e820_64.c 21.3 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
/* Check for already reserved areas */
98 99
static inline int
bad_addr(unsigned long *addrp, unsigned long size, unsigned long align)
100 101 102 103 104 105 106 107 108
{
	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) {
109
			*addrp = addr = round_up(r->end, align);
110 111 112
			changed = 1;
			goto again;
		}
113
	}
114
	return changed;
115
}
L
Linus Torvalds 已提交
116

117 118 119 120
/*
 * This function checks if any part of the range <start,end> is mapped
 * with type.
 */
121
int
122
e820_any_mapped(unsigned long start, unsigned long end, unsigned type)
123
{
L
Linus Torvalds 已提交
124
	int i;
125 126 127 128 129

	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
130
			continue;
131
		if (ei->addr >= end || ei->addr + ei->size <= start)
132 133 134
			continue;
		return 1;
	}
L
Linus Torvalds 已提交
135 136
	return 0;
}
137
EXPORT_SYMBOL_GPL(e820_any_mapped);
L
Linus Torvalds 已提交
138

139 140 141 142 143 144
/*
 * 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
 */
145 146
int __init e820_all_mapped(unsigned long start, unsigned long end,
			   unsigned type)
147 148
{
	int i;
149

150 151
	for (i = 0; i < e820.nr_map; i++) {
		struct e820entry *ei = &e820.map[i];
152

153 154 155 156 157 158 159 160 161 162 163
		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;
164 165 166 167
		/*
		 * if start is now at or beyond end, we're done, full
		 * coverage
		 */
168
		if (start >= end)
169
			return 1;
170 171 172 173
	}
	return 0;
}

174
/*
175
 * Find a free area with specified alignment in a specific range.
176 177
 */
unsigned long __init find_e820_area(unsigned long start, unsigned long end,
178
				    unsigned long size, unsigned long align)
179 180 181 182 183
{
	int i;

	for (i = 0; i < e820.nr_map; i++) {
		struct e820entry *ei = &e820.map[i];
184 185
		unsigned long addr, last;
		unsigned long ei_last;
186 187 188

		if (ei->type != E820_RAM)
			continue;
189 190
		addr = round_up(ei->addr, align);
		ei_last = ei->addr + ei->size;
191
		if (addr < start)
192 193
			addr = round_up(start, align);
		if (addr > ei_last)
194
			continue;
195
		while (bad_addr(&addr, size, align) && addr+size <= ei_last)
L
Linus Torvalds 已提交
196
			;
197
		last = addr + size;
198
		if (last > ei_last)
L
Linus Torvalds 已提交
199
			continue;
200
		if (last > end)
L
Linus Torvalds 已提交
201
			continue;
202 203 204 205
		return addr;
	}
	return -1UL;
}
L
Linus Torvalds 已提交
206 207 208 209 210 211

/*
 * Find the highest page frame number we have available
 */
unsigned long __init e820_end_of_ram(void)
{
212 213
	unsigned long end_pfn;

214
	end_pfn = find_max_pfn_with_active_regions();
215 216

	if (end_pfn > end_pfn_map)
L
Linus Torvalds 已提交
217 218 219 220 221
		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;
222 223
	if (end_pfn > end_pfn_map)
		end_pfn = end_pfn_map;
L
Linus Torvalds 已提交
224

225 226
	printk(KERN_INFO "end_pfn_map = %lu\n", end_pfn_map);
	return end_pfn;
L
Linus Torvalds 已提交
227 228
}

229
/*
L
Linus Torvalds 已提交
230 231
 * Mark e820 reserved areas as busy for the resource manager.
 */
232
void __init e820_reserve_resources(void)
L
Linus Torvalds 已提交
233 234 235 236 237 238 239 240 241 242 243 244 245 246
{
	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;
247
		insert_resource(&iomem_resource, res);
L
Linus Torvalds 已提交
248 249 250
	}
}

251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268
/*
 * 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)
269 270
			register_nosave_region(PFN_DOWN(paddr),
						PFN_UP(ei->addr));
271 272 273

		paddr = round_down(ei->addr + ei->size, PAGE_SIZE);
		if (ei->type != E820_RAM)
274 275
			register_nosave_region(PFN_UP(ei->addr),
						PFN_DOWN(paddr));
276 277 278 279 280 281

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

282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322
/*
 * 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;
}

323 324 325 326 327
/* 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)
{
328 329
	unsigned long ei_startpfn;
	unsigned long ei_endpfn;
330 331
	int i;

332 333 334 335 336
	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);
337 338
}

339
/*
L
Linus Torvalds 已提交
340
 * Add a memory region to the kernel e820 map.
341
 */
L
Linus Torvalds 已提交
342 343 344 345 346 347 348 349 350 351 352 353 354 355 356
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++;
}

357 358 359 360 361 362 363 364 365
/*
 * 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;
366
	unsigned long ei_startpfn, ei_endpfn, ram = 0;
367 368 369 370 371 372 373 374 375 376 377
	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);
}

378
static void __init e820_print_map(char *who)
L
Linus Torvalds 已提交
379 380 381 382
{
	int i;

	for (i = 0; i < e820.nr_map; i++) {
383
		printk(KERN_INFO " %s: %016Lx - %016Lx ", who,
384 385 386
		       (unsigned long long) e820.map[i].addr,
		       (unsigned long long)
		       (e820.map[i].addr + e820.map[i].size));
L
Linus Torvalds 已提交
387
		switch (e820.map[i].type) {
388 389 390
		case E820_RAM:
			printk(KERN_CONT "(usable)\n");
			break;
L
Linus Torvalds 已提交
391
		case E820_RESERVED:
392 393
			printk(KERN_CONT "(reserved)\n");
			break;
L
Linus Torvalds 已提交
394
		case E820_ACPI:
395 396
			printk(KERN_CONT "(ACPI data)\n");
			break;
L
Linus Torvalds 已提交
397
		case E820_NVS:
398 399 400 401 402
			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
403 404 405 406 407 408 409
		}
	}
}

/*
 * Sanitize the BIOS e820 map.
 *
410
 * Some e820 responses include overlapping entries. The following
L
Linus Torvalds 已提交
411 412 413
 * replaces the original e820 map with a new one, removing overlaps.
 *
 */
414
static int __init sanitize_e820_map(struct e820entry *biosmap, char *pnr_map)
L
Linus Torvalds 已提交
415 416 417 418 419 420 421 422 423 424 425 426 427 428 429
{
	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;
430
	int old_nr, new_nr, chg_nr;
L
Linus Torvalds 已提交
431 432 433
	int i;

	/*
434 435
		Visually we're performing the following
		(1,2,3,4 = memory types)...
L
Linus Torvalds 已提交
436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476

		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 */
477
	for (i = 0; i < old_nr; i++)
L
Linus Torvalds 已提交
478 479 480 481
		if (biosmap[i].addr + biosmap[i].size < biosmap[i].addr)
			return -1;

	/* create pointers for initial change-point information (for sorting) */
482
	for (i = 0; i < 2 * old_nr; i++)
L
Linus Torvalds 已提交
483 484
		change_point[i] = &change_point_list[i];

485 486
	/* record all known change-points (starting and ending addresses),
	   omitting those that are for empty memory regions */
L
Linus Torvalds 已提交
487
	chgidx = 0;
488
	for (i = 0; i < old_nr; i++)	{
489 490 491
		if (biosmap[i].size != 0) {
			change_point[chgidx]->addr = biosmap[i].addr;
			change_point[chgidx++]->pbios = &biosmap[i];
492 493
			change_point[chgidx]->addr = biosmap[i].addr +
				biosmap[i].size;
494 495
			change_point[chgidx++]->pbios = &biosmap[i];
		}
L
Linus Torvalds 已提交
496
	}
497
	chg_nr = chgidx;
L
Linus Torvalds 已提交
498 499 500 501 502

	/* sort change-point list by memory addresses (low -> high) */
	still_changing = 1;
	while (still_changing)	{
		still_changing = 0;
503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521
		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
522 523 524
				change_tmp = change_point[i];
				change_point[i] = change_point[i-1];
				change_point[i-1] = change_tmp;
525
				still_changing = 1;
L
Linus Torvalds 已提交
526 527 528 529 530
			}
		}
	}

	/* create a new bios memory map, removing overlaps */
531 532
	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
533 534
	last_type = 0;		 /* start with undefined memory type */
	last_addr = 0;		 /* start with 0 as last starting address */
535

L
Linus Torvalds 已提交
536
	/* loop through change-points, determining affect on the new bios map */
537
	for (chgidx = 0; chgidx < chg_nr; chgidx++) {
L
Linus Torvalds 已提交
538
		/* keep track of all overlapping bios entries */
539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556
		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
557 558 559
			}
			overlap_entries--;
		}
560 561 562 563 564
		/*
		 * 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
565
		current_type = 0;
566
		for (i = 0; i < overlap_entries; i++)
L
Linus Torvalds 已提交
567 568
			if (overlap_list[i]->type > current_type)
				current_type = overlap_list[i]->type;
569 570 571 572
		/*
		 * continue building up new bios map based on this
		 * information
		 */
L
Linus Torvalds 已提交
573 574 575 576
		if (current_type != last_type)	{
			if (last_type != 0)	 {
				new_bios[new_bios_entry].size =
					change_point[chgidx]->addr - last_addr;
577 578 579 580
				/*
				 * move forward only if the new size
				 * was non-zero
				 */
L
Linus Torvalds 已提交
581
				if (new_bios[new_bios_entry].size != 0)
582 583 584 585
					/*
					 * no more space left for new
					 * bios entries ?
					 */
L
Linus Torvalds 已提交
586
					if (++new_bios_entry >= E820MAX)
587
						break;
L
Linus Torvalds 已提交
588 589
			}
			if (current_type != 0)	{
590 591
				new_bios[new_bios_entry].addr =
					change_point[chgidx]->addr;
L
Linus Torvalds 已提交
592
				new_bios[new_bios_entry].type = current_type;
593
				last_addr = change_point[chgidx]->addr;
L
Linus Torvalds 已提交
594 595 596 597
			}
			last_type = current_type;
		}
	}
598 599
	/* retain count for new bios entries */
	new_nr = new_bios_entry;
L
Linus Torvalds 已提交
600 601

	/* copy new bios mapping into original location */
602
	memcpy(biosmap, new_bios, new_nr * sizeof(struct e820entry));
L
Linus Torvalds 已提交
603 604 605 606 607 608 609 610 611 612 613 614 615 616
	*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.
 */
617
static int __init copy_e820_map(struct e820entry *biosmap, int nr_map)
L
Linus Torvalds 已提交
618 619 620 621 622 623
{
	/* Only one memory region (or negative)? Ignore it */
	if (nr_map < 2)
		return -1;

	do {
624 625 626 627
		u64 start = biosmap->addr;
		u64 size = biosmap->size;
		u64 end = start + size;
		u32 type = biosmap->type;
L
Linus Torvalds 已提交
628 629 630 631 632 633

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

		add_memory_region(start, size, type);
634
	} while (biosmap++, --nr_map);
L
Linus Torvalds 已提交
635 636 637
	return 0;
}

638
static void early_panic(char *msg)
L
Linus Torvalds 已提交
639
{
640 641 642
	early_printk(msg);
	panic(msg);
}
L
Linus Torvalds 已提交
643

644 645
/* We're not void only for x86 32-bit compat */
char * __init machine_specific_memory_setup(void)
646
{
647
	char *who = "BIOS-e820";
L
Linus Torvalds 已提交
648 649 650 651 652 653
	/*
	 * 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
	 */
654 655
	sanitize_e820_map(boot_params.e820_map, &boot_params.e820_entries);
	if (copy_e820_map(boot_params.e820_map, boot_params.e820_entries) < 0)
656
		early_panic("Cannot find a valid memory map");
L
Linus Torvalds 已提交
657
	printk(KERN_INFO "BIOS-provided physical RAM map:\n");
658 659 660 661
	e820_print_map(who);

	/* In case someone cares... */
	return who;
L
Linus Torvalds 已提交
662 663
}

664 665 666 667 668
static int __init parse_memopt(char *p)
{
	if (!p)
		return -EINVAL;
	end_user_pfn = memparse(p, &p);
669
	end_user_pfn >>= PAGE_SHIFT;
670
	return 0;
671
}
672
early_param("mem", parse_memopt);
L
Linus Torvalds 已提交
673

674
static int userdef __initdata;
L
Linus Torvalds 已提交
675

676
static int __init parse_memmap_opt(char *p)
677
{
678
	char *oldp;
679 680
	unsigned long long start_at, mem_size;

681 682
	if (!strcmp(p, "exactmap")) {
#ifdef CONFIG_CRASH_DUMP
683 684 685
		/*
		 * If we are doing a crash dump, we still need to know
		 * the real mem size before original memory map is
686 687
		 * reset.
		 */
688
		e820_register_active_regions(0, 0, -1UL);
689
		saved_max_pfn = e820_end_of_ram();
690
		remove_all_active_ranges();
691 692 693 694 695 696 697 698 699 700 701
#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;
702 703

	userdef = 1;
704
	if (*p == '@') {
705
		start_at = memparse(p+1, &p);
706 707
		add_memory_region(start_at, mem_size, E820_RAM);
	} else if (*p == '#') {
708
		start_at = memparse(p+1, &p);
709 710
		add_memory_region(start_at, mem_size, E820_ACPI);
	} else if (*p == '$') {
711
		start_at = memparse(p+1, &p);
712 713 714 715
		add_memory_region(start_at, mem_size, E820_RESERVED);
	} else {
		end_user_pfn = (mem_size >> PAGE_SHIFT);
	}
716 717 718 719
	return *p == '\0' ? 0 : -EINVAL;
}
early_param("memmap", parse_memmap_opt);

720
void __init finish_e820_parsing(void)
721 722
{
	if (userdef) {
723 724 725 726 727 728
		char nr = e820.nr_map;

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

729 730 731
		printk(KERN_INFO "user-defined physical RAM map:\n");
		e820_print_map("user");
	}
732 733
}

734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760
void __init update_memory_range(u64 start, u64 size, unsigned old_type,
				unsigned new_type)
{
	int i;

	BUG_ON(old_type == new_type);

	for (i = 0; i < e820.nr_map; i++) {
		struct e820entry *ei = &e820.map[i];
		u64 final_start, final_end;
		if (ei->type != old_type)
			continue;
		/* totally covered? */
		if (ei->addr >= start && ei->size <= size) {
			ei->type = new_type;
			continue;
		}
		/* partially covered */
		final_start = max(start, ei->addr);
		final_end = min(start + size, ei->addr + ei->size);
		if (final_start >= final_end)
			continue;
		add_memory_region(final_start, final_end - final_start,
					 new_type);
	}
}

Y
Yinghai Lu 已提交
761 762 763 764 765 766 767 768 769 770 771 772
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");
}

773
unsigned long pci_mem_start = 0xaeedbabe;
774
EXPORT_SYMBOL(pci_mem_start);
775 776 777 778 779 780 781 782 783

/*
 * 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)
{
784
	unsigned long gapstart, gapsize, round;
785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815
	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;
816 817 818 819
		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");
820 821 822
	}

	/*
823 824
	 * See how much we want to round up: start off with
	 * rounding to the next 1MB area.
825
	 */
826 827 828 829 830
	round = 0x100000;
	while ((gapsize >> 4) > round)
		round += round;
	/* Fun with two's complement */
	pci_mem_start = (gapstart + round) & -round;
831

832 833 834
	printk(KERN_INFO
	       "Allocating PCI resources starting at %lx (gap: %lx:%lx)\n",
	       pci_mem_start, gapstart, gapsize);
835
}
836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854

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;
}