init.c 19.0 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
/*
 *  linux/arch/x86_64/mm/init.c
 *
 *  Copyright (C) 1995  Linus Torvalds
 *  Copyright (C) 2000  Pavel Machek <pavel@suse.cz>
 *  Copyright (C) 2002,2003 Andi Kleen <ak@suse.de>
 */

#include <linux/signal.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/types.h>
#include <linux/ptrace.h>
#include <linux/mman.h>
#include <linux/mm.h>
#include <linux/swap.h>
#include <linux/smp.h>
#include <linux/init.h>
#include <linux/pagemap.h>
#include <linux/bootmem.h>
#include <linux/proc_fs.h>
24
#include <linux/pci.h>
25
#include <linux/poison.h>
26
#include <linux/dma-mapping.h>
27 28
#include <linux/module.h>
#include <linux/memory_hotplug.h>
L
Linus Torvalds 已提交
29 30 31 32 33 34 35 36 37 38 39 40 41 42

#include <asm/processor.h>
#include <asm/system.h>
#include <asm/uaccess.h>
#include <asm/pgtable.h>
#include <asm/pgalloc.h>
#include <asm/dma.h>
#include <asm/fixmap.h>
#include <asm/e820.h>
#include <asm/apic.h>
#include <asm/tlb.h>
#include <asm/mmu_context.h>
#include <asm/proto.h>
#include <asm/smp.h>
43
#include <asm/sections.h>
L
Linus Torvalds 已提交
44 45 46 47 48

#ifndef Dprintk
#define Dprintk(x...)
#endif

49 50 51
struct dma_mapping_ops* dma_ops;
EXPORT_SYMBOL(dma_ops);

52 53
static unsigned long dma_reserve __initdata;

L
Linus Torvalds 已提交
54 55 56 57 58 59 60 61 62 63
DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);

/*
 * NOTE: pagetable_init alloc all the fixmap pagetables contiguous on the
 * physical space so we can cache the place of the first one and move
 * around without checking the pgd every time.
 */

void show_mem(void)
{
64 65
	long i, total = 0, reserved = 0;
	long shared = 0, cached = 0;
L
Linus Torvalds 已提交
66 67 68
	pg_data_t *pgdat;
	struct page *page;

69
	printk(KERN_INFO "Mem-info:\n");
L
Linus Torvalds 已提交
70
	show_free_areas();
71
	printk(KERN_INFO "Free swap:       %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
L
Linus Torvalds 已提交
72

73
	for_each_online_pgdat(pgdat) {
L
Linus Torvalds 已提交
74 75 76
               for (i = 0; i < pgdat->node_spanned_pages; ++i) {
			page = pfn_to_page(pgdat->node_start_pfn + i);
			total++;
77 78 79 80 81 82
			if (PageReserved(page))
				reserved++;
			else if (PageSwapCache(page))
				cached++;
			else if (page_count(page))
				shared += page_count(page) - 1;
L
Linus Torvalds 已提交
83 84
               }
	}
85 86 87 88
	printk(KERN_INFO "%lu pages of RAM\n", total);
	printk(KERN_INFO "%lu reserved pages\n",reserved);
	printk(KERN_INFO "%lu pages shared\n",shared);
	printk(KERN_INFO "%lu pages swap cached\n",cached);
L
Linus Torvalds 已提交
89 90 91 92
}

int after_bootmem;

93
static __init void *spp_getpage(void)
L
Linus Torvalds 已提交
94 95 96 97 98 99 100 101 102 103 104 105 106
{ 
	void *ptr;
	if (after_bootmem)
		ptr = (void *) get_zeroed_page(GFP_ATOMIC); 
	else
		ptr = alloc_bootmem_pages(PAGE_SIZE);
	if (!ptr || ((unsigned long)ptr & ~PAGE_MASK))
		panic("set_pte_phys: cannot allocate page data %s\n", after_bootmem?"after bootmem":"");

	Dprintk("spp_getpage %p\n", ptr);
	return ptr;
} 

107
static __init void set_pte_phys(unsigned long vaddr,
L
Linus Torvalds 已提交
108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155
			 unsigned long phys, pgprot_t prot)
{
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *pte, new_pte;

	Dprintk("set_pte_phys %lx to %lx\n", vaddr, phys);

	pgd = pgd_offset_k(vaddr);
	if (pgd_none(*pgd)) {
		printk("PGD FIXMAP MISSING, it should be setup in head.S!\n");
		return;
	}
	pud = pud_offset(pgd, vaddr);
	if (pud_none(*pud)) {
		pmd = (pmd_t *) spp_getpage(); 
		set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE | _PAGE_USER));
		if (pmd != pmd_offset(pud, 0)) {
			printk("PAGETABLE BUG #01! %p <-> %p\n", pmd, pmd_offset(pud,0));
			return;
		}
	}
	pmd = pmd_offset(pud, vaddr);
	if (pmd_none(*pmd)) {
		pte = (pte_t *) spp_getpage();
		set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE | _PAGE_USER));
		if (pte != pte_offset_kernel(pmd, 0)) {
			printk("PAGETABLE BUG #02!\n");
			return;
		}
	}
	new_pte = pfn_pte(phys >> PAGE_SHIFT, prot);

	pte = pte_offset_kernel(pmd, vaddr);
	if (!pte_none(*pte) &&
	    pte_val(*pte) != (pte_val(new_pte) & __supported_pte_mask))
		pte_ERROR(*pte);
	set_pte(pte, new_pte);

	/*
	 * It's enough to flush this one mapping.
	 * (PGE mappings get flushed as well)
	 */
	__flush_tlb_one(vaddr);
}

/* NOTE: this is meant to be run only at boot */
156 157
void __init 
__set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t prot)
L
Linus Torvalds 已提交
158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181
{
	unsigned long address = __fix_to_virt(idx);

	if (idx >= __end_of_fixed_addresses) {
		printk("Invalid __set_fixmap\n");
		return;
	}
	set_pte_phys(address, phys, prot);
}

unsigned long __initdata table_start, table_end; 

extern pmd_t temp_boot_pmds[]; 

static  struct temp_map { 
	pmd_t *pmd;
	void  *address; 
	int    allocated; 
} temp_mappings[] __initdata = { 
	{ &temp_boot_pmds[0], (void *)(40UL * 1024 * 1024) },
	{ &temp_boot_pmds[1], (void *)(42UL * 1024 * 1024) }, 
	{}
}; 

182
static __meminit void *alloc_low_page(int *index, unsigned long *phys)
L
Linus Torvalds 已提交
183 184 185 186 187 188
{ 
	struct temp_map *ti;
	int i; 
	unsigned long pfn = table_end++, paddr; 
	void *adr;

189 190 191 192 193 194
	if (after_bootmem) {
		adr = (void *)get_zeroed_page(GFP_ATOMIC);
		*phys = __pa(adr);
		return adr;
	}

L
Linus Torvalds 已提交
195 196 197 198 199 200 201 202 203 204 205 206
	if (pfn >= end_pfn) 
		panic("alloc_low_page: ran out of memory"); 
	for (i = 0; temp_mappings[i].allocated; i++) {
		if (!temp_mappings[i].pmd) 
			panic("alloc_low_page: ran out of temp mappings"); 
	} 
	ti = &temp_mappings[i];
	paddr = (pfn << PAGE_SHIFT) & PMD_MASK; 
	set_pmd(ti->pmd, __pmd(paddr | _KERNPG_TABLE | _PAGE_PSE)); 
	ti->allocated = 1; 
	__flush_tlb(); 	       
	adr = ti->address + ((pfn << PAGE_SHIFT) & ~PMD_MASK); 
207
	memset(adr, 0, PAGE_SIZE);
L
Linus Torvalds 已提交
208 209 210 211 212
	*index = i; 
	*phys  = pfn * PAGE_SIZE;  
	return adr; 
} 

213
static __meminit void unmap_low_page(int i)
L
Linus Torvalds 已提交
214
{ 
215 216 217 218 219 220
	struct temp_map *ti;

	if (after_bootmem)
		return;

	ti = &temp_mappings[i];
L
Linus Torvalds 已提交
221 222 223 224
	set_pmd(ti->pmd, __pmd(0));
	ti->allocated = 0; 
} 

225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250
/* Must run before zap_low_mappings */
__init void *early_ioremap(unsigned long addr, unsigned long size)
{
	unsigned long map = round_down(addr, LARGE_PAGE_SIZE); 

	/* actually usually some more */
	if (size >= LARGE_PAGE_SIZE) { 
		return NULL;
	}
	set_pmd(temp_mappings[0].pmd,  __pmd(map | _KERNPG_TABLE | _PAGE_PSE));
	map += LARGE_PAGE_SIZE;
	set_pmd(temp_mappings[1].pmd,  __pmd(map | _KERNPG_TABLE | _PAGE_PSE));
	__flush_tlb();
	return temp_mappings[0].address + (addr & (LARGE_PAGE_SIZE-1));
}

/* To avoid virtual aliases later */
__init void early_iounmap(void *addr, unsigned long size)
{
	if ((void *)round_down((unsigned long)addr, LARGE_PAGE_SIZE) != temp_mappings[0].address)
		printk("early_iounmap: bad address %p\n", addr);
	set_pmd(temp_mappings[0].pmd, __pmd(0));
	set_pmd(temp_mappings[1].pmd, __pmd(0));
	__flush_tlb();
}

251
static void __meminit
252
phys_pmd_init(pmd_t *pmd_page, unsigned long address, unsigned long end)
253
{
254
	int i = pmd_index(address);
255

256
	for (; i < PTRS_PER_PMD; i++, address += PMD_SIZE) {
257
		unsigned long entry;
258
		pmd_t *pmd = pmd_page + pmd_index(address);
259

260 261 262 263
		if (address >= end) {
			if (!after_bootmem)
				for (; i < PTRS_PER_PMD; i++, pmd++)
					set_pmd(pmd, __pmd(0));
264 265
			break;
		}
266 267 268 269

		if (pmd_val(*pmd))
			continue;

270 271 272 273 274 275 276 277 278
		entry = _PAGE_NX|_PAGE_PSE|_KERNPG_TABLE|_PAGE_GLOBAL|address;
		entry &= __supported_pte_mask;
		set_pmd(pmd, __pmd(entry));
	}
}

static void __meminit
phys_pmd_update(pud_t *pud, unsigned long address, unsigned long end)
{
279 280 281 282 283
	pmd_t *pmd = pmd_offset(pud,0);
	spin_lock(&init_mm.page_table_lock);
	phys_pmd_init(pmd, address, end);
	spin_unlock(&init_mm.page_table_lock);
	__flush_tlb_all();
284 285
}

286
static void __meminit phys_pud_init(pud_t *pud_page, unsigned long addr, unsigned long end)
L
Linus Torvalds 已提交
287
{ 
288
	int i = pud_index(addr);
289 290


291
	for (; i < PTRS_PER_PUD; i++, addr = (addr & PUD_MASK) + PUD_SIZE ) {
L
Linus Torvalds 已提交
292
		int map; 
293 294
		unsigned long pmd_phys;
		pud_t *pud = pud_page + pud_index(addr);
L
Linus Torvalds 已提交
295 296
		pmd_t *pmd;

297
		if (addr >= end)
L
Linus Torvalds 已提交
298 299
			break;

300
		if (!after_bootmem && !e820_any_mapped(addr,addr+PUD_SIZE,0)) {
L
Linus Torvalds 已提交
301 302 303 304
			set_pud(pud, __pud(0)); 
			continue;
		} 

305 306 307 308 309
		if (pud_val(*pud)) {
			phys_pmd_update(pud, addr, end);
			continue;
		}

L
Linus Torvalds 已提交
310
		pmd = alloc_low_page(&map, &pmd_phys);
311
		spin_lock(&init_mm.page_table_lock);
L
Linus Torvalds 已提交
312
		set_pud(pud, __pud(pmd_phys | _KERNPG_TABLE));
313
		phys_pmd_init(pmd, addr, end);
314
		spin_unlock(&init_mm.page_table_lock);
L
Linus Torvalds 已提交
315 316 317 318 319 320 321
		unmap_low_page(map);
	}
	__flush_tlb();
} 

static void __init find_early_table_space(unsigned long end)
{
322
	unsigned long puds, pmds, tables, start;
L
Linus Torvalds 已提交
323 324 325 326 327 328

	puds = (end + PUD_SIZE - 1) >> PUD_SHIFT;
	pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT;
	tables = round_up(puds * sizeof(pud_t), PAGE_SIZE) +
		 round_up(pmds * sizeof(pmd_t), PAGE_SIZE);

329 330 331 332 333
 	/* RED-PEN putting page tables only on node 0 could
 	   cause a hotspot and fill up ZONE_DMA. The page tables
 	   need roughly 0.5KB per GB. */
 	start = 0x8000;
 	table_start = find_e820_area(start, end, tables);
L
Linus Torvalds 已提交
334 335 336 337 338
	if (table_start == -1UL)
		panic("Cannot find space for the kernel page tables");

	table_start >>= PAGE_SHIFT;
	table_end = table_start;
339 340

	early_printk("kernel direct mapping tables up to %lx @ %lx-%lx\n",
341 342
		end, table_start << PAGE_SHIFT,
		(table_start << PAGE_SHIFT) + tables);
L
Linus Torvalds 已提交
343 344 345 346 347
}

/* Setup the direct mapping of the physical memory at PAGE_OFFSET.
   This runs before bootmem is initialized and gets pages directly from the 
   physical memory. To access them they are temporarily mapped. */
348
void __meminit init_memory_mapping(unsigned long start, unsigned long end)
L
Linus Torvalds 已提交
349 350 351 352 353 354 355 356 357 358 359
{ 
	unsigned long next; 

	Dprintk("init_memory_mapping\n");

	/* 
	 * Find space for the kernel direct mapping tables.
	 * Later we should allocate these tables in the local node of the memory
	 * mapped.  Unfortunately this is done currently before the nodes are 
	 * discovered.
	 */
360 361
	if (!after_bootmem)
		find_early_table_space(end);
L
Linus Torvalds 已提交
362 363 364 365 366 367 368

	start = (unsigned long)__va(start);
	end = (unsigned long)__va(end);

	for (; start < end; start = next) {
		int map;
		unsigned long pud_phys; 
369 370 371 372
		pgd_t *pgd = pgd_offset_k(start);
		pud_t *pud;

		if (after_bootmem)
373
			pud = pud_offset(pgd, start & PGDIR_MASK);
374 375 376
		else
			pud = alloc_low_page(&map, &pud_phys);

L
Linus Torvalds 已提交
377 378 379 380
		next = start + PGDIR_SIZE;
		if (next > end) 
			next = end; 
		phys_pud_init(pud, __pa(start), __pa(next));
381 382
		if (!after_bootmem)
			set_pgd(pgd_offset_k(start), mk_kernel_pgd(pud_phys));
L
Linus Torvalds 已提交
383 384 385
		unmap_low_page(map);   
	} 

386 387
	if (!after_bootmem)
		asm volatile("movq %%cr4,%0" : "=r" (mmu_cr4_features));
L
Linus Torvalds 已提交
388 389 390
	__flush_tlb_all();
}

391
void __cpuinit zap_low_mappings(int cpu)
L
Linus Torvalds 已提交
392
{
393 394 395 396 397 398 399 400 401 402 403
	if (cpu == 0) {
		pgd_t *pgd = pgd_offset_k(0UL);
		pgd_clear(pgd);
	} else {
		/*
		 * For AP's, zap the low identity mappings by changing the cr3
		 * to init_level4_pgt and doing local flush tlb all
		 */
		asm volatile("movq %0,%%cr3" :: "r" (__pa_symbol(&init_level4_pgt)));
	}
	__flush_tlb_all();
L
Linus Torvalds 已提交
404 405
}

406
#ifndef CONFIG_NUMA
L
Linus Torvalds 已提交
407 408
void __init paging_init(void)
{
409 410 411 412 413 414
	unsigned long max_zone_pfns[MAX_NR_ZONES];
	memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
	max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN;
	max_zone_pfns[ZONE_DMA32] = MAX_DMA32_PFN;
	max_zone_pfns[ZONE_NORMAL] = end_pfn;

415 416
	memory_present(0, 0, end_pfn);
	sparse_init();
417
	free_area_init_nodes(max_zone_pfns);
L
Linus Torvalds 已提交
418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454
}
#endif

/* Unmap a kernel mapping if it exists. This is useful to avoid prefetches
   from the CPU leading to inconsistent cache lines. address and size
   must be aligned to 2MB boundaries. 
   Does nothing when the mapping doesn't exist. */
void __init clear_kernel_mapping(unsigned long address, unsigned long size) 
{
	unsigned long end = address + size;

	BUG_ON(address & ~LARGE_PAGE_MASK);
	BUG_ON(size & ~LARGE_PAGE_MASK); 
	
	for (; address < end; address += LARGE_PAGE_SIZE) { 
		pgd_t *pgd = pgd_offset_k(address);
		pud_t *pud;
		pmd_t *pmd;
		if (pgd_none(*pgd))
			continue;
		pud = pud_offset(pgd, address);
		if (pud_none(*pud))
			continue; 
		pmd = pmd_offset(pud, address);
		if (!pmd || pmd_none(*pmd))
			continue; 
		if (0 == (pmd_val(*pmd) & _PAGE_PSE)) { 
			/* Could handle this, but it should not happen currently. */
			printk(KERN_ERR 
	       "clear_kernel_mapping: mapping has been split. will leak memory\n"); 
			pmd_ERROR(*pmd); 
		}
		set_pmd(pmd, __pmd(0)); 		
	}
	__flush_tlb_all();
} 

455 456 457 458 459 460
/*
 * Memory hotplug specific functions
 */
void online_page(struct page *page)
{
	ClearPageReserved(page);
461
	init_page_count(page);
462 463 464 465 466
	__free_page(page);
	totalram_pages++;
	num_physpages++;
}

467
#ifdef CONFIG_MEMORY_HOTPLUG
468 469 470 471
/*
 * Memory is added always to NORMAL zone. This means you will never get
 * additional DMA/DMA32 memory.
 */
472
int arch_add_memory(int nid, u64 start, u64 size)
473
{
474
	struct pglist_data *pgdat = NODE_DATA(nid);
475
	struct zone *zone = pgdat->node_zones + ZONE_NORMAL;
476 477 478 479
	unsigned long start_pfn = start >> PAGE_SHIFT;
	unsigned long nr_pages = size >> PAGE_SHIFT;
	int ret;

480 481
	init_memory_mapping(start, (start + size -1));

482 483 484 485 486 487 488 489 490
	ret = __add_pages(zone, start_pfn, nr_pages);
	if (ret)
		goto error;

	return ret;
error:
	printk("%s: Problem encountered in __add_pages!\n", __func__);
	return ret;
}
491
EXPORT_SYMBOL_GPL(arch_add_memory);
492 493 494 495 496 497 498

int remove_memory(u64 start, u64 size)
{
	return -EINVAL;
}
EXPORT_SYMBOL_GPL(remove_memory);

499
#if !defined(CONFIG_ACPI_NUMA) && defined(CONFIG_NUMA)
500 501 502 503
int memory_add_physaddr_to_nid(u64 start)
{
	return 0;
}
504
EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid);
505 506
#endif

507 508 509
#endif /* CONFIG_MEMORY_HOTPLUG */

#ifdef CONFIG_MEMORY_HOTPLUG_RESERVE
510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534
/*
 * Memory Hotadd without sparsemem. The mem_maps have been allocated in advance,
 * just online the pages.
 */
int __add_pages(struct zone *z, unsigned long start_pfn, unsigned long nr_pages)
{
	int err = -EIO;
	unsigned long pfn;
	unsigned long total = 0, mem = 0;
	for (pfn = start_pfn; pfn < start_pfn + nr_pages; pfn++) {
		if (pfn_valid(pfn)) {
			online_page(pfn_to_page(pfn));
			err = 0;
			mem++;
		}
		total++;
	}
	if (!err) {
		z->spanned_pages += total;
		z->present_pages += mem;
		z->zone_pgdat->node_spanned_pages += total;
		z->zone_pgdat->node_present_pages += mem;
	}
	return err;
}
535
#endif
536

L
Linus Torvalds 已提交
537 538 539 540 541
static struct kcore_list kcore_mem, kcore_vmalloc, kcore_kernel, kcore_modules,
			 kcore_vsyscall;

void __init mem_init(void)
{
542
	long codesize, reservedpages, datasize, initsize;
L
Linus Torvalds 已提交
543

544
	pci_iommu_alloc();
L
Linus Torvalds 已提交
545 546 547 548 549 550 551

	/* clear the zero-page */
	memset(empty_zero_page, 0, PAGE_SIZE);

	reservedpages = 0;

	/* this will put all low memory onto the freelists */
552
#ifdef CONFIG_NUMA
553
	totalram_pages = numa_free_all_bootmem();
L
Linus Torvalds 已提交
554
#else
555
	totalram_pages = free_all_bootmem();
L
Linus Torvalds 已提交
556
#endif
557 558
	reservedpages = end_pfn - totalram_pages -
					absent_pages_in_range(0, end_pfn);
L
Linus Torvalds 已提交
559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574

	after_bootmem = 1;

	codesize =  (unsigned long) &_etext - (unsigned long) &_text;
	datasize =  (unsigned long) &_edata - (unsigned long) &_etext;
	initsize =  (unsigned long) &__init_end - (unsigned long) &__init_begin;

	/* Register memory areas for /proc/kcore */
	kclist_add(&kcore_mem, __va(0), max_low_pfn << PAGE_SHIFT); 
	kclist_add(&kcore_vmalloc, (void *)VMALLOC_START, 
		   VMALLOC_END-VMALLOC_START);
	kclist_add(&kcore_kernel, &_stext, _end - _stext);
	kclist_add(&kcore_modules, (void *)MODULES_VADDR, MODULES_LEN);
	kclist_add(&kcore_vsyscall, (void *)VSYSCALL_START, 
				 VSYSCALL_END - VSYSCALL_START);

575
	printk("Memory: %luk/%luk available (%ldk kernel code, %ldk reserved, %ldk data, %ldk init)\n",
L
Linus Torvalds 已提交
576 577 578 579 580 581 582
		(unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
		end_pfn << (PAGE_SHIFT-10),
		codesize >> 10,
		reservedpages << (PAGE_SHIFT-10),
		datasize >> 10,
		initsize >> 10);

583
#ifdef CONFIG_SMP
L
Linus Torvalds 已提交
584
	/*
585 586 587
	 * Sync boot_level4_pgt mappings with the init_level4_pgt
	 * except for the low identity mappings which are already zapped
	 * in init_level4_pgt. This sync-up is essential for AP's bringup
L
Linus Torvalds 已提交
588
	 */
589
	memcpy(boot_level4_pgt+1, init_level4_pgt+1, (PTRS_PER_PGD-1)*sizeof(pgd_t));
L
Linus Torvalds 已提交
590 591 592
#endif
}

593
void free_init_pages(char *what, unsigned long begin, unsigned long end)
L
Linus Torvalds 已提交
594 595 596
{
	unsigned long addr;

597 598 599 600 601
	if (begin >= end)
		return;

	printk(KERN_INFO "Freeing %s: %ldk freed\n", what, (end - begin) >> 10);
	for (addr = begin; addr < end; addr += PAGE_SIZE) {
L
Linus Torvalds 已提交
602
		ClearPageReserved(virt_to_page(addr));
603
		init_page_count(virt_to_page(addr));
604 605
		memset((void *)(addr & ~(PAGE_SIZE-1)),
			POISON_FREE_INITMEM, PAGE_SIZE);
L
Linus Torvalds 已提交
606 607 608
		free_page(addr);
		totalram_pages++;
	}
609 610 611 612
}

void free_initmem(void)
{
613 614
	memset(__initdata_begin, POISON_FREE_INITDATA,
		__initdata_end - __initdata_begin);
615 616 617
	free_init_pages("unused kernel memory",
			(unsigned long)(&__init_begin),
			(unsigned long)(&__init_end));
L
Linus Torvalds 已提交
618 619
}

620 621 622 623
#ifdef CONFIG_DEBUG_RODATA

void mark_rodata_ro(void)
{
624
	unsigned long addr = (unsigned long)__start_rodata;
625

626
	for (; addr < (unsigned long)__end_rodata; addr += PAGE_SIZE)
627 628 629
		change_page_attr_addr(addr, 1, PAGE_KERNEL_RO);

	printk ("Write protecting the kernel read-only data: %luk\n",
630
			(__end_rodata - __start_rodata) >> 10);
631 632 633 634 635 636 637 638 639 640 641

	/*
	 * change_page_attr_addr() requires a global_flush_tlb() call after it.
	 * We do this after the printk so that if something went wrong in the
	 * change, the printk gets out at least to give a better debug hint
	 * of who is the culprit.
	 */
	global_flush_tlb();
}
#endif

L
Linus Torvalds 已提交
642 643 644
#ifdef CONFIG_BLK_DEV_INITRD
void free_initrd_mem(unsigned long start, unsigned long end)
{
645
	free_init_pages("initrd memory", start, end);
L
Linus Torvalds 已提交
646 647 648 649 650
}
#endif

void __init reserve_bootmem_generic(unsigned long phys, unsigned len) 
{ 
651
#ifdef CONFIG_NUMA
L
Linus Torvalds 已提交
652
	int nid = phys_to_nid(phys);
653 654 655 656 657 658 659 660 661 662 663 664 665 666
#endif
	unsigned long pfn = phys >> PAGE_SHIFT;
	if (pfn >= end_pfn) {
		/* This can happen with kdump kernels when accessing firmware
		   tables. */
		if (pfn < end_pfn_map)
			return;
		printk(KERN_ERR "reserve_bootmem: illegal reserve %lx %u\n",
				phys, len);
		return;
	}

	/* Should check here against the e820 map to avoid double free */
#ifdef CONFIG_NUMA
L
Linus Torvalds 已提交
667 668 669 670
  	reserve_bootmem_node(NODE_DATA(nid), phys, len);
#else       		
	reserve_bootmem(phys, len);    
#endif
671
	if (phys+len <= MAX_DMA_PFN*PAGE_SIZE) {
672
		dma_reserve += len / PAGE_SIZE;
673 674
		set_dma_reserve(dma_reserve);
	}
L
Linus Torvalds 已提交
675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732
}

int kern_addr_valid(unsigned long addr) 
{ 
	unsigned long above = ((long)addr) >> __VIRTUAL_MASK_SHIFT;
       pgd_t *pgd;
       pud_t *pud;
       pmd_t *pmd;
       pte_t *pte;

	if (above != 0 && above != -1UL)
		return 0; 
	
	pgd = pgd_offset_k(addr);
	if (pgd_none(*pgd))
		return 0;

	pud = pud_offset(pgd, addr);
	if (pud_none(*pud))
		return 0; 

	pmd = pmd_offset(pud, addr);
	if (pmd_none(*pmd))
		return 0;
	if (pmd_large(*pmd))
		return pfn_valid(pmd_pfn(*pmd));

	pte = pte_offset_kernel(pmd, addr);
	if (pte_none(*pte))
		return 0;
	return pfn_valid(pte_pfn(*pte));
}

#ifdef CONFIG_SYSCTL
#include <linux/sysctl.h>

extern int exception_trace, page_fault_trace;

static ctl_table debug_table2[] = {
	{ 99, "exception-trace", &exception_trace, sizeof(int), 0644, NULL,
	  proc_dointvec },
	{ 0, }
}; 

static ctl_table debug_root_table2[] = { 
	{ .ctl_name = CTL_DEBUG, .procname = "debug", .mode = 0555, 
	   .child = debug_table2 }, 
	{ 0 }, 
}; 

static __init int x8664_sysctl_init(void)
{ 
	register_sysctl_table(debug_root_table2, 1);
	return 0;
}
__initcall(x8664_sysctl_init);
#endif

733
/* A pseudo VMA to allow ptrace access for the vsyscall page.  This only
734 735
   covers the 64bit vsyscall page now. 32bit has a real VMA now and does
   not need special handling anymore. */
L
Linus Torvalds 已提交
736 737 738

static struct vm_area_struct gate_vma = {
	.vm_start = VSYSCALL_START,
739 740 741
	.vm_end = VSYSCALL_START + (VSYSCALL_MAPPED_PAGES << PAGE_SHIFT),
	.vm_page_prot = PAGE_READONLY_EXEC,
	.vm_flags = VM_READ | VM_EXEC
L
Linus Torvalds 已提交
742 743 744 745 746
};

struct vm_area_struct *get_gate_vma(struct task_struct *tsk)
{
#ifdef CONFIG_IA32_EMULATION
747 748
	if (test_tsk_thread_flag(tsk, TIF_IA32))
		return NULL;
L
Linus Torvalds 已提交
749 750 751 752 753 754 755
#endif
	return &gate_vma;
}

int in_gate_area(struct task_struct *task, unsigned long addr)
{
	struct vm_area_struct *vma = get_gate_vma(task);
756 757
	if (!vma)
		return 0;
L
Linus Torvalds 已提交
758 759 760 761 762 763 764 765 766
	return (addr >= vma->vm_start) && (addr < vma->vm_end);
}

/* Use this when you have no reliable task/vma, typically from interrupt
 * context.  It is less reliable than using the task's vma and may give
 * false positives.
 */
int in_gate_area_no_task(unsigned long addr)
{
767
	return (addr >= VSYSCALL_START) && (addr < VSYSCALL_END);
L
Linus Torvalds 已提交
768
}