[GitHub/exynos8895/android_kernel_samsung_universal8895.git] / mm / sparse-vmemmap.c

/*
 * Virtual Memory Map support
 *
 * (C) 2007 sgi. Christoph Lameter.
 *
 * Virtual memory maps allow VM primitives pfn_to_page, page_to_pfn,
 * virt_to_page, page_address() to be implemented as a base offset
 * calculation without memory access.
 *
 * However, virtual mappings need a page table and TLBs. Many Linux
 * architectures already map their physical space using 1-1 mappings
 * via TLBs. For those arches the virtual memory map is essentially
 * for free if we use the same page size as the 1-1 mappings. In that
 * case the overhead consists of a few additional pages that are
 * allocated to create a view of memory for vmemmap.
 *
 * The architecture is expected to provide a vmemmap_populate() function
 * to instantiate the mapping.
 */
#include <linux/mm.h>
#include <linux/mmzone.h>
#include <linux/bootmem.h>
#include <linux/highmem.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/vmalloc.h>
#include <linux/sched.h>
#include <asm/dma.h>
#include <asm/pgalloc.h>
#include <asm/pgtable.h>

/*
 * Allocate a block of memory to be used to back the virtual memory map
 * or to back the page tables that are used to create the mapping.
 * Uses the main allocators if they are available, else bootmem.
 */

static void * __init_refok __earlyonly_bootmem_alloc(int node,
				unsigned long size,
				unsigned long align,
				unsigned long goal)
{
	return memblock_virt_alloc_try_nid(size, align, goal,
					    BOOTMEM_ALLOC_ACCESSIBLE, node);
}

static void *vmemmap_buf;
static void *vmemmap_buf_end;

void * __meminit vmemmap_alloc_block(unsigned long size, int node)
{
	/* If the main allocator is up use that, fallback to bootmem. */
	if (slab_is_available()) {
		struct page *page;

		if (node_state(node, N_HIGH_MEMORY))
			page = alloc_pages_node(
				node, GFP_KERNEL | __GFP_ZERO | __GFP_REPEAT,
				get_order(size));
		else
			page = alloc_pages(
				GFP_KERNEL | __GFP_ZERO | __GFP_REPEAT,
				get_order(size));
		if (page)
			return page_address(page);
		return NULL;
	} else
		return __earlyonly_bootmem_alloc(node, size, size,
				__pa(MAX_DMA_ADDRESS));
}

/* need to make sure size is all the same during early stage */
void * __meminit vmemmap_alloc_block_buf(unsigned long size, int node)
{
	void *ptr;

	if (!vmemmap_buf)
		return vmemmap_alloc_block(size, node);

	/* take the from buf */
	ptr = (void *)ALIGN((unsigned long)vmemmap_buf, size);
	if (ptr + size > vmemmap_buf_end)
		return vmemmap_alloc_block(size, node);

	vmemmap_buf = ptr + size;

	return ptr;
}

void __meminit vmemmap_verify(pte_t *pte, int node,
				unsigned long start, unsigned long end)
{
	unsigned long pfn = pte_pfn(*pte);
	int actual_node = early_pfn_to_nid(pfn);

	if (node_distance(actual_node, node) > LOCAL_DISTANCE)
		printk(KERN_WARNING "[%lx-%lx] potential offnode page_structs\n",
		       start, end - 1);
}

pte_t * __meminit vmemmap_pte_populate(pmd_t *pmd, unsigned long addr, int node)
{
	pte_t *pte = pte_offset_kernel(pmd, addr);
	if (pte_none(*pte)) {
		pte_t entry;
		void *p = vmemmap_alloc_block_buf(PAGE_SIZE, node);
		if (!p)
			return NULL;
		entry = pfn_pte(__pa(p) >> PAGE_SHIFT, PAGE_KERNEL);
		set_pte_at(&init_mm, addr, pte, entry);
	}
	return pte;
}

pmd_t * __meminit vmemmap_pmd_populate(pud_t *pud, unsigned long addr, int node)
{
	pmd_t *pmd = pmd_offset(pud, addr);
	if (pmd_none(*pmd)) {
		void *p = vmemmap_alloc_block(PAGE_SIZE, node);
		if (!p)
			return NULL;
		pmd_populate_kernel(&init_mm, pmd, p);
	}
	return pmd;
}

pud_t * __meminit vmemmap_pud_populate(pgd_t *pgd, unsigned long addr, int node)
{
#ifdef CONFIG_RKP
	void *p = NULL ;
#endif 	
	pud_t *pud = pud_offset(pgd, addr);
	if (pud_none(*pud)) {
#ifdef CONFIG_RKP
		p =  rkp_ro_alloc();
#else /* !CONFIG_RKP */
		void *p = vmemmap_alloc_block(PAGE_SIZE, node);
#endif
		if (!p)
			return NULL;
		pud_populate(&init_mm, pud, p);
	}
	return pud;
}

pgd_t * __meminit vmemmap_pgd_populate(unsigned long addr, int node)
{
	pgd_t *pgd = pgd_offset_k(addr);
	if (pgd_none(*pgd)) {
		void *p = vmemmap_alloc_block(PAGE_SIZE, node);
		if (!p)
			return NULL;
		pgd_populate(&init_mm, pgd, p);
	}
	return pgd;
}

int __meminit vmemmap_populate_basepages(unsigned long start,
					 unsigned long end, int node)
{
	unsigned long addr = start;
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *pte;

	for (; addr < end; addr += PAGE_SIZE) {
		pgd = vmemmap_pgd_populate(addr, node);
		if (!pgd)
			return -ENOMEM;
		pud = vmemmap_pud_populate(pgd, addr, node);
		if (!pud)
			return -ENOMEM;
		pmd = vmemmap_pmd_populate(pud, addr, node);
		if (!pmd)
			return -ENOMEM;
		pte = vmemmap_pte_populate(pmd, addr, node);
		if (!pte)
			return -ENOMEM;
		vmemmap_verify(pte, node, addr, addr + PAGE_SIZE);
	}

	return 0;
}

struct page * __meminit sparse_mem_map_populate(unsigned long pnum, int nid)
{
	unsigned long start;
	unsigned long end;
	struct page *map;

	map = pfn_to_page(pnum * PAGES_PER_SECTION);
	start = (unsigned long)map;
	end = (unsigned long)(map + PAGES_PER_SECTION);

	if (vmemmap_populate(start, end, nid))
		return NULL;

	return map;
}

void __init sparse_mem_maps_populate_node(struct page **map_map,
					  unsigned long pnum_begin,
					  unsigned long pnum_end,
					  unsigned long map_count, int nodeid)
{
	unsigned long pnum;
	unsigned long size = sizeof(struct page) * PAGES_PER_SECTION;
	void *vmemmap_buf_start;

	size = ALIGN(size, PMD_SIZE);
	vmemmap_buf_start = __earlyonly_bootmem_alloc(nodeid, size * map_count,
			 PMD_SIZE, __pa(MAX_DMA_ADDRESS));

	if (vmemmap_buf_start) {
		vmemmap_buf = vmemmap_buf_start;
		vmemmap_buf_end = vmemmap_buf_start + size * map_count;
	}

	for (pnum = pnum_begin; pnum < pnum_end; pnum++) {
		struct mem_section *ms;

		if (!present_section_nr(pnum))
			continue;

		map_map[pnum] = sparse_mem_map_populate(pnum, nodeid);
		if (map_map[pnum])
			continue;
		ms = __nr_to_section(pnum);
		printk(KERN_ERR "%s: sparsemem memory map backing failed some memory will not be available.\n",
		       __func__);
		ms->section_mem_map = 0;
	}

	if (vmemmap_buf_start) {
		/* need to free left buf */
		memblock_free_early(__pa(vmemmap_buf),
				    vmemmap_buf_end - vmemmap_buf);
		vmemmap_buf = NULL;
		vmemmap_buf_end = NULL;
	}
}
Commit	Line	Data
	1	/*
	2	* Virtual Memory Map support
	3	*
	4	* (C) 2007 sgi. Christoph Lameter.
	5	*
	6	* Virtual memory maps allow VM primitives pfn_to_page, page_to_pfn,
	7	* virt_to_page, page_address() to be implemented as a base offset
	8	* calculation without memory access.
	9	*
	10	* However, virtual mappings need a page table and TLBs. Many Linux
	11	* architectures already map their physical space using 1-1 mappings
	12	* via TLBs. For those arches the virtual memory map is essentially
	13	* for free if we use the same page size as the 1-1 mappings. In that
	14	* case the overhead consists of a few additional pages that are
	15	* allocated to create a view of memory for vmemmap.
	16	*
	17	* The architecture is expected to provide a vmemmap_populate() function
	18	* to instantiate the mapping.
	19	*/
	20	#include <linux/mm.h>
	21	#include <linux/mmzone.h>
	22	#include <linux/bootmem.h>
	23	#include <linux/highmem.h>
	24	#include <linux/slab.h>
	25	#include <linux/spinlock.h>
	26	#include <linux/vmalloc.h>
	27	#include <linux/sched.h>
	28	#include <asm/dma.h>
	29	#include <asm/pgalloc.h>
	30	#include <asm/pgtable.h>
	31
	32	/*
	33	* Allocate a block of memory to be used to back the virtual memory map
	34	* or to back the page tables that are used to create the mapping.
	35	* Uses the main allocators if they are available, else bootmem.
	36	*/
	37
	38	static void * __init_refok __earlyonly_bootmem_alloc(int node,
	39	unsigned long size,
	40	unsigned long align,
	41	unsigned long goal)
	42	{
	43	return memblock_virt_alloc_try_nid(size, align, goal,
	44	BOOTMEM_ALLOC_ACCESSIBLE, node);
	45	}
	46
	47	static void *vmemmap_buf;
	48	static void *vmemmap_buf_end;
	49
	50	void * __meminit vmemmap_alloc_block(unsigned long size, int node)
	51	{
	52	/* If the main allocator is up use that, fallback to bootmem. */
	53	if (slab_is_available()) {
	54	struct page *page;
	55
	56	if (node_state(node, N_HIGH_MEMORY))
	57	page = alloc_pages_node(
	58	node, GFP_KERNEL \| __GFP_ZERO \| __GFP_REPEAT,
	59	get_order(size));
	60	else
	61	page = alloc_pages(
	62	GFP_KERNEL \| __GFP_ZERO \| __GFP_REPEAT,
	63	get_order(size));
	64	if (page)
	65	return page_address(page);
	66	return NULL;
	67	} else
	68	return __earlyonly_bootmem_alloc(node, size, size,
	69	__pa(MAX_DMA_ADDRESS));
	70	}
	71
	72	/* need to make sure size is all the same during early stage */
	73	void * __meminit vmemmap_alloc_block_buf(unsigned long size, int node)
	74	{
	75	void *ptr;
	76
	77	if (!vmemmap_buf)
	78	return vmemmap_alloc_block(size, node);
	79
	80	/* take the from buf */
	81	ptr = (void *)ALIGN((unsigned long)vmemmap_buf, size);
	82	if (ptr + size > vmemmap_buf_end)
	83	return vmemmap_alloc_block(size, node);
	84
	85	vmemmap_buf = ptr + size;
	86
	87	return ptr;
	88	}
	89
	90	void __meminit vmemmap_verify(pte_t *pte, int node,
	91	unsigned long start, unsigned long end)
	92	{
	93	unsigned long pfn = pte_pfn(*pte);
	94	int actual_node = early_pfn_to_nid(pfn);
	95
	96	if (node_distance(actual_node, node) > LOCAL_DISTANCE)
	97	printk(KERN_WARNING "[%lx-%lx] potential offnode page_structs\n",
	98	start, end - 1);
	99	}
	100
	101	pte_t * __meminit vmemmap_pte_populate(pmd_t *pmd, unsigned long addr, int node)
	102	{
	103	pte_t *pte = pte_offset_kernel(pmd, addr);
	104	if (pte_none(*pte)) {
	105	pte_t entry;
	106	void *p = vmemmap_alloc_block_buf(PAGE_SIZE, node);
	107	if (!p)
	108	return NULL;
	109	entry = pfn_pte(__pa(p) >> PAGE_SHIFT, PAGE_KERNEL);
	110	set_pte_at(&init_mm, addr, pte, entry);
	111	}
	112	return pte;
	113	}
	114
	115	pmd_t * __meminit vmemmap_pmd_populate(pud_t *pud, unsigned long addr, int node)
	116	{
	117	pmd_t *pmd = pmd_offset(pud, addr);
	118	if (pmd_none(*pmd)) {
	119	void *p = vmemmap_alloc_block(PAGE_SIZE, node);
	120	if (!p)
	121	return NULL;
	122	pmd_populate_kernel(&init_mm, pmd, p);
	123	}
	124	return pmd;
	125	}
	126
	127	pud_t * __meminit vmemmap_pud_populate(pgd_t *pgd, unsigned long addr, int node)
	128	{
	129	#ifdef CONFIG_RKP
	130	void *p = NULL ;
	131	#endif
	132	pud_t *pud = pud_offset(pgd, addr);
	133	if (pud_none(*pud)) {
	134	#ifdef CONFIG_RKP
	135	p = rkp_ro_alloc();
	136	#else /* !CONFIG_RKP */
	137	void *p = vmemmap_alloc_block(PAGE_SIZE, node);
	138	#endif
	139	if (!p)
	140	return NULL;
	141	pud_populate(&init_mm, pud, p);
	142	}
	143	return pud;
	144	}
	145
	146	pgd_t * __meminit vmemmap_pgd_populate(unsigned long addr, int node)
	147	{
	148	pgd_t *pgd = pgd_offset_k(addr);
	149	if (pgd_none(*pgd)) {
	150	void *p = vmemmap_alloc_block(PAGE_SIZE, node);
	151	if (!p)
	152	return NULL;
	153	pgd_populate(&init_mm, pgd, p);
	154	}
	155	return pgd;
	156	}
	157
	158	int __meminit vmemmap_populate_basepages(unsigned long start,
	159	unsigned long end, int node)
	160	{
	161	unsigned long addr = start;
	162	pgd_t *pgd;
	163	pud_t *pud;
	164	pmd_t *pmd;
	165	pte_t *pte;
	166
	167	for (; addr < end; addr += PAGE_SIZE) {
	168	pgd = vmemmap_pgd_populate(addr, node);
	169	if (!pgd)
	170	return -ENOMEM;
	171	pud = vmemmap_pud_populate(pgd, addr, node);
	172	if (!pud)
	173	return -ENOMEM;
	174	pmd = vmemmap_pmd_populate(pud, addr, node);
	175	if (!pmd)
	176	return -ENOMEM;
	177	pte = vmemmap_pte_populate(pmd, addr, node);
	178	if (!pte)
	179	return -ENOMEM;
	180	vmemmap_verify(pte, node, addr, addr + PAGE_SIZE);
	181	}
	182
	183	return 0;
	184	}
	185
	186	struct page * __meminit sparse_mem_map_populate(unsigned long pnum, int nid)
	187	{
	188	unsigned long start;
	189	unsigned long end;
	190	struct page *map;
	191
	192	map = pfn_to_page(pnum * PAGES_PER_SECTION);
	193	start = (unsigned long)map;
	194	end = (unsigned long)(map + PAGES_PER_SECTION);
	195
	196	if (vmemmap_populate(start, end, nid))
	197	return NULL;
	198
	199	return map;
	200	}
	201
	202	void __init sparse_mem_maps_populate_node(struct page **map_map,
	203	unsigned long pnum_begin,
	204	unsigned long pnum_end,
	205	unsigned long map_count, int nodeid)
	206	{
	207	unsigned long pnum;
	208	unsigned long size = sizeof(struct page) * PAGES_PER_SECTION;
	209	void *vmemmap_buf_start;
	210
	211	size = ALIGN(size, PMD_SIZE);
	212	vmemmap_buf_start = __earlyonly_bootmem_alloc(nodeid, size * map_count,
	213	PMD_SIZE, __pa(MAX_DMA_ADDRESS));
	214
	215	if (vmemmap_buf_start) {
	216	vmemmap_buf = vmemmap_buf_start;
	217	vmemmap_buf_end = vmemmap_buf_start + size * map_count;
	218	}
	219
	220	for (pnum = pnum_begin; pnum < pnum_end; pnum++) {
	221	struct mem_section *ms;
	222
	223	if (!present_section_nr(pnum))
	224	continue;
	225
	226	map_map[pnum] = sparse_mem_map_populate(pnum, nodeid);
	227	if (map_map[pnum])
	228	continue;
	229	ms = __nr_to_section(pnum);
	230	printk(KERN_ERR "%s: sparsemem memory map backing failed some memory will not be available.\n",
	231	__func__);
	232	ms->section_mem_map = 0;
	233	}
	234
	235	if (vmemmap_buf_start) {
	236	/* need to free left buf */
	237	memblock_free_early(__pa(vmemmap_buf),
	238	vmemmap_buf_end - vmemmap_buf);
	239	vmemmap_buf = NULL;
	240	vmemmap_buf_end = NULL;
	241	}
	242	}