e820_64.c 21.2 KB
Newer Older
1
/*
L
Linus Torvalds 已提交
2 3
 * Handle the memory map.
 * The functions here do the job until bootmem takes over.
4 5 6 7 8 9
 *
 *  Getting sanitize_e820_map() in sync with i386 version by applying change:
 *  -  Provisions for empty E820 memory regions (reported by certain BIOSes).
 *     Alex Achenbach <xela@slit.de>, December 2002.
 *  Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
 *
L
Linus Torvalds 已提交
10 11 12 13 14 15 16
 */
#include <linux/kernel.h>
#include <linux/types.h>
#include <linux/init.h>
#include <linux/bootmem.h>
#include <linux/ioport.h>
#include <linux/string.h>
17
#include <linux/kexec.h>
18
#include <linux/module.h>
19
#include <linux/mm.h>
20 21
#include <linux/suspend.h>
#include <linux/pfn.h>
22

23
#include <asm/pgtable.h>
L
Linus Torvalds 已提交
24 25 26
#include <asm/page.h>
#include <asm/e820.h>
#include <asm/proto.h>
27
#include <asm/setup.h>
28
#include <asm/sections.h>
29
#include <asm/kdebug.h>
L
Linus Torvalds 已提交
30

31
struct e820map e820;
A
Andi Kleen 已提交
32

33
/*
L
Linus Torvalds 已提交
34 35
 * PFN of last memory page.
 */
36
unsigned long end_pfn;
L
Linus Torvalds 已提交
37

38
/*
L
Linus Torvalds 已提交
39 40 41
 * end_pfn only includes RAM, while end_pfn_map includes all e820 entries.
 * The direct mapping extends to end_pfn_map, so that we can directly access
 * apertures, ACPI and other tables without having to play with fixmaps.
42 43
 */
unsigned long end_pfn_map;
L
Linus Torvalds 已提交
44

45
/*
L
Linus Torvalds 已提交
46 47
 * Last pfn which the user wants to use.
 */
48
static unsigned long __initdata end_user_pfn = MAXMEM>>PAGE_SHIFT;
L
Linus Torvalds 已提交
49

50 51 52 53
/*
 * Early reserved memory areas.
 */
#define MAX_EARLY_RES 20
54

55 56
struct early_res {
	unsigned long start, end;
57
	char name[16];
58 59
};
static struct early_res early_res[MAX_EARLY_RES] __initdata = {
60
	{ 0, PAGE_SIZE, "BIOS data page" },			/* BIOS data page */
61
#ifdef CONFIG_SMP
62
	{ SMP_TRAMPOLINE_BASE, SMP_TRAMPOLINE_BASE + 2*PAGE_SIZE, "SMP_TRAMPOLINE" },
L
Linus Torvalds 已提交
63
#endif
64 65 66
	{}
};

67
void __init reserve_early(unsigned long start, unsigned long end, char *name)
68 69 70 71 72 73
{
	int i;
	struct early_res *r;
	for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
		r = &early_res[i];
		if (end > r->start && start < r->end)
74 75
			panic("Overlapping early reservations %lx-%lx %s to %lx-%lx %s\n",
			      start, end - 1, name?name:"", r->start, r->end - 1, r->name);
L
Linus Torvalds 已提交
76
	}
77 78 79 80 81
	if (i >= MAX_EARLY_RES)
		panic("Too many early reservations");
	r = &early_res[i];
	r->start = start;
	r->end = end;
82 83
	if (name)
		strncpy(r->name, name, sizeof(r->name) - 1);
84
}
85

86 87 88 89 90
void __init early_res_to_bootmem(void)
{
	int i;
	for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
		struct early_res *r = &early_res[i];
91 92
		printk(KERN_INFO "early res: %d [%lx-%lx] %s\n", i,
			r->start, r->end - 1, r->name);
93
		reserve_bootmem_generic(r->start, r->end - r->start);
94
	}
95
}
96

97 98 99 100 101 102 103 104 105 106 107 108 109 110 111
/* Check for already reserved areas */
static inline int bad_addr(unsigned long *addrp, unsigned long size)
{
	int i;
	unsigned long addr = *addrp, last;
	int changed = 0;
again:
	last = addr + size;
	for (i = 0; i < MAX_EARLY_RES && early_res[i].end; i++) {
		struct early_res *r = &early_res[i];
		if (last >= r->start && addr < r->end) {
			*addrp = addr = r->end;
			changed = 1;
			goto again;
		}
112
	}
113
	return changed;
114
}
L
Linus Torvalds 已提交
115

116 117 118 119
/*
 * This function checks if any part of the range <start,end> is mapped
 * with type.
 */
120
int
121
e820_any_mapped(unsigned long start, unsigned long end, unsigned type)
122
{
L
Linus Torvalds 已提交
123
	int i;
124 125 126 127 128

	for (i = 0; i < e820.nr_map; i++) {
		struct e820entry *ei = &e820.map[i];

		if (type && ei->type != type)
L
Linus Torvalds 已提交
129
			continue;
130
		if (ei->addr >= end || ei->addr + ei->size <= start)
131 132 133
			continue;
		return 1;
	}
L
Linus Torvalds 已提交
134 135
	return 0;
}
136
EXPORT_SYMBOL_GPL(e820_any_mapped);
L
Linus Torvalds 已提交
137

138 139 140 141 142 143
/*
 * This function checks if the entire range <start,end> is mapped with type.
 *
 * Note: this function only works correct if the e820 table is sorted and
 * not-overlapping, which is the case
 */
144 145
int __init e820_all_mapped(unsigned long start, unsigned long end,
			   unsigned type)
146 147
{
	int i;
148

149 150
	for (i = 0; i < e820.nr_map; i++) {
		struct e820entry *ei = &e820.map[i];
151

152 153 154 155 156 157 158 159 160 161 162
		if (type && ei->type != type)
			continue;
		/* is the region (part) in overlap with the current region ?*/
		if (ei->addr >= end || ei->addr + ei->size <= start)
			continue;

		/* if the region is at the beginning of <start,end> we move
		 * start to the end of the region since it's ok until there
		 */
		if (ei->addr <= start)
			start = ei->addr + ei->size;
163 164 165 166
		/*
		 * if start is now at or beyond end, we're done, full
		 * coverage
		 */
167
		if (start >= end)
168
			return 1;
169 170 171 172
	}
	return 0;
}

173
/*
174
 * Find a free area with specified alignment in a specific range.
175 176
 */
unsigned long __init find_e820_area(unsigned long start, unsigned long end,
177
				    unsigned size, unsigned long align)
178 179
{
	int i;
180
	unsigned long mask = ~(align - 1);
181 182 183 184 185 186 187 188

	for (i = 0; i < e820.nr_map; i++) {
		struct e820entry *ei = &e820.map[i];
		unsigned long addr = ei->addr, last;

		if (ei->type != E820_RAM)
			continue;
		if (addr < start)
L
Linus Torvalds 已提交
189
			addr = start;
190 191
		if (addr > ei->addr + ei->size)
			continue;
192
		while (bad_addr(&addr, size) && addr+size <= ei->addr+ei->size)
L
Linus Torvalds 已提交
193
			;
194 195
		addr = (addr + align - 1) & mask;
		last = addr + size;
L
Linus Torvalds 已提交
196 197
		if (last > ei->addr + ei->size)
			continue;
198
		if (last > end)
L
Linus Torvalds 已提交
199
			continue;
200 201 202 203
		return addr;
	}
	return -1UL;
}
L
Linus Torvalds 已提交
204 205 206 207 208 209

/*
 * Find the highest page frame number we have available
 */
unsigned long __init e820_end_of_ram(void)
{
210 211
	unsigned long end_pfn;

212
	end_pfn = find_max_pfn_with_active_regions();
213 214

	if (end_pfn > end_pfn_map)
L
Linus Torvalds 已提交
215 216 217 218 219
		end_pfn_map = end_pfn;
	if (end_pfn_map > MAXMEM>>PAGE_SHIFT)
		end_pfn_map = MAXMEM>>PAGE_SHIFT;
	if (end_pfn > end_user_pfn)
		end_pfn = end_user_pfn;
220 221
	if (end_pfn > end_pfn_map)
		end_pfn = end_pfn_map;
L
Linus Torvalds 已提交
222

223 224
	printk(KERN_INFO "end_pfn_map = %lu\n", end_pfn_map);
	return end_pfn;
L
Linus Torvalds 已提交
225 226
}

227
/*
L
Linus Torvalds 已提交
228 229
 * Mark e820 reserved areas as busy for the resource manager.
 */
230 231
void __init e820_reserve_resources(struct resource *code_resource,
		struct resource *data_resource, struct resource *bss_resource)
L
Linus Torvalds 已提交
232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248
{
	int i;
	for (i = 0; i < e820.nr_map; i++) {
		struct resource *res;
		res = alloc_bootmem_low(sizeof(struct resource));
		switch (e820.map[i].type) {
		case E820_RAM:	res->name = "System RAM"; break;
		case E820_ACPI:	res->name = "ACPI Tables"; break;
		case E820_NVS:	res->name = "ACPI Non-volatile Storage"; break;
		default:	res->name = "reserved";
		}
		res->start = e820.map[i].addr;
		res->end = res->start + e820.map[i].size - 1;
		res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
		request_resource(&iomem_resource, res);
		if (e820.map[i].type == E820_RAM) {
			/*
249 250 251
			 * We don't know which RAM region contains kernel data,
			 * so we try it repeatedly and let the resource manager
			 * test it.
L
Linus Torvalds 已提交
252
			 */
253 254 255
			request_resource(res, code_resource);
			request_resource(res, data_resource);
			request_resource(res, bss_resource);
256
#ifdef CONFIG_KEXEC
257 258
			if (crashk_res.start != crashk_res.end)
				request_resource(res, &crashk_res);
259
#endif
L
Linus Torvalds 已提交
260 261 262 263
		}
	}
}

264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281
/*
 * Find the ranges of physical addresses that do not correspond to
 * e820 RAM areas and mark the corresponding pages as nosave for software
 * suspend and suspend to RAM.
 *
 * This function requires the e820 map to be sorted and without any
 * overlapping entries and assumes the first e820 area to be RAM.
 */
void __init e820_mark_nosave_regions(void)
{
	int i;
	unsigned long paddr;

	paddr = round_down(e820.map[0].addr + e820.map[0].size, PAGE_SIZE);
	for (i = 1; i < e820.nr_map; i++) {
		struct e820entry *ei = &e820.map[i];

		if (paddr < ei->addr)
282 283
			register_nosave_region(PFN_DOWN(paddr),
						PFN_UP(ei->addr));
284 285 286

		paddr = round_down(ei->addr + ei->size, PAGE_SIZE);
		if (ei->type != E820_RAM)
287 288
			register_nosave_region(PFN_UP(ei->addr),
						PFN_DOWN(paddr));
289 290 291 292 293 294

		if (paddr >= (end_pfn << PAGE_SHIFT))
			break;
	}
}

295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335
/*
 * Finds an active region in the address range from start_pfn to end_pfn and
 * returns its range in ei_startpfn and ei_endpfn for the e820 entry.
 */
static int __init e820_find_active_region(const struct e820entry *ei,
					  unsigned long start_pfn,
					  unsigned long end_pfn,
					  unsigned long *ei_startpfn,
					  unsigned long *ei_endpfn)
{
	*ei_startpfn = round_up(ei->addr, PAGE_SIZE) >> PAGE_SHIFT;
	*ei_endpfn = round_down(ei->addr + ei->size, PAGE_SIZE) >> PAGE_SHIFT;

	/* Skip map entries smaller than a page */
	if (*ei_startpfn >= *ei_endpfn)
		return 0;

	/* Check if end_pfn_map should be updated */
	if (ei->type != E820_RAM && *ei_endpfn > end_pfn_map)
		end_pfn_map = *ei_endpfn;

	/* Skip if map is outside the node */
	if (ei->type != E820_RAM || *ei_endpfn <= start_pfn ||
				    *ei_startpfn >= end_pfn)
		return 0;

	/* Check for overlaps */
	if (*ei_startpfn < start_pfn)
		*ei_startpfn = start_pfn;
	if (*ei_endpfn > end_pfn)
		*ei_endpfn = end_pfn;

	/* Obey end_user_pfn to save on memmap */
	if (*ei_startpfn >= end_user_pfn)
		return 0;
	if (*ei_endpfn > end_user_pfn)
		*ei_endpfn = end_user_pfn;

	return 1;
}

336 337 338 339 340
/* Walk the e820 map and register active regions within a node */
void __init
e820_register_active_regions(int nid, unsigned long start_pfn,
							unsigned long end_pfn)
{
341 342
	unsigned long ei_startpfn;
	unsigned long ei_endpfn;
343 344
	int i;

345 346 347 348 349
	for (i = 0; i < e820.nr_map; i++)
		if (e820_find_active_region(&e820.map[i],
					    start_pfn, end_pfn,
					    &ei_startpfn, &ei_endpfn))
			add_active_range(nid, ei_startpfn, ei_endpfn);
350 351
}

352
/*
L
Linus Torvalds 已提交
353
 * Add a memory region to the kernel e820 map.
354
 */
L
Linus Torvalds 已提交
355 356 357 358 359 360 361 362 363 364 365 366 367 368 369
void __init add_memory_region(unsigned long start, unsigned long size, int type)
{
	int x = e820.nr_map;

	if (x == E820MAX) {
		printk(KERN_ERR "Ooops! Too many entries in the memory map!\n");
		return;
	}

	e820.map[x].addr = start;
	e820.map[x].size = size;
	e820.map[x].type = type;
	e820.nr_map++;
}

370 371 372 373 374 375 376 377 378
/*
 * Find the hole size (in bytes) in the memory range.
 * @start: starting address of the memory range to scan
 * @end: ending address of the memory range to scan
 */
unsigned long __init e820_hole_size(unsigned long start, unsigned long end)
{
	unsigned long start_pfn = start >> PAGE_SHIFT;
	unsigned long end_pfn = end >> PAGE_SHIFT;
379
	unsigned long ei_startpfn, ei_endpfn, ram = 0;
380 381 382 383 384 385 386 387 388 389 390
	int i;

	for (i = 0; i < e820.nr_map; i++) {
		if (e820_find_active_region(&e820.map[i],
					    start_pfn, end_pfn,
					    &ei_startpfn, &ei_endpfn))
			ram += ei_endpfn - ei_startpfn;
	}
	return end - start - (ram << PAGE_SHIFT);
}

391
static void __init e820_print_map(char *who)
L
Linus Torvalds 已提交
392 393 394 395
{
	int i;

	for (i = 0; i < e820.nr_map; i++) {
396
		printk(KERN_INFO " %s: %016Lx - %016Lx ", who,
397 398 399
		       (unsigned long long) e820.map[i].addr,
		       (unsigned long long)
		       (e820.map[i].addr + e820.map[i].size));
L
Linus Torvalds 已提交
400
		switch (e820.map[i].type) {
401 402 403
		case E820_RAM:
			printk(KERN_CONT "(usable)\n");
			break;
L
Linus Torvalds 已提交
404
		case E820_RESERVED:
405 406
			printk(KERN_CONT "(reserved)\n");
			break;
L
Linus Torvalds 已提交
407
		case E820_ACPI:
408 409
			printk(KERN_CONT "(ACPI data)\n");
			break;
L
Linus Torvalds 已提交
410
		case E820_NVS:
411 412 413 414 415
			printk(KERN_CONT "(ACPI NVS)\n");
			break;
		default:
			printk(KERN_CONT "type %u\n", e820.map[i].type);
			break;
L
Linus Torvalds 已提交
416 417 418 419 420 421 422
		}
	}
}

/*
 * Sanitize the BIOS e820 map.
 *
423
 * Some e820 responses include overlapping entries. The following
L
Linus Torvalds 已提交
424 425 426
 * replaces the original e820 map with a new one, removing overlaps.
 *
 */
427
static int __init sanitize_e820_map(struct e820entry *biosmap, char *pnr_map)
L
Linus Torvalds 已提交
428 429 430 431 432 433 434 435 436 437 438 439 440 441 442
{
	struct change_member {
		struct e820entry *pbios; /* pointer to original bios entry */
		unsigned long long addr; /* address for this change point */
	};
	static struct change_member change_point_list[2*E820MAX] __initdata;
	static struct change_member *change_point[2*E820MAX] __initdata;
	static struct e820entry *overlap_list[E820MAX] __initdata;
	static struct e820entry new_bios[E820MAX] __initdata;
	struct change_member *change_tmp;
	unsigned long current_type, last_type;
	unsigned long long last_addr;
	int chgidx, still_changing;
	int overlap_entries;
	int new_bios_entry;
443
	int old_nr, new_nr, chg_nr;
L
Linus Torvalds 已提交
444 445 446
	int i;

	/*
447 448
		Visually we're performing the following
		(1,2,3,4 = memory types)...
L
Linus Torvalds 已提交
449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489

		Sample memory map (w/overlaps):
		   ____22__________________
		   ______________________4_
		   ____1111________________
		   _44_____________________
		   11111111________________
		   ____________________33__
		   ___________44___________
		   __________33333_________
		   ______________22________
		   ___________________2222_
		   _________111111111______
		   _____________________11_
		   _________________4______

		Sanitized equivalent (no overlap):
		   1_______________________
		   _44_____________________
		   ___1____________________
		   ____22__________________
		   ______11________________
		   _________1______________
		   __________3_____________
		   ___________44___________
		   _____________33_________
		   _______________2________
		   ________________1_______
		   _________________4______
		   ___________________2____
		   ____________________33__
		   ______________________4_
	*/

	/* if there's only one memory region, don't bother */
	if (*pnr_map < 2)
		return -1;

	old_nr = *pnr_map;

	/* bail out if we find any unreasonable addresses in bios map */
490
	for (i = 0; i < old_nr; i++)
L
Linus Torvalds 已提交
491 492 493 494
		if (biosmap[i].addr + biosmap[i].size < biosmap[i].addr)
			return -1;

	/* create pointers for initial change-point information (for sorting) */
495
	for (i = 0; i < 2 * old_nr; i++)
L
Linus Torvalds 已提交
496 497
		change_point[i] = &change_point_list[i];

498 499
	/* record all known change-points (starting and ending addresses),
	   omitting those that are for empty memory regions */
L
Linus Torvalds 已提交
500
	chgidx = 0;
501
	for (i = 0; i < old_nr; i++)	{
502 503 504
		if (biosmap[i].size != 0) {
			change_point[chgidx]->addr = biosmap[i].addr;
			change_point[chgidx++]->pbios = &biosmap[i];
505 506
			change_point[chgidx]->addr = biosmap[i].addr +
				biosmap[i].size;
507 508
			change_point[chgidx++]->pbios = &biosmap[i];
		}
L
Linus Torvalds 已提交
509
	}
510
	chg_nr = chgidx;
L
Linus Torvalds 已提交
511 512 513 514 515

	/* sort change-point list by memory addresses (low -> high) */
	still_changing = 1;
	while (still_changing)	{
		still_changing = 0;
516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534
		for (i = 1; i < chg_nr; i++)  {
			unsigned long long curaddr, lastaddr;
			unsigned long long curpbaddr, lastpbaddr;

			curaddr = change_point[i]->addr;
			lastaddr = change_point[i - 1]->addr;
			curpbaddr = change_point[i]->pbios->addr;
			lastpbaddr = change_point[i - 1]->pbios->addr;

			/*
			 * swap entries, when:
			 *
			 * curaddr > lastaddr or
			 * curaddr == lastaddr and curaddr == curpbaddr and
			 * lastaddr != lastpbaddr
			 */
			if (curaddr < lastaddr ||
			    (curaddr == lastaddr && curaddr == curpbaddr &&
			     lastaddr != lastpbaddr)) {
L
Linus Torvalds 已提交
535 536 537
				change_tmp = change_point[i];
				change_point[i] = change_point[i-1];
				change_point[i-1] = change_tmp;
538
				still_changing = 1;
L
Linus Torvalds 已提交
539 540 541 542 543
			}
		}
	}

	/* create a new bios memory map, removing overlaps */
544 545
	overlap_entries = 0;	 /* number of entries in the overlap table */
	new_bios_entry = 0;	 /* index for creating new bios map entries */
L
Linus Torvalds 已提交
546 547
	last_type = 0;		 /* start with undefined memory type */
	last_addr = 0;		 /* start with 0 as last starting address */
548

L
Linus Torvalds 已提交
549
	/* loop through change-points, determining affect on the new bios map */
550
	for (chgidx = 0; chgidx < chg_nr; chgidx++) {
L
Linus Torvalds 已提交
551
		/* keep track of all overlapping bios entries */
552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569
		if (change_point[chgidx]->addr ==
		    change_point[chgidx]->pbios->addr) {
			/*
			 * add map entry to overlap list (> 1 entry
			 * implies an overlap)
			 */
			overlap_list[overlap_entries++] =
				change_point[chgidx]->pbios;
		} else {
			/*
			 * remove entry from list (order independent,
			 * so swap with last)
			 */
			for (i = 0; i < overlap_entries; i++) {
				if (overlap_list[i] ==
				    change_point[chgidx]->pbios)
					overlap_list[i] =
						overlap_list[overlap_entries-1];
L
Linus Torvalds 已提交
570 571 572
			}
			overlap_entries--;
		}
573 574 575 576 577
		/*
		 * if there are overlapping entries, decide which
		 * "type" to use (larger value takes precedence --
		 * 1=usable, 2,3,4,4+=unusable)
		 */
L
Linus Torvalds 已提交
578
		current_type = 0;
579
		for (i = 0; i < overlap_entries; i++)
L
Linus Torvalds 已提交
580 581
			if (overlap_list[i]->type > current_type)
				current_type = overlap_list[i]->type;
582 583 584 585
		/*
		 * continue building up new bios map based on this
		 * information
		 */
L
Linus Torvalds 已提交
586 587 588 589
		if (current_type != last_type)	{
			if (last_type != 0)	 {
				new_bios[new_bios_entry].size =
					change_point[chgidx]->addr - last_addr;
590 591 592 593
				/*
				 * move forward only if the new size
				 * was non-zero
				 */
L
Linus Torvalds 已提交
594
				if (new_bios[new_bios_entry].size != 0)
595 596 597 598
					/*
					 * no more space left for new
					 * bios entries ?
					 */
L
Linus Torvalds 已提交
599
					if (++new_bios_entry >= E820MAX)
600
						break;
L
Linus Torvalds 已提交
601 602
			}
			if (current_type != 0)	{
603 604
				new_bios[new_bios_entry].addr =
					change_point[chgidx]->addr;
L
Linus Torvalds 已提交
605
				new_bios[new_bios_entry].type = current_type;
606
				last_addr = change_point[chgidx]->addr;
L
Linus Torvalds 已提交
607 608 609 610
			}
			last_type = current_type;
		}
	}
611 612
	/* retain count for new bios entries */
	new_nr = new_bios_entry;
L
Linus Torvalds 已提交
613 614

	/* copy new bios mapping into original location */
615
	memcpy(biosmap, new_bios, new_nr * sizeof(struct e820entry));
L
Linus Torvalds 已提交
616 617 618 619 620 621 622 623 624 625 626 627 628 629
	*pnr_map = new_nr;

	return 0;
}

/*
 * Copy the BIOS e820 map into a safe place.
 *
 * Sanity-check it while we're at it..
 *
 * If we're lucky and live on a modern system, the setup code
 * will have given us a memory map that we can use to properly
 * set up memory.  If we aren't, we'll fake a memory map.
 */
630
static int __init copy_e820_map(struct e820entry *biosmap, int nr_map)
L
Linus Torvalds 已提交
631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646
{
	/* Only one memory region (or negative)? Ignore it */
	if (nr_map < 2)
		return -1;

	do {
		unsigned long start = biosmap->addr;
		unsigned long size = biosmap->size;
		unsigned long end = start + size;
		unsigned long type = biosmap->type;

		/* Overflow in 64 bits? Ignore the memory map. */
		if (start > end)
			return -1;

		add_memory_region(start, size, type);
647
	} while (biosmap++, --nr_map);
L
Linus Torvalds 已提交
648 649 650
	return 0;
}

651
static void early_panic(char *msg)
L
Linus Torvalds 已提交
652
{
653 654 655
	early_printk(msg);
	panic(msg);
}
L
Linus Torvalds 已提交
656

657 658
/* We're not void only for x86 32-bit compat */
char * __init machine_specific_memory_setup(void)
659
{
660
	char *who = "BIOS-e820";
L
Linus Torvalds 已提交
661 662 663 664 665 666
	/*
	 * Try to copy the BIOS-supplied E820-map.
	 *
	 * Otherwise fake a memory map; one section from 0k->640k,
	 * the next section from 1mb->appropriate_mem_k
	 */
667 668
	sanitize_e820_map(boot_params.e820_map, &boot_params.e820_entries);
	if (copy_e820_map(boot_params.e820_map, boot_params.e820_entries) < 0)
669
		early_panic("Cannot find a valid memory map");
L
Linus Torvalds 已提交
670
	printk(KERN_INFO "BIOS-provided physical RAM map:\n");
671 672 673 674
	e820_print_map(who);

	/* In case someone cares... */
	return who;
L
Linus Torvalds 已提交
675 676
}

677 678 679 680 681
static int __init parse_memopt(char *p)
{
	if (!p)
		return -EINVAL;
	end_user_pfn = memparse(p, &p);
682
	end_user_pfn >>= PAGE_SHIFT;
683
	return 0;
684
}
685
early_param("mem", parse_memopt);
L
Linus Torvalds 已提交
686

687
static int userdef __initdata;
L
Linus Torvalds 已提交
688

689
static int __init parse_memmap_opt(char *p)
690
{
691
	char *oldp;
692 693
	unsigned long long start_at, mem_size;

694 695
	if (!strcmp(p, "exactmap")) {
#ifdef CONFIG_CRASH_DUMP
696 697 698
		/*
		 * If we are doing a crash dump, we still need to know
		 * the real mem size before original memory map is
699 700
		 * reset.
		 */
701
		e820_register_active_regions(0, 0, -1UL);
702
		saved_max_pfn = e820_end_of_ram();
703
		remove_all_active_ranges();
704 705 706 707 708 709 710 711 712 713 714
#endif
		end_pfn_map = 0;
		e820.nr_map = 0;
		userdef = 1;
		return 0;
	}

	oldp = p;
	mem_size = memparse(p, &p);
	if (p == oldp)
		return -EINVAL;
715 716

	userdef = 1;
717
	if (*p == '@') {
718
		start_at = memparse(p+1, &p);
719 720
		add_memory_region(start_at, mem_size, E820_RAM);
	} else if (*p == '#') {
721
		start_at = memparse(p+1, &p);
722 723
		add_memory_region(start_at, mem_size, E820_ACPI);
	} else if (*p == '$') {
724
		start_at = memparse(p+1, &p);
725 726 727 728
		add_memory_region(start_at, mem_size, E820_RESERVED);
	} else {
		end_user_pfn = (mem_size >> PAGE_SHIFT);
	}
729 730 731 732
	return *p == '\0' ? 0 : -EINVAL;
}
early_param("memmap", parse_memmap_opt);

733
void __init finish_e820_parsing(void)
734 735
{
	if (userdef) {
736 737 738 739 740 741
		char nr = e820.nr_map;

		if (sanitize_e820_map(e820.map, &nr) < 0)
			early_panic("Invalid user supplied memory map");
		e820.nr_map = nr;

742 743 744
		printk(KERN_INFO "user-defined physical RAM map:\n");
		e820_print_map("user");
	}
745 746
}

Y
Yinghai Lu 已提交
747 748 749 750 751 752 753 754 755 756 757 758
void __init update_e820(void)
{
	u8 nr_map;

	nr_map = e820.nr_map;
	if (sanitize_e820_map(e820.map, &nr_map))
		return;
	e820.nr_map = nr_map;
	printk(KERN_INFO "modified physical RAM map:\n");
	e820_print_map("modified");
}

759
unsigned long pci_mem_start = 0xaeedbabe;
760
EXPORT_SYMBOL(pci_mem_start);
761 762 763 764 765 766 767 768 769

/*
 * Search for the biggest gap in the low 32 bits of the e820
 * memory space.  We pass this space to PCI to assign MMIO resources
 * for hotplug or unconfigured devices in.
 * Hopefully the BIOS let enough space left.
 */
__init void e820_setup_gap(void)
{
770
	unsigned long gapstart, gapsize, round;
771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801
	unsigned long last;
	int i;
	int found = 0;

	last = 0x100000000ull;
	gapstart = 0x10000000;
	gapsize = 0x400000;
	i = e820.nr_map;
	while (--i >= 0) {
		unsigned long long start = e820.map[i].addr;
		unsigned long long end = start + e820.map[i].size;

		/*
		 * Since "last" is at most 4GB, we know we'll
		 * fit in 32 bits if this condition is true
		 */
		if (last > end) {
			unsigned long gap = last - end;

			if (gap > gapsize) {
				gapsize = gap;
				gapstart = end;
				found = 1;
			}
		}
		if (start < last)
			last = start;
	}

	if (!found) {
		gapstart = (end_pfn << PAGE_SHIFT) + 1024*1024;
802 803 804 805
		printk(KERN_ERR "PCI: Warning: Cannot find a gap in the 32bit "
		       "address range\n"
		       KERN_ERR "PCI: Unassigned devices with 32bit resource "
		       "registers may break!\n");
806 807 808
	}

	/*
809 810
	 * See how much we want to round up: start off with
	 * rounding to the next 1MB area.
811
	 */
812 813 814 815 816
	round = 0x100000;
	while ((gapsize >> 4) > round)
		round += round;
	/* Fun with two's complement */
	pci_mem_start = (gapstart + round) & -round;
817

818 819 820
	printk(KERN_INFO
	       "Allocating PCI resources starting at %lx (gap: %lx:%lx)\n",
	       pci_mem_start, gapstart, gapsize);
821
}
822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840

int __init arch_get_ram_range(int slot, u64 *addr, u64 *size)
{
	int i;

	if (slot < 0 || slot >= e820.nr_map)
		return -1;
	for (i = slot; i < e820.nr_map; i++) {
		if (e820.map[i].type != E820_RAM)
			continue;
		break;
	}
	if (i == e820.nr_map || e820.map[i].addr > (max_pfn << PAGE_SHIFT))
		return -1;
	*addr = e820.map[i].addr;
	*size = min_t(u64, e820.map[i].size + e820.map[i].addr,
		max_pfn << PAGE_SHIFT) - *addr;
	return i + 1;
}