

	start = start_pfn & ~(PAGES_PER_SECTION - 1);
	end = ALIGN(start_pfn + nr_pages, PAGES_PER_SECTION);

	for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION)
		__free_page_cgroup(pfn);
	return 0;

}

static int __meminit page_cgroup_callback(struct notifier_block *self,
			       unsigned long action, void *arg)
{
	struct memory_notify *mn = arg;
	int ret = 0;
	switch (action) {
	case MEM_GOING_ONLINE:
		ret = online_page_cgroup(mn->start_pfn,
				   mn->nr_pages, mn->status_change_nid);
		break;
	case MEM_CANCEL_ONLINE:
	case MEM_OFFLINE:
		offline_page_cgroup(mn->start_pfn,
				mn->nr_pages, mn->status_change_nid);
		break;
	case MEM_GOING_OFFLINE:
		break;
	case MEM_ONLINE:
	case MEM_CANCEL_OFFLINE:
		break;
	}
	ret = notifier_from_errno(ret);
	return ret;
}

#endif

void __init page_cgroup_init(void)
{
	unsigned long pfn;
	int fail = 0;

	if (mem_cgroup_subsys.disabled)
		return;

	for (pfn = 0; !fail && pfn < max_pfn; pfn += PAGES_PER_SECTION) {
		if (!pfn_present(pfn))
			continue;
		fail = init_section_page_cgroup(pfn);
	}
	if (fail) {
		printk(KERN_CRIT "try cgroup_disable=memory boot option\n");
		panic("Out of memory");
	} else {
		hotplug_memory_notifier(page_cgroup_callback, 0);
	}
	printk(KERN_INFO "allocated %ld bytes of page_cgroup\n", total_usage);
	printk(KERN_INFO "please try cgroup_disable=memory option if you don't"
	" want\n");
}

void __meminit pgdat_page_cgroup_init(struct pglist_data *pgdat)
{
	return;
}

#endif
Commit	Line	Data
52d4b9ac KH	1	#include <linux/mm.h>
	2	#include <linux/mmzone.h>
	3	#include <linux/bootmem.h>
	4	#include <linux/bit_spinlock.h>
	5	#include <linux/page_cgroup.h>
	6	#include <linux/hash.h>
94b6da5a	7	#include <linux/slab.h>
52d4b9ac	8	#include <linux/memory.h>
4c821042	9	#include <linux/vmalloc.h>
94b6da5a	10	#include <linux/cgroup.h>
52d4b9ac KH	11
	12	static void __meminit
	13	__init_page_cgroup(struct page_cgroup *pc, unsigned long pfn)
	14	{
	15	pc->flags = 0;
	16	pc->mem_cgroup = NULL;
	17	pc->page = pfn_to_page(pfn);
	18	}
	19	static unsigned long total_usage;
	20
	21	#if !defined(CONFIG_SPARSEMEM)
	22
	23
31168481	24	void __meminit pgdat_page_cgroup_init(struct pglist_data *pgdat)
52d4b9ac KH	25	{
	26	pgdat->node_page_cgroup = NULL;
	27	}
	28
	29	struct page_cgroup lookup_page_cgroup(struct page page)
	30	{
	31	unsigned long pfn = page_to_pfn(page);
	32	unsigned long offset;
	33	struct page_cgroup *base;
	34
	35	base = NODE_DATA(page_to_nid(page))->node_page_cgroup;
	36	if (unlikely(!base))
	37	return NULL;
	38
	39	offset = pfn - NODE_DATA(page_to_nid(page))->node_start_pfn;
	40	return base + offset;
	41	}
	42
	43	static int __init alloc_node_page_cgroup(int nid)
	44	{
	45	struct page_cgroup base, pc;
	46	unsigned long table_size;
	47	unsigned long start_pfn, nr_pages, index;
	48
	49	start_pfn = NODE_DATA(nid)->node_start_pfn;
	50	nr_pages = NODE_DATA(nid)->node_spanned_pages;
	51
	52	table_size = sizeof(struct page_cgroup) * nr_pages;
	53
	54	base = __alloc_bootmem_node_nopanic(NODE_DATA(nid),
	55	table_size, PAGE_SIZE, __pa(MAX_DMA_ADDRESS));
	56	if (!base)
	57	return -ENOMEM;
	58	for (index = 0; index < nr_pages; index++) {
	59	pc = base + index;
	60	__init_page_cgroup(pc, start_pfn + index);
	61	}
	62	NODE_DATA(nid)->node_page_cgroup = base;
	63	total_usage += table_size;
	64	return 0;
	65	}
	66
	67	void __init page_cgroup_init(void)
	68	{
	69
	70	int nid, fail;
	71
94b6da5a KH	72	if (mem_cgroup_subsys.disabled)
	73	return;
	74
52d4b9ac KH	75	for_each_online_node(nid) {
	76	fail = alloc_node_page_cgroup(nid);
	77	if (fail)
	78	goto fail;
	79	}
	80	printk(KERN_INFO "allocated %ld bytes of page_cgroup\n", total_usage);
	81	printk(KERN_INFO "please try cgroup_disable=memory option if you"
	82	" don't want\n");
	83	return;
	84	fail:
	85	printk(KERN_CRIT "allocation of page_cgroup was failed.\n");
	86	printk(KERN_CRIT "please try cgroup_disable=memory boot option\n");
	87	panic("Out of memory");
	88	}
	89
	90	#else /* CONFIG_FLAT_NODE_MEM_MAP */
	91
	92	struct page_cgroup lookup_page_cgroup(struct page page)
	93	{
	94	unsigned long pfn = page_to_pfn(page);
	95	struct mem_section *section = __pfn_to_section(pfn);
	96
	97	return section->page_cgroup + pfn;
	98	}
	99
31168481 AV	100	/* __alloc_bootmem...() is protected by !slab_available() */
31168481 AV	101	int __init_refok init_section_page_cgroup(unsigned long pfn)
52d4b9ac KH	102	{
	103	struct mem_section *section;
	104	struct page_cgroup base, pc;
	105	unsigned long table_size;
	106	int nid, index;
	107
	108	section = __pfn_to_section(pfn);
	109
	110	if (section->page_cgroup)
	111	return 0;
	112
	113	nid = page_to_nid(pfn_to_page(pfn));
	114
	115	table_size = sizeof(struct page_cgroup) * PAGES_PER_SECTION;
94b6da5a KH	116	if (slab_is_available()) {
	117	base = kmalloc_node(table_size, GFP_KERNEL, nid);
	118	if (!base)
	119	base = vmalloc_node(table_size, nid);
	120	} else {
	121	base = __alloc_bootmem_node_nopanic(NODE_DATA(nid), table_size,
	122	PAGE_SIZE, __pa(MAX_DMA_ADDRESS));
	123	}
52d4b9ac KH	124
	125	if (!base) {
	126	printk(KERN_ERR "page cgroup allocation failure\n");
	127	return -ENOMEM;
	128	}
	129
	130	for (index = 0; index < PAGES_PER_SECTION; index++) {
	131	pc = base + index;
	132	__init_page_cgroup(pc, pfn + index);
	133	}
	134
	135	section = __pfn_to_section(pfn);
	136	section->page_cgroup = base - pfn;
	137	total_usage += table_size;
	138	return 0;
	139	}
	140	#ifdef CONFIG_MEMORY_HOTPLUG
	141	void __free_page_cgroup(unsigned long pfn)
	142	{
	143	struct mem_section *ms;
	144	struct page_cgroup *base;
	145
	146	ms = __pfn_to_section(pfn);
	147	if (!ms \|\| !ms->page_cgroup)
	148	return;
	149	base = ms->page_cgroup + pfn;
94b6da5a	150	if (is_vmalloc_addr(base)) {
52d4b9ac	151	vfree(base);
94b6da5a KH	152	ms->page_cgroup = NULL;
	153	} else {
	154	struct page *page = virt_to_page(base);
	155	if (!PageReserved(page)) { /* Is bootmem ? */
	156	kfree(base);
	157	ms->page_cgroup = NULL;
	158	}
	159	}
52d4b9ac KH	160	}
52d4b9ac KH	161
31168481	162	int __meminit online_page_cgroup(unsigned long start_pfn,
52d4b9ac KH	163	unsigned long nr_pages,
	164	int nid)
	165	{
	166	unsigned long start, end, pfn;
	167	int fail = 0;
	168
33c5d3d6	169	start = start_pfn & ~(PAGES_PER_SECTION - 1);
52d4b9ac KH	170	end = ALIGN(start_pfn + nr_pages, PAGES_PER_SECTION);
	171
	172	for (pfn = start; !fail && pfn < end; pfn += PAGES_PER_SECTION) {
	173	if (!pfn_present(pfn))
	174	continue;
	175	fail = init_section_page_cgroup(pfn);
	176	}
	177	if (!fail)
	178	return 0;
	179
	180	/* rollback */
	181	for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION)
	182	__free_page_cgroup(pfn);
	183
	184	return -ENOMEM;
	185	}
	186
31168481	187	int __meminit offline_page_cgroup(unsigned long start_pfn,
52d4b9ac KH	188	unsigned long nr_pages, int nid)
	189	{
	190	unsigned long start, end, pfn;
	191
33c5d3d6	192	start = start_pfn & ~(PAGES_PER_SECTION - 1);
52d4b9ac KH	193	end = ALIGN(start_pfn + nr_pages, PAGES_PER_SECTION);
	194
	195	for (pfn = start; pfn < end; pfn += PAGES_PER_SECTION)
	196	__free_page_cgroup(pfn);
	197	return 0;
	198
	199	}
	200
31168481	201	static int __meminit page_cgroup_callback(struct notifier_block *self,
52d4b9ac KH	202	unsigned long action, void *arg)
	203	{
	204	struct memory_notify *mn = arg;
	205	int ret = 0;
	206	switch (action) {
	207	case MEM_GOING_ONLINE:
	208	ret = online_page_cgroup(mn->start_pfn,
	209	mn->nr_pages, mn->status_change_nid);
	210	break;
	211	case MEM_CANCEL_ONLINE:
	212	case MEM_OFFLINE:
	213	offline_page_cgroup(mn->start_pfn,
	214	mn->nr_pages, mn->status_change_nid);
	215	break;
	216	case MEM_GOING_OFFLINE:
	217	break;
	218	case MEM_ONLINE:
	219	case MEM_CANCEL_OFFLINE:
	220	break;
	221	}
	222	ret = notifier_from_errno(ret);
	223	return ret;
	224	}
	225
	226	#endif
	227
	228	void __init page_cgroup_init(void)
	229	{
	230	unsigned long pfn;
	231	int fail = 0;
	232
94b6da5a KH	233	if (mem_cgroup_subsys.disabled)
	234	return;
	235
52d4b9ac KH	236	for (pfn = 0; !fail && pfn < max_pfn; pfn += PAGES_PER_SECTION) {
	237	if (!pfn_present(pfn))
	238	continue;
	239	fail = init_section_page_cgroup(pfn);
	240	}
	241	if (fail) {
	242	printk(KERN_CRIT "try cgroup_disable=memory boot option\n");
	243	panic("Out of memory");
	244	} else {
	245	hotplug_memory_notifier(page_cgroup_callback, 0);
	246	}
	247	printk(KERN_INFO "allocated %ld bytes of page_cgroup\n", total_usage);
	248	printk(KERN_INFO "please try cgroup_disable=memory option if you don't"
	249	" want\n");
	250	}
	251
31168481	252	void __meminit pgdat_page_cgroup_init(struct pglist_data *pgdat)
52d4b9ac KH	253	{
	254	return;
	255	}
	256
	257	#endif