memblock.c 22.4 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
 */
44
static unsigned long __init_memblock memblock_addrs_overlap(phys_addr_t base1, phys_addr_t size1,
45
				       phys_addr_t base2, phys_addr_t size2)
Y
Yinghai Lu 已提交
46 47 48 49
{
	return ((base1 < (base2 + size2)) && (base2 < (base1 + size1)));
}

50
long __init_memblock memblock_overlaps_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size)
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68
{
	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.
 */

69
static phys_addr_t __init_memblock memblock_find_region(phys_addr_t start, phys_addr_t end,
70 71 72 73 74
					  phys_addr_t size, phys_addr_t align)
{
	phys_addr_t base, res_base;
	long j;

75 76 77 78
	/* In case, huge size is requested */
	if (end < size)
		return MEMBLOCK_ERROR;

79
	base = round_down(end - size, align);
80

81 82 83 84 85 86
	/* Prevent allocations returning 0 as it's also used to
	 * indicate an allocation failure
	 */
	if (start == 0)
		start = PAGE_SIZE;

87 88 89 90 91 92 93
	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;
94
		base = round_down(res_base - size, align);
95 96 97 98 99
	}

	return MEMBLOCK_ERROR;
}

100 101
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)
102 103 104 105 106 107
{
	long i;

	BUG_ON(0 == size);

	/* Pump up max_addr */
108 109
	if (end == MEMBLOCK_ALLOC_ACCESSIBLE)
		end = memblock.current_limit;
110 111 112 113 114 115 116 117

	/* 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;
118
		phys_addr_t bottom, top, found;
119 120 121

		if (memblocksize < size)
			continue;
122 123 124 125 126 127 128 129 130
		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;
131 132 133 134
	}
	return MEMBLOCK_ERROR;
}

135 136 137 138 139 140 141 142
/*
 * 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);
}

143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166
/*
 * 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);
}

167
static void __init_memblock memblock_remove_region(struct memblock_type *type, unsigned long r)
Y
Yinghai Lu 已提交
168 169 170
{
	unsigned long i;

171 172 173
	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
174
	}
175
	type->cnt--;
Y
Yinghai Lu 已提交
176

177 178 179 180 181 182
	/* Special case for empty arrays */
	if (type->cnt == 0) {
		type->cnt = 1;
		type->regions[0].base = 0;
		type->regions[0].size = 0;
	}
Y
Yinghai Lu 已提交
183 184
}

185 186 187
/* Defined below but needed now */
static long memblock_add_region(struct memblock_type *type, phys_addr_t base, phys_addr_t size);

188
static int __init_memblock memblock_double_array(struct memblock_type *type)
189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218
{
	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
219
		addr = memblock_find_base(new_size, sizeof(phys_addr_t), 0, MEMBLOCK_ALLOC_ACCESSIBLE);
220 221 222 223 224 225 226
	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);

227 228 229
	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);

230 231 232 233 234 235 236 237 238 239 240 241 242 243 244
	/* 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 ! */
245
	BUG_ON(memblock_add_region(&memblock.reserved, addr, new_size));
246 247 248 249 250 251 252 253 254 255 256 257 258

	/* 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;
}

259
extern int __init_memblock __weak memblock_memory_can_coalesce(phys_addr_t addr1, phys_addr_t size1,
260 261 262 263 264
					  phys_addr_t addr2, phys_addr_t size2)
{
	return 1;
}

265 266
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
267
{
268 269
	phys_addr_t end = base + size;
	int i, slot = -1;
Y
Yinghai Lu 已提交
270

271
	/* First try and coalesce this MEMBLOCK with others */
272
	for (i = 0; i < type->cnt; i++) {
273 274 275 276 277 278
		struct memblock_region *rgn = &type->regions[i];
		phys_addr_t rend = rgn->base + rgn->size;

		/* Exit if there's no possible hits */
		if (rgn->base > end || rgn->size == 0)
			break;
Y
Yinghai Lu 已提交
279

280 281 282 283
		/* Check if we are fully enclosed within an existing
		 * block
		 */
		if (rgn->base <= base && rend >= end)
Y
Yinghai Lu 已提交
284 285
			return 0;

286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344
		/* Check if we overlap or are adjacent with the bottom
		 * of a block.
		 */
		if (base < rgn->base && end >= rgn->base) {
			/* If we can't coalesce, create a new block */
			if (!memblock_memory_can_coalesce(base, size,
							  rgn->base,
							  rgn->size)) {
				/* Overlap & can't coalesce are mutually
				 * exclusive, if you do that, be prepared
				 * for trouble
				 */
				WARN_ON(end != rgn->base);
				goto new_block;
			}
			/* We extend the bottom of the block down to our
			 * base
			 */
			rgn->base = base;
			rgn->size = rend - base;

			/* Return if we have nothing else to allocate
			 * (fully coalesced)
			 */
			if (rend >= end)
				return 0;

			/* We continue processing from the end of the
			 * coalesced block.
			 */
			base = rend;
			size = end - base;
		}

		/* Now check if we overlap or are adjacent with the
		 * top of a block
		 */
		if (base <= rend && end >= rend) {
			/* If we can't coalesce, create a new block */
			if (!memblock_memory_can_coalesce(rgn->base,
							  rgn->size,
							  base, size)) {
				/* Overlap & can't coalesce are mutually
				 * exclusive, if you do that, be prepared
				 * for trouble
				 */
				WARN_ON(rend != base);
				goto new_block;
			}
			/* We adjust our base down to enclose the
			 * original block and destroy it. It will be
			 * part of our new allocation. Since we've
			 * freed an entry, we know we won't fail
			 * to allocate one later, so we won't risk
			 * losing the original block allocation.
			 */
			size += (base - rgn->base);
			base = rgn->base;
			memblock_remove_region(type, i--);
Y
Yinghai Lu 已提交
345 346 347
		}
	}

348 349
	/* If the array is empty, special case, replace the fake
	 * filler region and return
350
	 */
351 352 353 354
	if ((type->cnt == 1) && (type->regions[0].size == 0)) {
		type->regions[0].base = base;
		type->regions[0].size = size;
		return 0;
Y
Yinghai Lu 已提交
355 356
	}

357
 new_block:
358 359 360 361
	/* 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
362 363 364
		return -1;

	/* Couldn't coalesce the MEMBLOCK, so add it to the sorted table. */
365 366 367 368
	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
369
		} else {
370 371
			type->regions[i+1].base = base;
			type->regions[i+1].size = size;
372
			slot = i + 1;
Y
Yinghai Lu 已提交
373 374 375
			break;
		}
	}
376 377 378
	if (base < type->regions[0].base) {
		type->regions[0].base = base;
		type->regions[0].size = size;
379
		slot = 0;
Y
Yinghai Lu 已提交
380
	}
381
	type->cnt++;
Y
Yinghai Lu 已提交
382

383 384 385 386
	/* 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)) {
387 388
		BUG_ON(slot < 0);
		memblock_remove_region(type, slot);
389 390 391
		return -1;
	}

Y
Yinghai Lu 已提交
392 393 394
	return 0;
}

395
long __init_memblock memblock_add(phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
396
{
397
	return memblock_add_region(&memblock.memory, base, size);
Y
Yinghai Lu 已提交
398 399 400

}

401 402
static long __init_memblock __memblock_remove(struct memblock_type *type,
					      phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
403
{
404
	phys_addr_t end = base + size;
Y
Yinghai Lu 已提交
405 406
	int i;

407 408 409 410
	/* Walk through the array for collisions */
	for (i = 0; i < type->cnt; i++) {
		struct memblock_region *rgn = &type->regions[i];
		phys_addr_t rend = rgn->base + rgn->size;
Y
Yinghai Lu 已提交
411

412 413
		/* Nothing more to do, exit */
		if (rgn->base > end || rgn->size == 0)
Y
Yinghai Lu 已提交
414 415
			break;

416 417 418 419 420
		/* If we fully enclose the block, drop it */
		if (base <= rgn->base && end >= rend) {
			memblock_remove_region(type, i--);
			continue;
		}
Y
Yinghai Lu 已提交
421

422 423 424 425 426 427 428 429 430 431 432 433 434 435
		/* If we are fully enclosed within a block
		 * then we need to split it and we are done
		 */
		if (base > rgn->base && end < rend) {
			rgn->size = base - rgn->base;
			if (!memblock_add_region(type, end, rend - end))
				return 0;
			/* Failure to split is bad, we at least
			 * restore the block before erroring
			 */
			rgn->size = rend - rgn->base;
			WARN_ON(1);
			return -1;
		}
Y
Yinghai Lu 已提交
436

437 438 439 440 441 442
		/* Check if we need to trim the bottom of a block */
		if (rgn->base < end && rend > end) {
			rgn->size -= end - rgn->base;
			rgn->base = end;
			break;
		}
Y
Yinghai Lu 已提交
443

444 445 446
		/* And check if we need to trim the top of a block */
		if (base < rend)
			rgn->size -= rend - base;
Y
Yinghai Lu 已提交
447

448 449
	}
	return 0;
Y
Yinghai Lu 已提交
450 451
}

452
long __init_memblock memblock_remove(phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
453 454 455 456
{
	return __memblock_remove(&memblock.memory, base, size);
}

457
long __init_memblock memblock_free(phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
458 459 460 461
{
	return __memblock_remove(&memblock.reserved, base, size);
}

462
long __init_memblock memblock_reserve(phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
463
{
464
	struct memblock_type *_rgn = &memblock.reserved;
Y
Yinghai Lu 已提交
465 466 467 468 469 470

	BUG_ON(0 == size);

	return memblock_add_region(_rgn, base, size);
}

471
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
472
{
473
	phys_addr_t found;
Y
Yinghai Lu 已提交
474

475 476 477
	/* We align the size to limit fragmentation. Without this, a lot of
	 * small allocs quickly eat up the whole reserve array on sparc
	 */
478
	size = round_up(size, align);
Y
Yinghai Lu 已提交
479

480
	found = memblock_find_base(size, align, 0, max_addr);
481
	if (found != MEMBLOCK_ERROR &&
482
	    !memblock_add_region(&memblock.reserved, found, size))
483
		return found;
Y
Yinghai Lu 已提交
484

485
	return 0;
Y
Yinghai Lu 已提交
486 487
}

488
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
489
{
490 491 492 493 494 495 496 497 498
	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
499 500
}

501
phys_addr_t __init memblock_alloc(phys_addr_t size, phys_addr_t align)
Y
Yinghai Lu 已提交
502
{
503 504
	return memblock_alloc_base(size, align, MEMBLOCK_ALLOC_ACCESSIBLE);
}
Y
Yinghai Lu 已提交
505 506


507 508 509
/*
 * Additional node-local allocators. Search for node memory is bottom up
 * and walks memblock regions within that node bottom-up as well, but allocation
510 511 512 513 514
 * 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.
515
 */
Y
Yinghai Lu 已提交
516

517
phys_addr_t __weak __init memblock_nid_range(phys_addr_t start, phys_addr_t end, int *nid)
518
{
519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539
#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
540 541 542 543 544
	*nid = 0;

	return end;
}

545 546 547
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
548
{
549
	phys_addr_t start, end;
Y
Yinghai Lu 已提交
550 551 552 553

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

554
	start = round_up(start, align);
Y
Yinghai Lu 已提交
555
	while (start < end) {
556
		phys_addr_t this_end;
Y
Yinghai Lu 已提交
557 558
		int this_nid;

559
		this_end = memblock_nid_range(start, end, &this_nid);
Y
Yinghai Lu 已提交
560
		if (this_nid == nid) {
561
			phys_addr_t ret = memblock_find_region(start, this_end, size, align);
562
			if (ret != MEMBLOCK_ERROR &&
563
			    !memblock_add_region(&memblock.reserved, ret, size))
Y
Yinghai Lu 已提交
564 565 566 567 568
				return ret;
		}
		start = this_end;
	}

569
	return MEMBLOCK_ERROR;
Y
Yinghai Lu 已提交
570 571
}

572
phys_addr_t __init memblock_alloc_nid(phys_addr_t size, phys_addr_t align, int nid)
Y
Yinghai Lu 已提交
573
{
574
	struct memblock_type *mem = &memblock.memory;
Y
Yinghai Lu 已提交
575 576 577 578
	int i;

	BUG_ON(0 == size);

579 580 581
	/* We align the size to limit fragmentation. Without this, a lot of
	 * small allocs quickly eat up the whole reserve array on sparc
	 */
582
	size = round_up(size, align);
583

584 585 586 587
	/* 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
588
	for (i = 0; i < mem->cnt; i++) {
589
		phys_addr_t ret = memblock_alloc_nid_region(&mem->regions[i],
Y
Yinghai Lu 已提交
590
					       size, align, nid);
591
		if (ret != MEMBLOCK_ERROR)
Y
Yinghai Lu 已提交
592 593 594
			return ret;
	}

595 596 597 598 599 600 601 602 603
	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;
604
	return memblock_alloc_base(size, align, MEMBLOCK_ALLOC_ACCESSIBLE);
Y
Yinghai Lu 已提交
605 606
}

607 608 609 610 611

/*
 * Remaining API functions
 */

Y
Yinghai Lu 已提交
612
/* You must call memblock_analyze() before this. */
613
phys_addr_t __init memblock_phys_mem_size(void)
Y
Yinghai Lu 已提交
614
{
615
	return memblock.memory_size;
Y
Yinghai Lu 已提交
616 617
}

618
phys_addr_t __init_memblock memblock_end_of_DRAM(void)
Y
Yinghai Lu 已提交
619 620 621
{
	int idx = memblock.memory.cnt - 1;

622
	return (memblock.memory.regions[idx].base + memblock.memory.regions[idx].size);
Y
Yinghai Lu 已提交
623 624 625
}

/* You must call memblock_analyze() after this. */
626
void __init memblock_enforce_memory_limit(phys_addr_t memory_limit)
Y
Yinghai Lu 已提交
627 628
{
	unsigned long i;
629
	phys_addr_t limit;
630
	struct memblock_region *p;
Y
Yinghai Lu 已提交
631 632 633 634 635 636 637

	if (!memory_limit)
		return;

	/* Truncate the memblock regions to satisfy the memory limit. */
	limit = memory_limit;
	for (i = 0; i < memblock.memory.cnt; i++) {
638 639
		if (limit > memblock.memory.regions[i].size) {
			limit -= memblock.memory.regions[i].size;
Y
Yinghai Lu 已提交
640 641 642
			continue;
		}

643
		memblock.memory.regions[i].size = limit;
Y
Yinghai Lu 已提交
644 645 646 647 648 649 650 651
		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++) {
652
		p = &memblock.reserved.regions[i];
Y
Yinghai Lu 已提交
653 654 655 656 657 658 659 660 661 662 663 664 665

		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--;
		}
	}
}

666
static int __init_memblock memblock_search(struct memblock_type *type, phys_addr_t addr)
667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683
{
	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;
}

684
int __init memblock_is_reserved(phys_addr_t addr)
Y
Yinghai Lu 已提交
685
{
686 687
	return memblock_search(&memblock.reserved, addr) != -1;
}
Y
Yinghai Lu 已提交
688

689
int __init_memblock memblock_is_memory(phys_addr_t addr)
690 691 692 693
{
	return memblock_search(&memblock.memory, addr) != -1;
}

694
int __init_memblock memblock_is_region_memory(phys_addr_t base, phys_addr_t size)
695
{
696
	int idx = memblock_search(&memblock.memory, base);
697 698 699

	if (idx == -1)
		return 0;
700 701 702
	return memblock.memory.regions[idx].base <= base &&
		(memblock.memory.regions[idx].base +
		 memblock.memory.regions[idx].size) >= (base + size);
Y
Yinghai Lu 已提交
703 704
}

705
int __init_memblock memblock_is_region_reserved(phys_addr_t base, phys_addr_t size)
Y
Yinghai Lu 已提交
706
{
707
	return memblock_overlaps_region(&memblock.reserved, base, size) >= 0;
Y
Yinghai Lu 已提交
708 709
}

710

711
void __init_memblock memblock_set_current_limit(phys_addr_t limit)
712 713 714 715
{
	memblock.current_limit = limit;
}

716
static void __init_memblock memblock_dump(struct memblock_type *region, char *name)
717 718 719 720 721 722 723 724 725 726
{
	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;

727
		pr_info(" %s[%#x]\t[%#016llx-%#016llx], %#llx bytes\n",
728 729 730 731
		    name, i, base, base + size - 1, size);
	}
}

732
void __init_memblock memblock_dump_all(void)
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
{
	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;
758 759 760

	/* We allow resizing from there */
	memblock_can_resize = 1;
761 762
}

763 764
void __init memblock_init(void)
{
765 766 767 768 769 770
	static int init_done __initdata = 0;

	if (init_done)
		return;
	init_done = 1;

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
	/* 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;
}

796 797 798 799 800 801 802 803
static int __init early_memblock(char *p)
{
	if (p && strstr(p, "debug"))
		memblock_debug = 1;
	return 0;
}
early_param("memblock", early_memblock);

804
#if defined(CONFIG_DEBUG_FS) && !defined(ARCH_DISCARD_MEMBLOCK)
805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852

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 */