memblock.c 22.5 KB
Newer Older
Y
Yinghai Lu 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13
/*
 * Procedures for maintaining information about logical memory blocks.
 *
 * Peter Bergner, IBM Corp.	June 2001.
 * Copyright (C) 2001 Peter Bergner.
 *
 *      This program is free software; you can redistribute it and/or
 *      modify it under the terms of the GNU General Public License
 *      as published by the Free Software Foundation; either version
 *      2 of the License, or (at your option) any later version.
 */

#include <linux/kernel.h>
14
#include <linux/slab.h>
Y
Yinghai Lu 已提交
15 16
#include <linux/init.h>
#include <linux/bitops.h>
17
#include <linux/poison.h>
18
#include <linux/pfn.h>
19 20
#include <linux/debugfs.h>
#include <linux/seq_file.h>
Y
Yinghai Lu 已提交
21 22
#include <linux/memblock.h>

23
struct memblock memblock __initdata_memblock;
Y
Yinghai Lu 已提交
24

25 26 27 28
int memblock_debug __initdata_memblock;
int memblock_can_resize __initdata_memblock;
static struct memblock_region memblock_memory_init_regions[INIT_MEMBLOCK_REGIONS + 1] __initdata_memblock;
static struct memblock_region memblock_reserved_init_regions[INIT_MEMBLOCK_REGIONS + 1] __initdata_memblock;
Y
Yinghai Lu 已提交
29

30 31 32 33 34 35 36 37 38 39 40
/* inline so we don't get a warning when pr_debug is compiled out */
static inline const char *memblock_type_name(struct memblock_type *type)
{
	if (type == &memblock.memory)
		return "memory";
	else if (type == &memblock.reserved)
		return "reserved";
	else
		return "unknown";
}

41 42 43
/*
 * Address comparison utilities
 */
Y
Yinghai Lu 已提交
44

45
static phys_addr_t __init_memblock memblock_align_down(phys_addr_t addr, phys_addr_t size)
Y
Yinghai Lu 已提交
46
{
47
	return addr & ~(size - 1);
Y
Yinghai Lu 已提交
48 49
}

50
static phys_addr_t __init_memblock memblock_align_up(phys_addr_t addr, phys_addr_t size)
Y
Yinghai Lu 已提交
51
{
52
	return (addr + (size - 1)) & ~(size - 1);
Y
Yinghai Lu 已提交
53 54
}

55
static unsigned long __init_memblock memblock_addrs_overlap(phys_addr_t base1, phys_addr_t size1,
56
				       phys_addr_t base2, phys_addr_t size2)
Y
Yinghai Lu 已提交
57 58 59 60
{
	return ((base1 < (base2 + size2)) && (base2 < (base1 + size1)));
}

61
static long __init_memblock memblock_addrs_adjacent(phys_addr_t base1, phys_addr_t size1,
62
			       phys_addr_t base2, phys_addr_t size2)
Y
Yinghai Lu 已提交
63 64 65 66 67 68 69 70 71
{
	if (base2 == base1 + size1)
		return 1;
	else if (base1 == base2 + size2)
		return -1;

	return 0;
}

72
static long __init_memblock memblock_regions_adjacent(struct memblock_type *type,
73
				 unsigned long r1, unsigned long r2)
Y
Yinghai Lu 已提交
74
{
75 76 77 78
	phys_addr_t base1 = type->regions[r1].base;
	phys_addr_t size1 = type->regions[r1].size;
	phys_addr_t base2 = type->regions[r2].base;
	phys_addr_t size2 = type->regions[r2].size;
Y
Yinghai Lu 已提交
79 80 81 82

	return memblock_addrs_adjacent(base1, size1, base2, size2);
}

83
long __init_memblock memblock_overlaps_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101
{
	unsigned long i;

	for (i = 0; i < type->cnt; i++) {
		phys_addr_t rgnbase = type->regions[i].base;
		phys_addr_t rgnsize = type->regions[i].size;
		if (memblock_addrs_overlap(base, size, rgnbase, rgnsize))
			break;
	}

	return (i < type->cnt) ? i : -1;
}

/*
 * Find, allocate, deallocate or reserve unreserved regions. All allocations
 * are top-down.
 */

102
static phys_addr_t __init_memblock memblock_find_region(phys_addr_t start, phys_addr_t end,
103 104 105 106 107
					  phys_addr_t size, phys_addr_t align)
{
	phys_addr_t base, res_base;
	long j;

108 109 110 111 112 113
	/* In case, huge size is requested */
	if (end < size)
		return MEMBLOCK_ERROR;

	base = memblock_align_down((end - size), align);

114 115 116 117 118 119
	/* Prevent allocations returning 0 as it's also used to
	 * indicate an allocation failure
	 */
	if (start == 0)
		start = PAGE_SIZE;

120 121 122 123 124 125 126 127 128 129 130 131 132
	while (start <= base) {
		j = memblock_overlaps_region(&memblock.reserved, base, size);
		if (j < 0)
			return base;
		res_base = memblock.reserved.regions[j].base;
		if (res_base < size)
			break;
		base = memblock_align_down(res_base - size, align);
	}

	return MEMBLOCK_ERROR;
}

133 134
static phys_addr_t __init_memblock memblock_find_base(phys_addr_t size,
			phys_addr_t align, phys_addr_t start, phys_addr_t end)
135 136 137 138 139 140 141 142
{
	long i;

	BUG_ON(0 == size);

	size = memblock_align_up(size, align);

	/* Pump up max_addr */
143 144
	if (end == MEMBLOCK_ALLOC_ACCESSIBLE)
		end = memblock.current_limit;
145 146 147 148 149 150 151 152

	/* We do a top-down search, this tends to limit memory
	 * fragmentation by keeping early boot allocs near the
	 * top of memory
	 */
	for (i = memblock.memory.cnt - 1; i >= 0; i--) {
		phys_addr_t memblockbase = memblock.memory.regions[i].base;
		phys_addr_t memblocksize = memblock.memory.regions[i].size;
153
		phys_addr_t bottom, top, found;
154 155 156

		if (memblocksize < size)
			continue;
157 158 159 160 161 162 163 164 165
		if ((memblockbase + memblocksize) <= start)
			break;
		bottom = max(memblockbase, start);
		top = min(memblockbase + memblocksize, end);
		if (bottom >= top)
			continue;
		found = memblock_find_region(bottom, top, size, align);
		if (found != MEMBLOCK_ERROR)
			return found;
166 167 168 169
	}
	return MEMBLOCK_ERROR;
}

170 171 172 173 174 175 176 177
/*
 * Find a free area with specified alignment in a specific range.
 */
u64 __init_memblock memblock_find_in_range(u64 start, u64 end, u64 size, u64 align)
{
	return memblock_find_base(size, align, start, end);
}

178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201
/*
 * Free memblock.reserved.regions
 */
int __init_memblock memblock_free_reserved_regions(void)
{
	if (memblock.reserved.regions == memblock_reserved_init_regions)
		return 0;

	return memblock_free(__pa(memblock.reserved.regions),
		 sizeof(struct memblock_region) * memblock.reserved.max);
}

/*
 * Reserve memblock.reserved.regions
 */
int __init_memblock memblock_reserve_reserved_regions(void)
{
	if (memblock.reserved.regions == memblock_reserved_init_regions)
		return 0;

	return memblock_reserve(__pa(memblock.reserved.regions),
		 sizeof(struct memblock_region) * memblock.reserved.max);
}

202
static void __init_memblock memblock_remove_region(struct memblock_type *type, unsigned long r)
Y
Yinghai Lu 已提交
203 204 205
{
	unsigned long i;

206 207 208
	for (i = r; i < type->cnt - 1; i++) {
		type->regions[i].base = type->regions[i + 1].base;
		type->regions[i].size = type->regions[i + 1].size;
Y
Yinghai Lu 已提交
209
	}
210
	type->cnt--;
Y
Yinghai Lu 已提交
211 212 213
}

/* Assumption: base addr of region 1 < base addr of region 2 */
214
static void __init_memblock memblock_coalesce_regions(struct memblock_type *type,
Y
Yinghai Lu 已提交
215 216
		unsigned long r1, unsigned long r2)
{
217 218
	type->regions[r1].size += type->regions[r2].size;
	memblock_remove_region(type, r2);
Y
Yinghai Lu 已提交
219 220
}

221 222 223
/* Defined below but needed now */
static long memblock_add_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size);

224
static int __init_memblock memblock_double_array(struct memblock_type *type)
225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254
{
	struct memblock_region *new_array, *old_array;
	phys_addr_t old_size, new_size, addr;
	int use_slab = slab_is_available();

	/* We don't allow resizing until we know about the reserved regions
	 * of memory that aren't suitable for allocation
	 */
	if (!memblock_can_resize)
		return -1;

	/* Calculate new doubled size */
	old_size = type->max * sizeof(struct memblock_region);
	new_size = old_size << 1;

	/* Try to find some space for it.
	 *
	 * WARNING: We assume that either slab_is_available() and we use it or
	 * we use MEMBLOCK for allocations. That means that this is unsafe to use
	 * when bootmem is currently active (unless bootmem itself is implemented
	 * on top of MEMBLOCK which isn't the case yet)
	 *
	 * This should however not be an issue for now, as we currently only
	 * call into MEMBLOCK while it's still active, or much later when slab is
	 * active for memory hotplug operations
	 */
	if (use_slab) {
		new_array = kmalloc(new_size, GFP_KERNEL);
		addr = new_array == NULL ? MEMBLOCK_ERROR : __pa(new_array);
	} else
255
		addr = memblock_find_base(new_size, sizeof(phys_addr_t), 0, MEMBLOCK_ALLOC_ACCESSIBLE);
256 257 258 259 260 261 262
	if (addr == MEMBLOCK_ERROR) {
		pr_err("memblock: Failed to double %s array from %ld to %ld entries !\n",
		       memblock_type_name(type), type->max, type->max * 2);
		return -1;
	}
	new_array = __va(addr);

263 264 265
	memblock_dbg("memblock: %s array is doubled to %ld at [%#010llx-%#010llx]",
		 memblock_type_name(type), type->max * 2, (u64)addr, (u64)addr + new_size - 1);

266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294
	/* Found space, we now need to move the array over before
	 * we add the reserved region since it may be our reserved
	 * array itself that is full.
	 */
	memcpy(new_array, type->regions, old_size);
	memset(new_array + type->max, 0, old_size);
	old_array = type->regions;
	type->regions = new_array;
	type->max <<= 1;

	/* If we use SLAB that's it, we are done */
	if (use_slab)
		return 0;

	/* Add the new reserved region now. Should not fail ! */
	BUG_ON(memblock_add_region(&memblock.reserved, addr, new_size) < 0);

	/* If the array wasn't our static init one, then free it. We only do
	 * that before SLAB is available as later on, we don't know whether
	 * to use kfree or free_bootmem_pages(). Shouldn't be a big deal
	 * anyways
	 */
	if (old_array != memblock_memory_init_regions &&
	    old_array != memblock_reserved_init_regions)
		memblock_free(__pa(old_array), old_size);

	return 0;
}

295
extern int __init_memblock __weak memblock_memory_can_coalesce(phys_addr_t addr1, phys_addr_t size1,
296 297 298 299 300
					  phys_addr_t addr2, phys_addr_t size2)
{
	return 1;
}

301
static long __init_memblock memblock_add_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
302 303 304 305
{
	unsigned long coalesced = 0;
	long adjacent, i;

306 307 308
	if ((type->cnt == 1) && (type->regions[0].size == 0)) {
		type->regions[0].base = base;
		type->regions[0].size = size;
Y
Yinghai Lu 已提交
309 310 311 312
		return 0;
	}

	/* First try and coalesce this MEMBLOCK with another. */
313
	for (i = 0; i < type->cnt; i++) {
314 315
		phys_addr_t rgnbase = type->regions[i].base;
		phys_addr_t rgnsize = type->regions[i].size;
Y
Yinghai Lu 已提交
316 317 318 319 320 321

		if ((rgnbase == base) && (rgnsize == size))
			/* Already have this region, so we're done */
			return 0;

		adjacent = memblock_addrs_adjacent(base, size, rgnbase, rgnsize);
322 323 324 325
		/* Check if arch allows coalescing */
		if (adjacent != 0 && type == &memblock.memory &&
		    !memblock_memory_can_coalesce(base, size, rgnbase, rgnsize))
			break;
Y
Yinghai Lu 已提交
326
		if (adjacent > 0) {
327 328
			type->regions[i].base -= size;
			type->regions[i].size += size;
Y
Yinghai Lu 已提交
329 330 331
			coalesced++;
			break;
		} else if (adjacent < 0) {
332
			type->regions[i].size += size;
Y
Yinghai Lu 已提交
333 334 335 336 337
			coalesced++;
			break;
		}
	}

338 339 340 341 342 343 344 345
	/* If we plugged a hole, we may want to also coalesce with the
	 * next region
	 */
	if ((i < type->cnt - 1) && memblock_regions_adjacent(type, i, i+1) &&
	    ((type != &memblock.memory || memblock_memory_can_coalesce(type->regions[i].base,
							     type->regions[i].size,
							     type->regions[i+1].base,
							     type->regions[i+1].size)))) {
346
		memblock_coalesce_regions(type, i, i+1);
Y
Yinghai Lu 已提交
347 348 349 350 351
		coalesced++;
	}

	if (coalesced)
		return coalesced;
352 353 354 355 356

	/* If we are out of space, we fail. It's too late to resize the array
	 * but then this shouldn't have happened in the first place.
	 */
	if (WARN_ON(type->cnt >= type->max))
Y
Yinghai Lu 已提交
357 358 359
		return -1;

	/* Couldn't coalesce the MEMBLOCK, so add it to the sorted table. */
360 361 362 363
	for (i = type->cnt - 1; i >= 0; i--) {
		if (base < type->regions[i].base) {
			type->regions[i+1].base = type->regions[i].base;
			type->regions[i+1].size = type->regions[i].size;
Y
Yinghai Lu 已提交
364
		} else {
365 366
			type->regions[i+1].base = base;
			type->regions[i+1].size = size;
Y
Yinghai Lu 已提交
367 368 369 370
			break;
		}
	}

371 372 373
	if (base < type->regions[0].base) {
		type->regions[0].base = base;
		type->regions[0].size = size;
Y
Yinghai Lu 已提交
374
	}
375
	type->cnt++;
Y
Yinghai Lu 已提交
376

377 378 379 380 381 382 383 384
	/* The array is full ? Try to resize it. If that fails, we undo
	 * our allocation and return an error
	 */
	if (type->cnt == type->max && memblock_double_array(type)) {
		type->cnt--;
		return -1;
	}

Y
Yinghai Lu 已提交
385 386 387
	return 0;
}

388
long __init_memblock memblock_add(phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
389
{
390
	return memblock_add_region(&memblock.memory, base, size);
Y
Yinghai Lu 已提交
391 392 393

}

394
static long __init_memblock __memblock_remove(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
395
{
396 397
	phys_addr_t rgnbegin, rgnend;
	phys_addr_t end = base + size;
Y
Yinghai Lu 已提交
398 399 400 401 402
	int i;

	rgnbegin = rgnend = 0; /* supress gcc warnings */

	/* Find the region where (base, size) belongs to */
403 404 405
	for (i=0; i < type->cnt; i++) {
		rgnbegin = type->regions[i].base;
		rgnend = rgnbegin + type->regions[i].size;
Y
Yinghai Lu 已提交
406 407 408 409 410 411

		if ((rgnbegin <= base) && (end <= rgnend))
			break;
	}

	/* Didn't find the region */
412
	if (i == type->cnt)
Y
Yinghai Lu 已提交
413 414 415 416
		return -1;

	/* Check to see if we are removing entire region */
	if ((rgnbegin == base) && (rgnend == end)) {
417
		memblock_remove_region(type, i);
Y
Yinghai Lu 已提交
418 419 420 421 422
		return 0;
	}

	/* Check to see if region is matching at the front */
	if (rgnbegin == base) {
423 424
		type->regions[i].base = end;
		type->regions[i].size -= size;
Y
Yinghai Lu 已提交
425 426 427 428 429
		return 0;
	}

	/* Check to see if the region is matching at the end */
	if (rgnend == end) {
430
		type->regions[i].size -= size;
Y
Yinghai Lu 已提交
431 432 433 434 435 436 437
		return 0;
	}

	/*
	 * We need to split the entry -  adjust the current one to the
	 * beginging of the hole and add the region after hole.
	 */
438 439
	type->regions[i].size = base - type->regions[i].base;
	return memblock_add_region(type, end, rgnend - end);
Y
Yinghai Lu 已提交
440 441
}

442
long __init_memblock memblock_remove(phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
443 444 445 446
{
	return __memblock_remove(&memblock.memory, base, size);
}

447
long __init_memblock memblock_free(phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
448 449 450 451
{
	return __memblock_remove(&memblock.reserved, base, size);
}

452
long __init_memblock memblock_reserve(phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
453
{
454
	struct memblock_type *_rgn = &memblock.reserved;
Y
Yinghai Lu 已提交
455 456 457 458 459 460

	BUG_ON(0 == size);

	return memblock_add_region(_rgn, base, size);
}

461
phys_addr_t __init __memblock_alloc_base(phys_addr_t size, phys_addr_t align, phys_addr_t max_addr)
Y
Yinghai Lu 已提交
462
{
463
	phys_addr_t found;
Y
Yinghai Lu 已提交
464

465 466 467 468
	/* We align the size to limit fragmentation. Without this, a lot of
	 * small allocs quickly eat up the whole reserve array on sparc
	 */
	size = memblock_align_up(size, align);
Y
Yinghai Lu 已提交
469

470
	found = memblock_find_base(size, align, 0, max_addr);
471 472 473
	if (found != MEMBLOCK_ERROR &&
	    memblock_add_region(&memblock.reserved, found, size) >= 0)
		return found;
Y
Yinghai Lu 已提交
474

475
	return 0;
Y
Yinghai Lu 已提交
476 477
}

478
phys_addr_t __init memblock_alloc_base(phys_addr_t size, phys_addr_t align, phys_addr_t max_addr)
Y
Yinghai Lu 已提交
479
{
480 481 482 483 484 485 486 487 488
	phys_addr_t alloc;

	alloc = __memblock_alloc_base(size, align, max_addr);

	if (alloc == 0)
		panic("ERROR: Failed to allocate 0x%llx bytes below 0x%llx.\n",
		      (unsigned long long) size, (unsigned long long) max_addr);

	return alloc;
Y
Yinghai Lu 已提交
489 490
}

491
phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align)
Y
Yinghai Lu 已提交
492
{
493 494
	return memblock_alloc_base(size, align, MEMBLOCK_ALLOC_ACCESSIBLE);
}
Y
Yinghai Lu 已提交
495 496


497 498 499
/*
 * Additional node-local allocators. Search for node memory is bottom up
 * and walks memblock regions within that node bottom-up as well, but allocation
500 501 502 503 504
 * within an memblock region is top-down. XXX I plan to fix that at some stage
 *
 * WARNING: Only available after early_node_map[] has been populated,
 * on some architectures, that is after all the calls to add_active_range()
 * have been done to populate it.
505
 */
Y
Yinghai Lu 已提交
506

507
phys_addr_t __weak __init memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid)
508
{
509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529
#ifdef CONFIG_ARCH_POPULATES_NODE_MAP
	/*
	 * This code originates from sparc which really wants use to walk by addresses
	 * and returns the nid. This is not very convenient for early_pfn_map[] users
	 * as the map isn't sorted yet, and it really wants to be walked by nid.
	 *
	 * For now, I implement the inefficient method below which walks the early
	 * map multiple times. Eventually we may want to use an ARCH config option
	 * to implement a completely different method for both case.
	 */
	unsigned long start_pfn, end_pfn;
	int i;

	for (i = 0; i < MAX_NUMNODES; i++) {
		get_pfn_range_for_nid(i, &start_pfn, &end_pfn);
		if (start < PFN_PHYS(start_pfn) || start >= PFN_PHYS(end_pfn))
			continue;
		*nid = i;
		return min(end, PFN_PHYS(end_pfn));
	}
#endif
530 531 532 533 534
	*nid = 0;

	return end;
}

535 536 537
static phys_addr_t __init memblock_alloc_nid_region(struct memblock_region *mp,
					       phys_addr_t size,
					       phys_addr_t align, int nid)
Y
Yinghai Lu 已提交
538
{
539
	phys_addr_t start, end;
Y
Yinghai Lu 已提交
540 541 542 543 544 545

	start = mp->base;
	end = start + mp->size;

	start = memblock_align_up(start, align);
	while (start < end) {
546
		phys_addr_t this_end;
Y
Yinghai Lu 已提交
547 548
		int this_nid;

549
		this_end = memblock_nid_range(start, end, &this_nid);
Y
Yinghai Lu 已提交
550
		if (this_nid == nid) {
551
			phys_addr_t ret = memblock_find_region(start, this_end, size, align);
552
			if (ret != MEMBLOCK_ERROR &&
553
			    memblock_add_region(&memblock.reserved, ret, size) >= 0)
Y
Yinghai Lu 已提交
554 555 556 557 558
				return ret;
		}
		start = this_end;
	}

559
	return MEMBLOCK_ERROR;
Y
Yinghai Lu 已提交
560 561
}

562
phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int nid)
Y
Yinghai Lu 已提交
563
{
564
	struct memblock_type *mem = &memblock.memory;
Y
Yinghai Lu 已提交
565 566 567 568
	int i;

	BUG_ON(0 == size);

569 570 571 572 573
	/* We align the size to limit fragmentation. Without this, a lot of
	 * small allocs quickly eat up the whole reserve array on sparc
	 */
	size = memblock_align_up(size, align);

574 575 576 577
	/* We do a bottom-up search for a region with the right
	 * nid since that's easier considering how memblock_nid_range()
	 * works
	 */
Y
Yinghai Lu 已提交
578
	for (i = 0; i < mem->cnt; i++) {
579
		phys_addr_t ret = memblock_alloc_nid_region(&mem->regions[i],
Y
Yinghai Lu 已提交
580
					       size, align, nid);
581
		if (ret != MEMBLOCK_ERROR)
Y
Yinghai Lu 已提交
582 583 584
			return ret;
	}

585 586 587 588 589 590 591 592 593
	return 0;
}

phys_addr_t __init memblock_alloc_try_nid(phys_addr_t size, phys_addr_t align, int nid)
{
	phys_addr_t res = memblock_alloc_nid(size, align, nid);

	if (res)
		return res;
594
	return memblock_alloc_base(size, align, MEMBLOCK_ALLOC_ANYWHERE);
Y
Yinghai Lu 已提交
595 596
}

597 598 599 600 601

/*
 * Remaining API functions
 */

Y
Yinghai Lu 已提交
602
/* You must call memblock_analyze() before this. */
603
phys_addr_t __init memblock_phys_mem_size(void)
Y
Yinghai Lu 已提交
604
{
605
	return memblock.memory_size;
Y
Yinghai Lu 已提交
606 607
}

608
phys_addr_t __init_memblock memblock_end_of_DRAM(void)
Y
Yinghai Lu 已提交
609 610 611
{
	int idx = memblock.memory.cnt - 1;

612
	return (memblock.memory.regions[idx].base + memblock.memory.regions[idx].size);
Y
Yinghai Lu 已提交
613 614 615
}

/* You must call memblock_analyze() after this. */
616
void __init memblock_enforce_memory_limit(phys_addr_t memory_limit)
Y
Yinghai Lu 已提交
617 618
{
	unsigned long i;
619
	phys_addr_t limit;
620
	struct memblock_region *p;
Y
Yinghai Lu 已提交
621 622 623 624 625 626 627

	if (!memory_limit)
		return;

	/* Truncate the memblock regions to satisfy the memory limit. */
	limit = memory_limit;
	for (i = 0; i < memblock.memory.cnt; i++) {
628 629
		if (limit > memblock.memory.regions[i].size) {
			limit -= memblock.memory.regions[i].size;
Y
Yinghai Lu 已提交
630 631 632
			continue;
		}

633
		memblock.memory.regions[i].size = limit;
Y
Yinghai Lu 已提交
634 635 636 637 638 639 640 641
		memblock.memory.cnt = i + 1;
		break;
	}

	memory_limit = memblock_end_of_DRAM();

	/* And truncate any reserves above the limit also. */
	for (i = 0; i < memblock.reserved.cnt; i++) {
642
		p = &memblock.reserved.regions[i];
Y
Yinghai Lu 已提交
643 644 645 646 647 648 649 650 651 652 653 654 655

		if (p->base > memory_limit)
			p->size = 0;
		else if ((p->base + p->size) > memory_limit)
			p->size = memory_limit - p->base;

		if (p->size == 0) {
			memblock_remove_region(&memblock.reserved, i);
			i--;
		}
	}
}

656
static int __init_memblock memblock_search(struct memblock_type *type, phys_addr_t addr)
657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673
{
	unsigned int left = 0, right = type->cnt;

	do {
		unsigned int mid = (right + left) / 2;

		if (addr < type->regions[mid].base)
			right = mid;
		else if (addr >= (type->regions[mid].base +
				  type->regions[mid].size))
			left = mid + 1;
		else
			return mid;
	} while (left < right);
	return -1;
}

674
int __init memblock_is_reserved(phys_addr_t addr)
Y
Yinghai Lu 已提交
675
{
676 677
	return memblock_search(&memblock.reserved, addr) != -1;
}
Y
Yinghai Lu 已提交
678

679
int __init_memblock memblock_is_memory(phys_addr_t addr)
680 681 682 683
{
	return memblock_search(&memblock.memory, addr) != -1;
}

684
int __init_memblock memblock_is_region_memory(phys_addr_t base, phys_addr_t size)
685 686 687 688 689 690 691 692
{
	int idx = memblock_search(&memblock.reserved, base);

	if (idx == -1)
		return 0;
	return memblock.reserved.regions[idx].base <= base &&
		(memblock.reserved.regions[idx].base +
		 memblock.reserved.regions[idx].size) >= (base + size);
Y
Yinghai Lu 已提交
693 694
}

695
int __init_memblock memblock_is_region_reserved(phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
696
{
697
	return memblock_overlaps_region(&memblock.reserved, base, size) >= 0;
Y
Yinghai Lu 已提交
698 699
}

700

701
void __init_memblock memblock_set_current_limit(phys_addr_t limit)
702 703 704 705
{
	memblock.current_limit = limit;
}

706
static void __init_memblock memblock_dump(struct memblock_type *region, char *name)
707 708 709 710 711 712 713 714 715 716
{
	unsigned long long base, size;
	int i;

	pr_info(" %s.cnt  = 0x%lx\n", name, region->cnt);

	for (i = 0; i < region->cnt; i++) {
		base = region->regions[i].base;
		size = region->regions[i].size;

717
		pr_info(" %s[%#x]\t[%#016llx-%#016llx], %#llx bytes\n",
718 719 720 721
		    name, i, base, base + size - 1, size);
	}
}

722
void __init_memblock memblock_dump_all(void)
723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747
{
	if (!memblock_debug)
		return;

	pr_info("MEMBLOCK configuration:\n");
	pr_info(" memory size = 0x%llx\n", (unsigned long long)memblock.memory_size);

	memblock_dump(&memblock.memory, "memory");
	memblock_dump(&memblock.reserved, "reserved");
}

void __init memblock_analyze(void)
{
	int i;

	/* Check marker in the unused last array entry */
	WARN_ON(memblock_memory_init_regions[INIT_MEMBLOCK_REGIONS].base
		!= (phys_addr_t)RED_INACTIVE);
	WARN_ON(memblock_reserved_init_regions[INIT_MEMBLOCK_REGIONS].base
		!= (phys_addr_t)RED_INACTIVE);

	memblock.memory_size = 0;

	for (i = 0; i < memblock.memory.cnt; i++)
		memblock.memory_size += memblock.memory.regions[i].size;
748 749 750

	/* We allow resizing from there */
	memblock_can_resize = 1;
751 752
}

753 754
void __init memblock_init(void)
{
755 756 757 758 759 760
	static int init_done __initdata = 0;

	if (init_done)
		return;
	init_done = 1;

761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785
	/* Hookup the initial arrays */
	memblock.memory.regions	= memblock_memory_init_regions;
	memblock.memory.max		= INIT_MEMBLOCK_REGIONS;
	memblock.reserved.regions	= memblock_reserved_init_regions;
	memblock.reserved.max	= INIT_MEMBLOCK_REGIONS;

	/* Write a marker in the unused last array entry */
	memblock.memory.regions[INIT_MEMBLOCK_REGIONS].base = (phys_addr_t)RED_INACTIVE;
	memblock.reserved.regions[INIT_MEMBLOCK_REGIONS].base = (phys_addr_t)RED_INACTIVE;

	/* Create a dummy zero size MEMBLOCK which will get coalesced away later.
	 * This simplifies the memblock_add() code below...
	 */
	memblock.memory.regions[0].base = 0;
	memblock.memory.regions[0].size = 0;
	memblock.memory.cnt = 1;

	/* Ditto. */
	memblock.reserved.regions[0].base = 0;
	memblock.reserved.regions[0].size = 0;
	memblock.reserved.cnt = 1;

	memblock.current_limit = MEMBLOCK_ALLOC_ANYWHERE;
}

786 787 788 789 790 791 792 793
static int __init early_memblock(char *p)
{
	if (p && strstr(p, "debug"))
		memblock_debug = 1;
	return 0;
}
early_param("memblock", early_memblock);

794
#if defined(CONFIG_DEBUG_FS) && !defined(ARCH_DISCARD_MEMBLOCK)
795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842

static int memblock_debug_show(struct seq_file *m, void *private)
{
	struct memblock_type *type = m->private;
	struct memblock_region *reg;
	int i;

	for (i = 0; i < type->cnt; i++) {
		reg = &type->regions[i];
		seq_printf(m, "%4d: ", i);
		if (sizeof(phys_addr_t) == 4)
			seq_printf(m, "0x%08lx..0x%08lx\n",
				   (unsigned long)reg->base,
				   (unsigned long)(reg->base + reg->size - 1));
		else
			seq_printf(m, "0x%016llx..0x%016llx\n",
				   (unsigned long long)reg->base,
				   (unsigned long long)(reg->base + reg->size - 1));

	}
	return 0;
}

static int memblock_debug_open(struct inode *inode, struct file *file)
{
	return single_open(file, memblock_debug_show, inode->i_private);
}

static const struct file_operations memblock_debug_fops = {
	.open = memblock_debug_open,
	.read = seq_read,
	.llseek = seq_lseek,
	.release = single_release,
};

static int __init memblock_init_debugfs(void)
{
	struct dentry *root = debugfs_create_dir("memblock", NULL);
	if (!root)
		return -ENXIO;
	debugfs_create_file("memory", S_IRUGO, root, &memblock.memory, &memblock_debug_fops);
	debugfs_create_file("reserved", S_IRUGO, root, &memblock.reserved, &memblock_debug_fops);

	return 0;
}
__initcall(memblock_init_debugfs);

#endif /* CONFIG_DEBUG_FS */