init_32.c 32.4 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
/*
 *
 *  Copyright (C) 1995  Linus Torvalds
 *
 *  Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
 */

#include <linux/module.h>
#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/hugetlb.h>
#include <linux/swap.h>
#include <linux/smp.h>
#include <linux/init.h>
#include <linux/highmem.h>
#include <linux/pagemap.h>
24
#include <linux/pci.h>
25
#include <linux/pfn.h>
26
#include <linux/poison.h>
L
Linus Torvalds 已提交
27 28 29
#include <linux/bootmem.h>
#include <linux/slab.h>
#include <linux/proc_fs.h>
30
#include <linux/memory_hotplug.h>
31
#include <linux/initrd.h>
32
#include <linux/cpumask.h>
L
Linus Torvalds 已提交
33

34
#include <asm/asm.h>
35
#include <asm/bios_ebda.h>
L
Linus Torvalds 已提交
36 37 38 39 40 41 42 43
#include <asm/processor.h>
#include <asm/system.h>
#include <asm/uaccess.h>
#include <asm/pgtable.h>
#include <asm/dma.h>
#include <asm/fixmap.h>
#include <asm/e820.h>
#include <asm/apic.h>
I
Ingo Molnar 已提交
44
#include <asm/bugs.h>
L
Linus Torvalds 已提交
45 46
#include <asm/tlb.h>
#include <asm/tlbflush.h>
47
#include <asm/pgalloc.h>
L
Linus Torvalds 已提交
48
#include <asm/sections.h>
49
#include <asm/paravirt.h>
50
#include <asm/setup.h>
51
#include <asm/cacheflush.h>
L
Linus Torvalds 已提交
52

53
unsigned long max_low_pfn_mapped;
54
unsigned long max_pfn_mapped;
55

L
Linus Torvalds 已提交
56 57 58
DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
unsigned long highstart_pfn, highend_pfn;

I
Ingo Molnar 已提交
59
static noinline int do_test_wp_bit(void);
L
Linus Torvalds 已提交
60

61 62 63 64 65 66 67

static unsigned long __initdata table_start;
static unsigned long __meminitdata table_end;
static unsigned long __meminitdata table_top;

static int __initdata after_init_bootmem;

68
static __init void *alloc_low_page(void)
69 70 71 72 73 74 75 76 77 78 79 80
{
	unsigned long pfn = table_end++;
	void *adr;

	if (pfn >= table_top)
		panic("alloc_low_page: ran out of memory");

	adr = __va(pfn * PAGE_SIZE);
	memset(adr, 0, PAGE_SIZE);
	return adr;
}

L
Linus Torvalds 已提交
81 82 83 84 85 86 87 88 89
/*
 * Creates a middle page table and puts a pointer to it in the
 * given global directory entry. This only returns the gd entry
 * in non-PAE compilation mode, since the middle layer is folded.
 */
static pmd_t * __init one_md_table_init(pgd_t *pgd)
{
	pud_t *pud;
	pmd_t *pmd_table;
I
Ingo Molnar 已提交
90

L
Linus Torvalds 已提交
91
#ifdef CONFIG_X86_PAE
92
	if (!(pgd_val(*pgd) & _PAGE_PRESENT)) {
93 94 95
		if (after_init_bootmem)
			pmd_table = (pmd_t *)alloc_bootmem_low_pages(PAGE_SIZE);
		else
96
			pmd_table = (pmd_t *)alloc_low_page();
97
		paravirt_alloc_pmd(&init_mm, __pa(pmd_table) >> PAGE_SHIFT);
98 99
		set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT));
		pud = pud_offset(pgd, 0);
I
Ingo Molnar 已提交
100
		BUG_ON(pmd_table != pmd_offset(pud, 0));
101 102

		return pmd_table;
103 104
	}
#endif
L
Linus Torvalds 已提交
105 106
	pud = pud_offset(pgd, 0);
	pmd_table = pmd_offset(pud, 0);
I
Ingo Molnar 已提交
107

L
Linus Torvalds 已提交
108 109 110 111 112
	return pmd_table;
}

/*
 * Create a page table and place a pointer to it in a middle page
I
Ingo Molnar 已提交
113
 * directory entry:
L
Linus Torvalds 已提交
114 115 116
 */
static pte_t * __init one_page_table_init(pmd_t *pmd)
{
117
	if (!(pmd_val(*pmd) & _PAGE_PRESENT)) {
118 119
		pte_t *page_table = NULL;

120
		if (after_init_bootmem) {
121
#ifdef CONFIG_DEBUG_PAGEALLOC
122
			page_table = (pte_t *) alloc_bootmem_pages(PAGE_SIZE);
123
#endif
124 125
			if (!page_table)
				page_table =
126
				(pte_t *)alloc_bootmem_low_pages(PAGE_SIZE);
127 128
		} else
			page_table = (pte_t *)alloc_low_page();
129

130
		paravirt_alloc_pte(&init_mm, __pa(page_table) >> PAGE_SHIFT);
L
Linus Torvalds 已提交
131
		set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_TABLE));
132
		BUG_ON(page_table != pte_offset_kernel(pmd, 0));
L
Linus Torvalds 已提交
133
	}
134

L
Linus Torvalds 已提交
135 136 137
	return pte_offset_kernel(pmd, 0);
}

138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178
static pte_t *__init page_table_kmap_check(pte_t *pte, pmd_t *pmd,
					   unsigned long vaddr, pte_t *lastpte)
{
#ifdef CONFIG_HIGHMEM
	/*
	 * Something (early fixmap) may already have put a pte
	 * page here, which causes the page table allocation
	 * to become nonlinear. Attempt to fix it, and if it
	 * is still nonlinear then we have to bug.
	 */
	int pmd_idx_kmap_begin = fix_to_virt(FIX_KMAP_END) >> PMD_SHIFT;
	int pmd_idx_kmap_end = fix_to_virt(FIX_KMAP_BEGIN) >> PMD_SHIFT;

	if (pmd_idx_kmap_begin != pmd_idx_kmap_end
	    && (vaddr >> PMD_SHIFT) >= pmd_idx_kmap_begin
	    && (vaddr >> PMD_SHIFT) <= pmd_idx_kmap_end
	    && ((__pa(pte) >> PAGE_SHIFT) < table_start
		|| (__pa(pte) >> PAGE_SHIFT) >= table_end)) {
		pte_t *newpte;
		int i;

		BUG_ON(after_init_bootmem);
		newpte = alloc_low_page();
		for (i = 0; i < PTRS_PER_PTE; i++)
			set_pte(newpte + i, pte[i]);

		paravirt_alloc_pte(&init_mm, __pa(newpte) >> PAGE_SHIFT);
		set_pmd(pmd, __pmd(__pa(newpte)|_PAGE_TABLE));
		BUG_ON(newpte != pte_offset_kernel(pmd, 0));
		__flush_tlb_all();

		paravirt_release_pte(__pa(pte) >> PAGE_SHIFT);
		pte = newpte;
	}
	BUG_ON(vaddr < fix_to_virt(FIX_KMAP_BEGIN - 1)
	       && vaddr > fix_to_virt(FIX_KMAP_END)
	       && lastpte && lastpte + PTRS_PER_PTE != pte);
#endif
	return pte;
}

L
Linus Torvalds 已提交
179
/*
I
Ingo Molnar 已提交
180
 * This function initializes a certain range of kernel virtual memory
L
Linus Torvalds 已提交
181 182
 * with new bootmem page tables, everywhere page tables are missing in
 * the given range.
I
Ingo Molnar 已提交
183 184 185
 *
 * NOTE: The pagetables are allocated contiguous on the physical space
 * so we can cache the place of the first one and move around without
L
Linus Torvalds 已提交
186 187
 * checking the pgd every time.
 */
I
Ingo Molnar 已提交
188 189
static void __init
page_table_range_init(unsigned long start, unsigned long end, pgd_t *pgd_base)
L
Linus Torvalds 已提交
190 191 192
{
	int pgd_idx, pmd_idx;
	unsigned long vaddr;
I
Ingo Molnar 已提交
193 194
	pgd_t *pgd;
	pmd_t *pmd;
195
	pte_t *pte = NULL;
L
Linus Torvalds 已提交
196 197 198 199 200 201 202

	vaddr = start;
	pgd_idx = pgd_index(vaddr);
	pmd_idx = pmd_index(vaddr);
	pgd = pgd_base + pgd_idx;

	for ( ; (pgd_idx < PTRS_PER_PGD) && (vaddr != end); pgd++, pgd_idx++) {
203 204
		pmd = one_md_table_init(pgd);
		pmd = pmd + pmd_index(vaddr);
I
Ingo Molnar 已提交
205 206
		for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end);
							pmd++, pmd_idx++) {
207 208
			pte = page_table_kmap_check(one_page_table_init(pmd),
			                            pmd, vaddr, pte);
L
Linus Torvalds 已提交
209 210 211 212 213 214 215 216 217 218 219 220 221 222 223

			vaddr += PMD_SIZE;
		}
		pmd_idx = 0;
	}
}

static inline int is_kernel_text(unsigned long addr)
{
	if (addr >= PAGE_OFFSET && addr <= (unsigned long)__init_end)
		return 1;
	return 0;
}

/*
I
Ingo Molnar 已提交
224 225 226
 * This maps the physical memory to kernel virtual address space, a total
 * of max_low_pfn pages, by creating page tables starting from address
 * PAGE_OFFSET:
L
Linus Torvalds 已提交
227
 */
228
static void __init kernel_physical_mapping_init(pgd_t *pgd_base,
229 230 231
						unsigned long start_pfn,
						unsigned long end_pfn,
						int use_pse)
L
Linus Torvalds 已提交
232
{
I
Ingo Molnar 已提交
233
	int pgd_idx, pmd_idx, pte_ofs;
L
Linus Torvalds 已提交
234 235 236 237
	unsigned long pfn;
	pgd_t *pgd;
	pmd_t *pmd;
	pte_t *pte;
238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255
	unsigned pages_2m, pages_4k;
	int mapping_iter;

	/*
	 * First iteration will setup identity mapping using large/small pages
	 * based on use_pse, with other attributes same as set by
	 * the early code in head_32.S
	 *
	 * Second iteration will setup the appropriate attributes (NX, GLOBAL..)
	 * as desired for the kernel identity mapping.
	 *
	 * This two pass mechanism conforms to the TLB app note which says:
	 *
	 *     "Software should not write to a paging-structure entry in a way
	 *      that would change, for any linear address, both the page size
	 *      and either the page frame or attributes."
	 */
	mapping_iter = 1;
L
Linus Torvalds 已提交
256

257 258
	if (!cpu_has_pse)
		use_pse = 0;
L
Linus Torvalds 已提交
259

260 261
repeat:
	pages_2m = pages_4k = 0;
262 263 264
	pfn = start_pfn;
	pgd_idx = pgd_index((pfn<<PAGE_SHIFT) + PAGE_OFFSET);
	pgd = pgd_base + pgd_idx;
L
Linus Torvalds 已提交
265 266
	for (; pgd_idx < PTRS_PER_PGD; pgd++, pgd_idx++) {
		pmd = one_md_table_init(pgd);
I
Ingo Molnar 已提交
267

268 269 270 271 272 273 274 275 276
		if (pfn >= end_pfn)
			continue;
#ifdef CONFIG_X86_PAE
		pmd_idx = pmd_index((pfn<<PAGE_SHIFT) + PAGE_OFFSET);
		pmd += pmd_idx;
#else
		pmd_idx = 0;
#endif
		for (; pmd_idx < PTRS_PER_PMD && pfn < end_pfn;
J
Jeremy Fitzhardinge 已提交
277
		     pmd++, pmd_idx++) {
I
Ingo Molnar 已提交
278
			unsigned int addr = pfn * PAGE_SIZE + PAGE_OFFSET;
L
Linus Torvalds 已提交
279

I
Ingo Molnar 已提交
280 281 282 283
			/*
			 * Map with big pages if possible, otherwise
			 * create normal page tables:
			 */
284
			if (use_pse) {
I
Ingo Molnar 已提交
285
				unsigned int addr2;
J
Jeremy Fitzhardinge 已提交
286
				pgprot_t prot = PAGE_KERNEL_LARGE;
287 288 289 290 291 292 293
				/*
				 * first pass will use the same initial
				 * identity mapping attribute + _PAGE_PSE.
				 */
				pgprot_t init_prot =
					__pgprot(PTE_IDENT_ATTR |
						 _PAGE_PSE);
J
Jeremy Fitzhardinge 已提交
294

I
Ingo Molnar 已提交
295
				addr2 = (pfn + PTRS_PER_PTE-1) * PAGE_SIZE +
J
Jeremy Fitzhardinge 已提交
296 297
					PAGE_OFFSET + PAGE_SIZE-1;

I
Ingo Molnar 已提交
298 299
				if (is_kernel_text(addr) ||
				    is_kernel_text(addr2))
J
Jeremy Fitzhardinge 已提交
300 301
					prot = PAGE_KERNEL_LARGE_EXEC;

302
				pages_2m++;
303 304 305 306
				if (mapping_iter == 1)
					set_pmd(pmd, pfn_pmd(pfn, init_prot));
				else
					set_pmd(pmd, pfn_pmd(pfn, prot));
307

L
Linus Torvalds 已提交
308
				pfn += PTRS_PER_PTE;
I
Ingo Molnar 已提交
309 310 311
				continue;
			}
			pte = one_page_table_init(pmd);
L
Linus Torvalds 已提交
312

313 314 315
			pte_ofs = pte_index((pfn<<PAGE_SHIFT) + PAGE_OFFSET);
			pte += pte_ofs;
			for (; pte_ofs < PTRS_PER_PTE && pfn < end_pfn;
I
Ingo Molnar 已提交
316 317
			     pte++, pfn++, pte_ofs++, addr += PAGE_SIZE) {
				pgprot_t prot = PAGE_KERNEL;
318 319 320 321 322
				/*
				 * first pass will use the same initial
				 * identity mapping attribute.
				 */
				pgprot_t init_prot = __pgprot(PTE_IDENT_ATTR);
J
Jeremy Fitzhardinge 已提交
323

I
Ingo Molnar 已提交
324 325
				if (is_kernel_text(addr))
					prot = PAGE_KERNEL_EXEC;
J
Jeremy Fitzhardinge 已提交
326

327
				pages_4k++;
328 329 330 331
				if (mapping_iter == 1)
					set_pte(pte, pfn_pte(pfn, init_prot));
				else
					set_pte(pte, pfn_pte(pfn, prot));
L
Linus Torvalds 已提交
332 333 334
			}
		}
	}
335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354
	if (mapping_iter == 1) {
		/*
		 * update direct mapping page count only in the first
		 * iteration.
		 */
		update_page_count(PG_LEVEL_2M, pages_2m);
		update_page_count(PG_LEVEL_4K, pages_4k);

		/*
		 * local global flush tlb, which will flush the previous
		 * mappings present in both small and large page TLB's.
		 */
		__flush_tlb_all();

		/*
		 * Second iteration will set the actual desired PTE attributes.
		 */
		mapping_iter = 2;
		goto repeat;
	}
L
Linus Torvalds 已提交
355 356 357 358 359
}

pte_t *kmap_pte;
pgprot_t kmap_prot;

I
Ingo Molnar 已提交
360 361 362 363 364
static inline pte_t *kmap_get_fixmap_pte(unsigned long vaddr)
{
	return pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(vaddr),
			vaddr), vaddr), vaddr);
}
L
Linus Torvalds 已提交
365 366 367 368 369

static void __init kmap_init(void)
{
	unsigned long kmap_vstart;

I
Ingo Molnar 已提交
370 371 372
	/*
	 * Cache the first kmap pte:
	 */
L
Linus Torvalds 已提交
373 374 375 376 377 378
	kmap_vstart = __fix_to_virt(FIX_KMAP_BEGIN);
	kmap_pte = kmap_get_fixmap_pte(kmap_vstart);

	kmap_prot = PAGE_KERNEL;
}

379
#ifdef CONFIG_HIGHMEM
L
Linus Torvalds 已提交
380 381
static void __init permanent_kmaps_init(pgd_t *pgd_base)
{
I
Ingo Molnar 已提交
382
	unsigned long vaddr;
L
Linus Torvalds 已提交
383 384 385 386 387 388 389 390 391 392 393 394
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *pte;

	vaddr = PKMAP_BASE;
	page_table_range_init(vaddr, vaddr + PAGE_SIZE*LAST_PKMAP, pgd_base);

	pgd = swapper_pg_dir + pgd_index(vaddr);
	pud = pud_offset(pgd, vaddr);
	pmd = pmd_offset(pud, vaddr);
	pte = pte_offset_kernel(pmd, vaddr);
I
Ingo Molnar 已提交
395
	pkmap_page_table = pte;
L
Linus Torvalds 已提交
396 397
}

Y
Yinghai Lu 已提交
398
static void __init add_one_highpage_init(struct page *page, int pfn)
L
Linus Torvalds 已提交
399
{
Y
Yinghai Lu 已提交
400 401 402 403
	ClearPageReserved(page);
	init_page_count(page);
	__free_page(page);
	totalhigh_pages++;
L
Linus Torvalds 已提交
404 405
}

406 407 408 409 410
struct add_highpages_data {
	unsigned long start_pfn;
	unsigned long end_pfn;
};

411
static int __init add_highpages_work_fn(unsigned long start_pfn,
412
					 unsigned long end_pfn, void *datax)
L
Linus Torvalds 已提交
413
{
414 415 416 417
	int node_pfn;
	struct page *page;
	unsigned long final_start_pfn, final_end_pfn;
	struct add_highpages_data *data;
I
Ingo Molnar 已提交
418

419 420 421 422 423
	data = (struct add_highpages_data *)datax;

	final_start_pfn = max(start_pfn, data->start_pfn);
	final_end_pfn = min(end_pfn, data->end_pfn);
	if (final_start_pfn >= final_end_pfn)
424
		return 0;
425 426 427 428 429 430

	for (node_pfn = final_start_pfn; node_pfn < final_end_pfn;
	     node_pfn++) {
		if (!pfn_valid(node_pfn))
			continue;
		page = pfn_to_page(node_pfn);
Y
Yinghai Lu 已提交
431
		add_one_highpage_init(page, node_pfn);
432
	}
433

434 435
	return 0;

436 437 438
}

void __init add_highpages_with_active_regions(int nid, unsigned long start_pfn,
Y
Yinghai Lu 已提交
439
					      unsigned long end_pfn)
440 441 442 443 444 445 446 447 448
{
	struct add_highpages_data data;

	data.start_pfn = start_pfn;
	data.end_pfn = end_pfn;

	work_with_active_regions(nid, add_highpages_work_fn, &data);
}

L
Linus Torvalds 已提交
449
#else
450 451 452
static inline void permanent_kmaps_init(pgd_t *pgd_base)
{
}
L
Linus Torvalds 已提交
453 454
#endif /* CONFIG_HIGHMEM */

455
void __init native_pagetable_setup_start(pgd_t *base)
L
Linus Torvalds 已提交
456
{
457 458 459 460 461
	unsigned long pfn, va;
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *pte;
462 463

	/*
464 465
	 * Remove any mappings which extend past the end of physical
	 * memory from the boot time page table:
466
	 */
467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483
	for (pfn = max_low_pfn + 1; pfn < 1<<(32-PAGE_SHIFT); pfn++) {
		va = PAGE_OFFSET + (pfn<<PAGE_SHIFT);
		pgd = base + pgd_index(va);
		if (!pgd_present(*pgd))
			break;

		pud = pud_offset(pgd, va);
		pmd = pmd_offset(pud, va);
		if (!pmd_present(*pmd))
			break;

		pte = pte_offset_kernel(pmd, va);
		if (!pte_present(*pte))
			break;

		pte_clear(NULL, va, pte);
	}
484
	paravirt_alloc_pmd(&init_mm, __pa(base) >> PAGE_SHIFT);
485 486 487 488 489 490 491 492 493 494 495 496
}

void __init native_pagetable_setup_done(pgd_t *base)
{
}

/*
 * Build a proper pagetable for the kernel mappings.  Up until this
 * point, we've been running on some set of pagetables constructed by
 * the boot process.
 *
 * If we're booting on native hardware, this will be a pagetable
497 498
 * constructed in arch/x86/kernel/head_32.S.  The root of the
 * pagetable will be swapper_pg_dir.
499 500 501 502 503 504 505 506 507 508 509
 *
 * If we're booting paravirtualized under a hypervisor, then there are
 * more options: we may already be running PAE, and the pagetable may
 * or may not be based in swapper_pg_dir.  In any case,
 * paravirt_pagetable_setup_start() will set up swapper_pg_dir
 * appropriately for the rest of the initialization to work.
 *
 * In general, pagetable_init() assumes that the pagetable may already
 * be partially populated, and so it avoids stomping on any existing
 * mappings.
 */
510
static void __init early_ioremap_page_table_range_init(pgd_t *pgd_base)
511
{
I
Ingo Molnar 已提交
512
	unsigned long vaddr, end;
513

L
Linus Torvalds 已提交
514 515 516 517 518
	/*
	 * Fixed mappings, only the page table structure has to be
	 * created - mappings will be set by set_fixmap():
	 */
	vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK;
519 520
	end = (FIXADDR_TOP + PMD_SIZE - 1) & PMD_MASK;
	page_table_range_init(vaddr, end, pgd_base);
521
	early_ioremap_reset();
522 523 524 525 526 527
}

static void __init pagetable_init(void)
{
	pgd_t *pgd_base = swapper_pg_dir;

L
Linus Torvalds 已提交
528 529 530
	permanent_kmaps_init(pgd_base);
}

531
#ifdef CONFIG_ACPI_SLEEP
L
Linus Torvalds 已提交
532
/*
533
 * ACPI suspend needs this for resume, because things like the intel-agp
L
Linus Torvalds 已提交
534 535
 * driver might have split up a kernel 4MB mapping.
 */
536
char swsusp_pg_dir[PAGE_SIZE]
I
Ingo Molnar 已提交
537
	__attribute__ ((aligned(PAGE_SIZE)));
L
Linus Torvalds 已提交
538 539 540 541 542

static inline void save_pg_dir(void)
{
	memcpy(swsusp_pg_dir, swapper_pg_dir, PAGE_SIZE);
}
543
#else /* !CONFIG_ACPI_SLEEP */
L
Linus Torvalds 已提交
544 545 546
static inline void save_pg_dir(void)
{
}
547
#endif /* !CONFIG_ACPI_SLEEP */
L
Linus Torvalds 已提交
548

I
Ingo Molnar 已提交
549
void zap_low_mappings(void)
L
Linus Torvalds 已提交
550 551 552 553 554 555 556 557 558
{
	int i;

	/*
	 * Zap initial low-memory mappings.
	 *
	 * Note that "pgd_clear()" doesn't do it for
	 * us, because pgd_clear() is a no-op on i386.
	 */
J
Jeremy Fitzhardinge 已提交
559
	for (i = 0; i < KERNEL_PGD_BOUNDARY; i++) {
L
Linus Torvalds 已提交
560 561 562 563 564
#ifdef CONFIG_X86_PAE
		set_pgd(swapper_pg_dir+i, __pgd(1 + __pa(empty_zero_page)));
#else
		set_pgd(swapper_pg_dir+i, __pgd(0));
#endif
I
Ingo Molnar 已提交
565
	}
L
Linus Torvalds 已提交
566 567 568
	flush_tlb_all();
}

I
Ingo Molnar 已提交
569
int nx_enabled;
J
Jan Beulich 已提交
570

571
pteval_t __supported_pte_mask __read_mostly = ~(_PAGE_NX | _PAGE_GLOBAL | _PAGE_IOMAP);
572 573
EXPORT_SYMBOL_GPL(__supported_pte_mask);

J
Jan Beulich 已提交
574 575
#ifdef CONFIG_X86_PAE

I
Ingo Molnar 已提交
576
static int disable_nx __initdata;
L
Linus Torvalds 已提交
577 578 579 580 581 582 583 584 585

/*
 * noexec = on|off
 *
 * Control non executable mappings.
 *
 * on      Enable
 * off     Disable
 */
586
static int __init noexec_setup(char *str)
L
Linus Torvalds 已提交
587
{
588 589 590 591 592
	if (!str || !strcmp(str, "on")) {
		if (cpu_has_nx) {
			__supported_pte_mask |= _PAGE_NX;
			disable_nx = 0;
		}
I
Ingo Molnar 已提交
593 594 595 596 597 598 599 600
	} else {
		if (!strcmp(str, "off")) {
			disable_nx = 1;
			__supported_pte_mask &= ~_PAGE_NX;
		} else {
			return -EINVAL;
		}
	}
601 602

	return 0;
L
Linus Torvalds 已提交
603
}
604
early_param("noexec", noexec_setup);
L
Linus Torvalds 已提交
605 606 607 608 609 610 611

static void __init set_nx(void)
{
	unsigned int v[4], l, h;

	if (cpu_has_pae && (cpuid_eax(0x80000000) > 0x80000001)) {
		cpuid(0x80000001, &v[0], &v[1], &v[2], &v[3]);
I
Ingo Molnar 已提交
612

L
Linus Torvalds 已提交
613 614 615 616 617 618 619 620 621 622 623
		if ((v[3] & (1 << 20)) && !disable_nx) {
			rdmsr(MSR_EFER, l, h);
			l |= EFER_NX;
			wrmsr(MSR_EFER, l, h);
			nx_enabled = 1;
			__supported_pte_mask |= _PAGE_NX;
		}
	}
}
#endif

624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641
/* user-defined highmem size */
static unsigned int highmem_pages = -1;

/*
 * highmem=size forces highmem to be exactly 'size' bytes.
 * This works even on boxes that have no highmem otherwise.
 * This also works to reduce highmem size on bigger boxes.
 */
static int __init parse_highmem(char *arg)
{
	if (!arg)
		return -EINVAL;

	highmem_pages = memparse(arg, &arg) >> PAGE_SHIFT;
	return 0;
}
early_param("highmem", parse_highmem);

642 643 644 645 646
#define MSG_HIGHMEM_TOO_BIG \
	"highmem size (%luMB) is bigger than pages available (%luMB)!\n"

#define MSG_LOWMEM_TOO_SMALL \
	"highmem size (%luMB) results in <64MB lowmem, ignoring it!\n"
647
/*
648 649 650
 * All of RAM fits into lowmem - but if user wants highmem
 * artificially via the highmem=x boot parameter then create
 * it:
651
 */
652
void __init lowmem_pfn_init(void)
653
{
Y
Yinghai Lu 已提交
654
	/* max_low_pfn is 0, we already have early_res support */
655
	max_low_pfn = max_pfn;
656

657 658 659 660 661 662 663 664 665 666 667
	if (highmem_pages == -1)
		highmem_pages = 0;
#ifdef CONFIG_HIGHMEM
	if (highmem_pages >= max_pfn) {
		printk(KERN_ERR MSG_HIGHMEM_TOO_BIG,
			pages_to_mb(highmem_pages), pages_to_mb(max_pfn));
		highmem_pages = 0;
	}
	if (highmem_pages) {
		if (max_low_pfn - highmem_pages < 64*1024*1024/PAGE_SIZE) {
			printk(KERN_ERR MSG_LOWMEM_TOO_SMALL,
668 669 670
				pages_to_mb(highmem_pages));
			highmem_pages = 0;
		}
671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689
		max_low_pfn -= highmem_pages;
	}
#else
	if (highmem_pages)
		printk(KERN_ERR "ignoring highmem size on non-highmem kernel!\n");
#endif
}

#define MSG_HIGHMEM_TOO_SMALL \
	"only %luMB highmem pages available, ignoring highmem size of %luMB!\n"

#define MSG_HIGHMEM_TRIMMED \
	"Warning: only 4GB will be used. Use a HIGHMEM64G enabled kernel!\n"
/*
 * We have more RAM than fits into lowmem - we try to put it into
 * highmem, also taking the highmem=x boot parameter into account:
 */
void __init highmem_pfn_init(void)
{
690 691
	max_low_pfn = MAXMEM_PFN;

692 693 694 695 696 697 698 699 700 701 702 703
	if (highmem_pages == -1)
		highmem_pages = max_pfn - MAXMEM_PFN;

	if (highmem_pages + MAXMEM_PFN < max_pfn)
		max_pfn = MAXMEM_PFN + highmem_pages;

	if (highmem_pages + MAXMEM_PFN > max_pfn) {
		printk(KERN_WARNING MSG_HIGHMEM_TOO_SMALL,
			pages_to_mb(max_pfn - MAXMEM_PFN),
			pages_to_mb(highmem_pages));
		highmem_pages = 0;
	}
704
#ifndef CONFIG_HIGHMEM
705 706 707 708 709 710 711
	/* Maximum memory usable is what is directly addressable */
	printk(KERN_WARNING "Warning only %ldMB will be used.\n", MAXMEM>>20);
	if (max_pfn > MAX_NONPAE_PFN)
		printk(KERN_WARNING "Use a HIGHMEM64G enabled kernel.\n");
	else
		printk(KERN_WARNING "Use a HIGHMEM enabled kernel.\n");
	max_pfn = MAXMEM_PFN;
712 713
#else /* !CONFIG_HIGHMEM */
#ifndef CONFIG_HIGHMEM64G
714 715 716 717
	if (max_pfn > MAX_NONPAE_PFN) {
		max_pfn = MAX_NONPAE_PFN;
		printk(KERN_WARNING MSG_HIGHMEM_TRIMMED);
	}
718 719
#endif /* !CONFIG_HIGHMEM64G */
#endif /* !CONFIG_HIGHMEM */
720 721 722 723 724 725 726 727 728
}

/*
 * Determine low and high memory ranges:
 */
void __init find_low_pfn_range(void)
{
	/* it could update max_pfn */

729
	if (max_pfn <= MAXMEM_PFN)
730
		lowmem_pfn_init();
731 732
	else
		highmem_pfn_init();
733 734
}

735
#ifndef CONFIG_NEED_MULTIPLE_NODES
736
void __init initmem_init(unsigned long start_pfn,
737 738 739 740 741 742 743
				  unsigned long end_pfn)
{
#ifdef CONFIG_HIGHMEM
	highstart_pfn = highend_pfn = max_pfn;
	if (max_pfn > max_low_pfn)
		highstart_pfn = max_low_pfn;
	memory_present(0, 0, highend_pfn);
744
	e820_register_active_regions(0, 0, highend_pfn);
745 746 747 748 749 750
	printk(KERN_NOTICE "%ldMB HIGHMEM available.\n",
		pages_to_mb(highend_pfn - highstart_pfn));
	num_physpages = highend_pfn;
	high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1;
#else
	memory_present(0, 0, max_low_pfn);
751
	e820_register_active_regions(0, 0, max_low_pfn);
752 753 754 755 756 757 758 759 760 761 762
	num_physpages = max_low_pfn;
	high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1;
#endif
#ifdef CONFIG_FLATMEM
	max_mapnr = num_physpages;
#endif
	printk(KERN_NOTICE "%ldMB LOWMEM available.\n",
			pages_to_mb(max_low_pfn));

	setup_bootmem_allocator();
}
763
#endif /* !CONFIG_NEED_MULTIPLE_NODES */
764

765
static void __init zone_sizes_init(void)
766 767 768 769 770 771 772 773 774 775 776 777 778
{
	unsigned long max_zone_pfns[MAX_NR_ZONES];
	memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
	max_zone_pfns[ZONE_DMA] =
		virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT;
	max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
#ifdef CONFIG_HIGHMEM
	max_zone_pfns[ZONE_HIGHMEM] = highend_pfn;
#endif

	free_area_init_nodes(max_zone_pfns);
}

779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
#ifdef CONFIG_NEED_MULTIPLE_NODES
static unsigned long __init setup_node_bootmem(int nodeid,
				 unsigned long start_pfn,
				 unsigned long end_pfn,
				 unsigned long bootmap)
{
	unsigned long bootmap_size;

	if (start_pfn > max_low_pfn)
		return bootmap;
	if (end_pfn > max_low_pfn)
		end_pfn = max_low_pfn;

	/* don't touch min_low_pfn */
	bootmap_size = init_bootmem_node(NODE_DATA(nodeid),
					 bootmap >> PAGE_SHIFT,
					 start_pfn, end_pfn);
	printk(KERN_INFO "  node %d low ram: %08lx - %08lx\n",
		nodeid, start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT);
	printk(KERN_INFO "  node %d bootmap %08lx - %08lx\n",
		 nodeid, bootmap, bootmap + bootmap_size);
	free_bootmem_with_active_regions(nodeid, end_pfn);
	early_res_to_bootmem(start_pfn<<PAGE_SHIFT, end_pfn<<PAGE_SHIFT);

	return bootmap + bootmap_size;
}
#endif

807 808
void __init setup_bootmem_allocator(void)
{
809
	int nodeid;
810 811 812 813 814 815 816 817 818 819 820
	unsigned long bootmap_size, bootmap;
	/*
	 * Initialize the boot-time allocator (with low memory only):
	 */
	bootmap_size = bootmem_bootmap_pages(max_low_pfn)<<PAGE_SHIFT;
	bootmap = find_e820_area(min_low_pfn<<PAGE_SHIFT,
				 max_pfn_mapped<<PAGE_SHIFT, bootmap_size,
				 PAGE_SIZE);
	if (bootmap == -1L)
		panic("Cannot find bootmem map of size %ld\n", bootmap_size);
	reserve_early(bootmap, bootmap + bootmap_size, "BOOTMAP");
Y
Yinghai Lu 已提交
821

822 823 824 825
	printk(KERN_INFO "  mapped low ram: 0 - %08lx\n",
		 max_pfn_mapped<<PAGE_SHIFT);
	printk(KERN_INFO "  low ram: %08lx - %08lx\n",
		 min_low_pfn<<PAGE_SHIFT, max_low_pfn<<PAGE_SHIFT);
826 827 828 829 830 831 832 833 834

#ifdef CONFIG_NEED_MULTIPLE_NODES
	for_each_online_node(nodeid)
		bootmap = setup_node_bootmem(nodeid, node_start_pfn[nodeid],
					node_end_pfn[nodeid], bootmap);
#else
	/* don't touch min_low_pfn */
	bootmap_size = init_bootmem_node(NODE_DATA(0), bootmap >> PAGE_SHIFT,
					 min_low_pfn, max_low_pfn);
835 836
	printk(KERN_INFO "  bootmap %08lx - %08lx\n",
		 bootmap, bootmap + bootmap_size);
837
	free_bootmem_with_active_regions(0, max_low_pfn);
838
	early_res_to_bootmem(0, max_low_pfn<<PAGE_SHIFT);
839
#endif
840

841
	after_init_bootmem = 1;
842 843
}

844
static void __init find_early_table_space(unsigned long end, int use_pse)
845
{
846
	unsigned long puds, pmds, ptes, tables, start;
847 848

	puds = (end + PUD_SIZE - 1) >> PUD_SHIFT;
849
	tables = roundup(puds * sizeof(pud_t), PAGE_SIZE);
850 851

	pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT;
852
	tables += roundup(pmds * sizeof(pmd_t), PAGE_SIZE);
853

854
	if (use_pse) {
855
		unsigned long extra;
856 857 858

		extra = end - ((end>>PMD_SHIFT) << PMD_SHIFT);
		extra += PMD_SIZE;
859 860 861 862
		ptes = (extra + PAGE_SIZE - 1) >> PAGE_SHIFT;
	} else
		ptes = (end + PAGE_SIZE - 1) >> PAGE_SHIFT;

863
	tables += roundup(ptes * sizeof(pte_t), PAGE_SIZE);
864

865
	/* for fixmap */
866
	tables += roundup(__end_of_fixed_addresses * sizeof(pte_t), PAGE_SIZE);
867

868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887
	/*
	 * 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 = 0x7000;
	table_start = find_e820_area(start, max_pfn_mapped<<PAGE_SHIFT,
					tables, PAGE_SIZE);
	if (table_start == -1UL)
		panic("Cannot find space for the kernel page tables");

	table_start >>= PAGE_SHIFT;
	table_end = table_start;
	table_top = table_start + (tables>>PAGE_SHIFT);

	printk(KERN_DEBUG "kernel direct mapping tables up to %lx @ %lx-%lx\n",
		end, table_start << PAGE_SHIFT,
		(table_start << PAGE_SHIFT) + tables);
}

888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911
struct map_range {
	unsigned long start;
	unsigned long end;
	unsigned page_size_mask;
};

#define NR_RANGE_MR 3

static int save_mr(struct map_range *mr, int nr_range,
		   unsigned long start_pfn, unsigned long end_pfn,
		   unsigned long page_size_mask)
{
	if (start_pfn < end_pfn) {
		if (nr_range >= NR_RANGE_MR)
			panic("run out of range for init_memory_mapping\n");
		mr[nr_range].start = start_pfn<<PAGE_SHIFT;
		mr[nr_range].end   = end_pfn<<PAGE_SHIFT;
		mr[nr_range].page_size_mask = page_size_mask;
		nr_range++;
	}

	return nr_range;
}

912 913 914 915
unsigned long __init_refok init_memory_mapping(unsigned long start,
						unsigned long end)
{
	pgd_t *pgd_base = swapper_pg_dir;
916
	unsigned long page_size_mask = 0;
917
	unsigned long start_pfn, end_pfn;
918 919 920 921 922 923 924 925
	unsigned long pos;

	struct map_range mr[NR_RANGE_MR];
	int nr_range, i;
	int use_pse;

	printk(KERN_INFO "init_memory_mapping: %08lx-%08lx\n", start, end);

926 927 928 929 930 931
#ifdef CONFIG_DEBUG_PAGEALLOC
	/*
	 * For CONFIG_DEBUG_PAGEALLOC, identity mapping will use small pages.
	 * This will simplify cpa(), which otherwise needs to support splitting
	 * large pages into small in interrupt context, etc.
	 */
932
	use_pse = 0;
933
#else
934
	use_pse = cpu_has_pse;
935
#endif
936 937 938 939 940 941 942 943 944 945 946 947 948 949

#ifdef CONFIG_X86_PAE
	set_nx();
	if (nx_enabled)
		printk(KERN_INFO "NX (Execute Disable) protection: active\n");
#endif

	/* Enable PSE if available */
	if (cpu_has_pse)
		set_in_cr4(X86_CR4_PSE);

	/* Enable PGE if available */
	if (cpu_has_pge) {
		set_in_cr4(X86_CR4_PGE);
950
		__supported_pte_mask |= _PAGE_GLOBAL;
951 952
	}

953 954 955 956 957 958
	memset(mr, 0, sizeof(mr));
	nr_range = 0;

	if (use_pse)
		page_size_mask |= 1 << PG_LEVEL_2M;

959 960 961 962 963 964
	/*
	 * Don't use a large page for the first 2/4MB of memory
	 * because there are often fixed size MTRRs in there
	 * and overlapping MTRRs into large pages can cause
	 * slowdowns.
	 */
965 966 967 968 969 970 971
	/* head could not be big page alignment ? */
	start_pfn = start >> PAGE_SHIFT;
	pos = start_pfn << PAGE_SHIFT;
	if (pos == 0)
		end_pfn = 1<<(PMD_SHIFT - PAGE_SHIFT);
	else
		end_pfn = ((pos + (PMD_SIZE - 1))>>PMD_SHIFT)
972
				 << (PMD_SHIFT - PAGE_SHIFT);
973 974 975 976 977
	if (end_pfn > (end>>PAGE_SHIFT))
		end_pfn = end>>PAGE_SHIFT;
	if (start_pfn < end_pfn) {
		nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0);
		pos = end_pfn << PAGE_SHIFT;
978 979 980
	}

	/* big page range */
981
	start_pfn = ((pos + (PMD_SIZE - 1))>>PMD_SHIFT)
982 983
			 << (PMD_SHIFT - PAGE_SHIFT);
	end_pfn = (end>>PMD_SHIFT) << (PMD_SHIFT - PAGE_SHIFT);
984 985 986 987 988
	if (start_pfn < end_pfn) {
		nr_range = save_mr(mr, nr_range, start_pfn, end_pfn,
				page_size_mask & (1<<PG_LEVEL_2M));
		pos = end_pfn << PAGE_SHIFT;
	}
989 990

	/* tail is not big page alignment ? */
991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007
	start_pfn = pos>>PAGE_SHIFT;
	end_pfn = end>>PAGE_SHIFT;
	if (start_pfn < end_pfn)
		nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0);

	/* try to merge same page size and continuous */
	for (i = 0; nr_range > 1 && i < nr_range - 1; i++) {
		unsigned long old_start;
		if (mr[i].end != mr[i+1].start ||
		    mr[i].page_size_mask != mr[i+1].page_size_mask)
			continue;
		/* move it */
		old_start = mr[i].start;
		memmove(&mr[i], &mr[i+1],
			(nr_range - 1 - i) * sizeof(struct map_range));
		mr[i--].start = old_start;
		nr_range--;
1008
	}
1009

1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027
	for (i = 0; i < nr_range; i++)
		printk(KERN_DEBUG " %08lx - %08lx page %s\n",
			mr[i].start, mr[i].end,
			(mr[i].page_size_mask & (1<<PG_LEVEL_2M)) ?
				  "big page" : "4k");

	/*
	 * Find space for the kernel direct mapping tables.
	 */
	if (!after_init_bootmem)
		find_early_table_space(end, use_pse);

	for (i = 0; i < nr_range; i++)
		kernel_physical_mapping_init(pgd_base,
				mr[i].start >> PAGE_SHIFT,
				mr[i].end >> PAGE_SHIFT,
				mr[i].page_size_mask == (1<<PG_LEVEL_2M));

1028 1029
	early_ioremap_page_table_range_init(pgd_base);

1030 1031 1032 1033 1034 1035 1036 1037
	load_cr3(swapper_pg_dir);

	__flush_tlb_all();

	if (!after_init_bootmem)
		reserve_early(table_start << PAGE_SHIFT,
				 table_end << PAGE_SHIFT, "PGTABLE");

1038 1039 1040
	if (!after_init_bootmem)
		early_memtest(start, end);

1041 1042 1043
	return end >> PAGE_SHIFT;
}

1044

L
Linus Torvalds 已提交
1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058
/*
 * paging_init() sets up the page tables - note that the first 8MB are
 * already mapped by head.S.
 *
 * This routines also unmaps the page at virtual kernel address 0, so
 * that we can trap those pesky NULL-reference errors in the kernel.
 */
void __init paging_init(void)
{
	pagetable_init();

	__flush_tlb_all();

	kmap_init();
1059 1060 1061 1062 1063 1064

	/*
	 * NOTE: at this point the bootmem allocator is fully available.
	 */
	sparse_init();
	zone_sizes_init();
L
Linus Torvalds 已提交
1065 1066 1067 1068
}

/*
 * Test if the WP bit works in supervisor mode. It isn't supported on 386's
D
Dmitri Vorobiev 已提交
1069 1070 1071
 * and also on some strange 486's. All 586+'s are OK. This used to involve
 * black magic jumps to work around some nasty CPU bugs, but fortunately the
 * switch to using exceptions got rid of all that.
L
Linus Torvalds 已提交
1072 1073 1074
 */
static void __init test_wp_bit(void)
{
1075 1076
	printk(KERN_INFO
  "Checking if this processor honours the WP bit even in supervisor mode...");
L
Linus Torvalds 已提交
1077 1078 1079 1080 1081 1082 1083

	/* Any page-aligned address will do, the test is non-destructive */
	__set_fixmap(FIX_WP_TEST, __pa(&swapper_pg_dir), PAGE_READONLY);
	boot_cpu_data.wp_works_ok = do_test_wp_bit();
	clear_fixmap(FIX_WP_TEST);

	if (!boot_cpu_data.wp_works_ok) {
1084
		printk(KERN_CONT "No.\n");
L
Linus Torvalds 已提交
1085
#ifdef CONFIG_X86_WP_WORKS_OK
1086 1087
		panic(
  "This kernel doesn't support CPU's with broken WP. Recompile it for a 386!");
L
Linus Torvalds 已提交
1088 1089
#endif
	} else {
1090
		printk(KERN_CONT "Ok.\n");
L
Linus Torvalds 已提交
1091 1092 1093
	}
}

I
Ingo Molnar 已提交
1094
static struct kcore_list kcore_mem, kcore_vmalloc;
L
Linus Torvalds 已提交
1095 1096 1097 1098

void __init mem_init(void)
{
	int codesize, reservedpages, datasize, initsize;
Y
Yinghai Lu 已提交
1099
	int tmp;
L
Linus Torvalds 已提交
1100

1101 1102
	pci_iommu_alloc();

1103
#ifdef CONFIG_FLATMEM
E
Eric Sesterhenn 已提交
1104
	BUG_ON(!mem_map);
L
Linus Torvalds 已提交
1105 1106 1107 1108 1109 1110 1111
#endif
	/* this will put all low memory onto the freelists */
	totalram_pages += free_all_bootmem();

	reservedpages = 0;
	for (tmp = 0; tmp < max_low_pfn; tmp++)
		/*
I
Ingo Molnar 已提交
1112
		 * Only count reserved RAM pages:
L
Linus Torvalds 已提交
1113 1114 1115 1116
		 */
		if (page_is_ram(tmp) && PageReserved(pfn_to_page(tmp)))
			reservedpages++;

Y
Yinghai Lu 已提交
1117
	set_highmem_pages_init();
L
Linus Torvalds 已提交
1118 1119 1120 1121 1122

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

I
Ingo Molnar 已提交
1123 1124
	kclist_add(&kcore_mem, __va(0), max_low_pfn << PAGE_SHIFT);
	kclist_add(&kcore_vmalloc, (void *)VMALLOC_START,
L
Linus Torvalds 已提交
1125 1126
		   VMALLOC_END-VMALLOC_START);

I
Ingo Molnar 已提交
1127 1128
	printk(KERN_INFO "Memory: %luk/%luk available (%dk kernel code, "
			"%dk reserved, %dk data, %dk init, %ldk highmem)\n",
L
Linus Torvalds 已提交
1129 1130 1131 1132 1133 1134 1135 1136 1137
		(unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
		num_physpages << (PAGE_SHIFT-10),
		codesize >> 10,
		reservedpages << (PAGE_SHIFT-10),
		datasize >> 10,
		initsize >> 10,
		(unsigned long) (totalhigh_pages << (PAGE_SHIFT-10))
	       );

1138
	printk(KERN_INFO "virtual kernel memory layout:\n"
I
Ingo Molnar 已提交
1139
		"    fixmap  : 0x%08lx - 0x%08lx   (%4ld kB)\n"
1140
#ifdef CONFIG_HIGHMEM
I
Ingo Molnar 已提交
1141
		"    pkmap   : 0x%08lx - 0x%08lx   (%4ld kB)\n"
1142
#endif
I
Ingo Molnar 已提交
1143 1144 1145 1146 1147 1148 1149
		"    vmalloc : 0x%08lx - 0x%08lx   (%4ld MB)\n"
		"    lowmem  : 0x%08lx - 0x%08lx   (%4ld MB)\n"
		"      .init : 0x%08lx - 0x%08lx   (%4ld kB)\n"
		"      .data : 0x%08lx - 0x%08lx   (%4ld kB)\n"
		"      .text : 0x%08lx - 0x%08lx   (%4ld kB)\n",
		FIXADDR_START, FIXADDR_TOP,
		(FIXADDR_TOP - FIXADDR_START) >> 10,
1150 1151

#ifdef CONFIG_HIGHMEM
I
Ingo Molnar 已提交
1152 1153
		PKMAP_BASE, PKMAP_BASE+LAST_PKMAP*PAGE_SIZE,
		(LAST_PKMAP*PAGE_SIZE) >> 10,
1154 1155
#endif

I
Ingo Molnar 已提交
1156 1157
		VMALLOC_START, VMALLOC_END,
		(VMALLOC_END - VMALLOC_START) >> 20,
1158

I
Ingo Molnar 已提交
1159 1160
		(unsigned long)__va(0), (unsigned long)high_memory,
		((unsigned long)high_memory - (unsigned long)__va(0)) >> 20,
1161

I
Ingo Molnar 已提交
1162 1163 1164
		(unsigned long)&__init_begin, (unsigned long)&__init_end,
		((unsigned long)&__init_end -
		 (unsigned long)&__init_begin) >> 10,
1165

I
Ingo Molnar 已提交
1166 1167
		(unsigned long)&_etext, (unsigned long)&_edata,
		((unsigned long)&_edata - (unsigned long)&_etext) >> 10,
1168

I
Ingo Molnar 已提交
1169 1170
		(unsigned long)&_text, (unsigned long)&_etext,
		((unsigned long)&_etext - (unsigned long)&_text) >> 10);
1171

1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185
	/*
	 * Check boundaries twice: Some fundamental inconsistencies can
	 * be detected at build time already.
	 */
#define __FIXADDR_TOP (-PAGE_SIZE)
#ifdef CONFIG_HIGHMEM
	BUILD_BUG_ON(PKMAP_BASE + LAST_PKMAP*PAGE_SIZE	> FIXADDR_START);
	BUILD_BUG_ON(VMALLOC_END			> PKMAP_BASE);
#endif
#define high_memory (-128UL << 20)
	BUILD_BUG_ON(VMALLOC_START			>= VMALLOC_END);
#undef high_memory
#undef __FIXADDR_TOP

1186
#ifdef CONFIG_HIGHMEM
I
Ingo Molnar 已提交
1187 1188
	BUG_ON(PKMAP_BASE + LAST_PKMAP*PAGE_SIZE	> FIXADDR_START);
	BUG_ON(VMALLOC_END				> PKMAP_BASE);
1189
#endif
1190
	BUG_ON(VMALLOC_START				>= VMALLOC_END);
I
Ingo Molnar 已提交
1191
	BUG_ON((unsigned long)high_memory		> VMALLOC_START);
1192

L
Linus Torvalds 已提交
1193 1194 1195
	if (boot_cpu_data.wp_works_ok < 0)
		test_wp_bit();

1196
	save_pg_dir();
L
Linus Torvalds 已提交
1197 1198 1199
	zap_low_mappings();
}

1200
#ifdef CONFIG_MEMORY_HOTPLUG
1201
int arch_add_memory(int nid, u64 start, u64 size)
1202
{
1203
	struct pglist_data *pgdata = NODE_DATA(nid);
1204
	struct zone *zone = pgdata->node_zones + ZONE_HIGHMEM;
1205 1206 1207
	unsigned long start_pfn = start >> PAGE_SHIFT;
	unsigned long nr_pages = size >> PAGE_SHIFT;

1208
	return __add_pages(nid, zone, start_pfn, nr_pages);
1209
}
1210
#endif
1211

L
Linus Torvalds 已提交
1212 1213 1214 1215
/*
 * This function cannot be __init, since exceptions don't work in that
 * section.  Put this after the callers, so that it cannot be inlined.
 */
I
Ingo Molnar 已提交
1216
static noinline int do_test_wp_bit(void)
L
Linus Torvalds 已提交
1217 1218 1219 1220 1221
{
	char tmp_reg;
	int flag;

	__asm__ __volatile__(
I
Ingo Molnar 已提交
1222 1223 1224
		"	movb %0, %1	\n"
		"1:	movb %1, %0	\n"
		"	xorl %2, %2	\n"
L
Linus Torvalds 已提交
1225
		"2:			\n"
1226
		_ASM_EXTABLE(1b,2b)
L
Linus Torvalds 已提交
1227 1228 1229 1230 1231
		:"=m" (*(char *)fix_to_virt(FIX_WP_TEST)),
		 "=q" (tmp_reg),
		 "=r" (flag)
		:"2" (1)
		:"memory");
I
Ingo Molnar 已提交
1232

L
Linus Torvalds 已提交
1233 1234 1235
	return flag;
}

1236
#ifdef CONFIG_DEBUG_RODATA
1237 1238
const int rodata_test_data = 0xC3;
EXPORT_SYMBOL_GPL(rodata_test_data);
1239 1240 1241

void mark_rodata_ro(void)
{
1242 1243
	unsigned long start = PFN_ALIGN(_text);
	unsigned long size = PFN_ALIGN(_etext) - start;
1244

1245 1246
#ifndef CONFIG_DYNAMIC_FTRACE
	/* Dynamic tracing modifies the kernel text section */
1247 1248 1249
	set_pages_ro(virt_to_page(start), size >> PAGE_SHIFT);
	printk(KERN_INFO "Write protecting the kernel text: %luk\n",
		size >> 10);
1250 1251

#ifdef CONFIG_CPA_DEBUG
1252 1253 1254
	printk(KERN_INFO "Testing CPA: Reverting %lx-%lx\n",
		start, start+size);
	set_pages_rw(virt_to_page(start), size>>PAGE_SHIFT);
1255

1256 1257
	printk(KERN_INFO "Testing CPA: write protecting again\n");
	set_pages_ro(virt_to_page(start), size>>PAGE_SHIFT);
1258
#endif
1259 1260
#endif /* CONFIG_DYNAMIC_FTRACE */

1261 1262
	start += size;
	size = (unsigned long)__end_rodata - start;
1263
	set_pages_ro(virt_to_page(start), size >> PAGE_SHIFT);
1264 1265
	printk(KERN_INFO "Write protecting the kernel read-only data: %luk\n",
		size >> 10);
1266
	rodata_test();
1267

1268
#ifdef CONFIG_CPA_DEBUG
1269
	printk(KERN_INFO "Testing CPA: undo %lx-%lx\n", start, start + size);
1270
	set_pages_rw(virt_to_page(start), size >> PAGE_SHIFT);
1271

1272
	printk(KERN_INFO "Testing CPA: write protecting again\n");
1273
	set_pages_ro(virt_to_page(start), size >> PAGE_SHIFT);
1274
#endif
1275 1276 1277
}
#endif

1278 1279 1280 1281 1282
int __init reserve_bootmem_generic(unsigned long phys, unsigned long len,
				   int flags)
{
	return reserve_bootmem(phys, len, flags);
}