hugetlbpage.c 24.1 KB
Newer Older
L
Linus Torvalds 已提交
1
/*
B
Becky Bruce 已提交
2
 * PPC Huge TLB Page Support for Kernel.
L
Linus Torvalds 已提交
3 4
 *
 * Copyright (C) 2003 David Gibson, IBM Corporation.
B
Becky Bruce 已提交
5
 * Copyright (C) 2011 Becky Bruce, Freescale Semiconductor
L
Linus Torvalds 已提交
6 7 8 9 10 11
 *
 * Based on the IA-32 version:
 * Copyright (C) 2002, Rohit Seth <rohit.seth@intel.com>
 */

#include <linux/mm.h>
12
#include <linux/io.h>
13
#include <linux/slab.h>
L
Linus Torvalds 已提交
14
#include <linux/hugetlb.h>
15
#include <linux/export.h>
B
Becky Bruce 已提交
16 17 18
#include <linux/of_fdt.h>
#include <linux/memblock.h>
#include <linux/bootmem.h>
19
#include <linux/moduleparam.h>
20
#include <asm/pgtable.h>
L
Linus Torvalds 已提交
21 22
#include <asm/pgalloc.h>
#include <asm/tlb.h>
B
Becky Bruce 已提交
23
#include <asm/setup.h>
L
Linus Torvalds 已提交
24

25 26 27
#define PAGE_SHIFT_64K	16
#define PAGE_SHIFT_16M	24
#define PAGE_SHIFT_16G	34
28

B
Becky Bruce 已提交
29
unsigned int HPAGE_SHIFT;
30

B
Becky Bruce 已提交
31 32
/*
 * Tracks gpages after the device tree is scanned and before the
33 34 35 36
 * huge_boot_pages list is ready.  On non-Freescale implementations, this is
 * just used to track 16G pages and so is a single array.  FSL-based
 * implementations may have more than one gpage size, so we need multiple
 * arrays
B
Becky Bruce 已提交
37
 */
38
#ifdef CONFIG_PPC_FSL_BOOK3E
B
Becky Bruce 已提交
39 40 41 42 43 44
#define MAX_NUMBER_GPAGES	128
struct psize_gpages {
	u64 gpage_list[MAX_NUMBER_GPAGES];
	unsigned int nr_gpages;
};
static struct psize_gpages gpage_freearray[MMU_PAGE_COUNT];
45 46 47 48
#else
#define MAX_NUMBER_GPAGES	1024
static u64 gpage_freearray[MAX_NUMBER_GPAGES];
static unsigned nr_gpages;
B
Becky Bruce 已提交
49
#endif
50

51 52
#define hugepd_none(hpd)	((hpd).pd == 0)

53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109
#ifdef CONFIG_PPC_BOOK3S_64
/*
 * At this point we do the placement change only for BOOK3S 64. This would
 * possibly work on other subarchs.
 */

/*
 * We have PGD_INDEX_SIZ = 12 and PTE_INDEX_SIZE = 8, so that we can have
 * 16GB hugepage pte in PGD and 16MB hugepage pte at PMD;
 */
int pmd_huge(pmd_t pmd)
{
	/*
	 * leaf pte for huge page, bottom two bits != 00
	 */
	return ((pmd_val(pmd) & 0x3) != 0x0);
}

int pud_huge(pud_t pud)
{
	/*
	 * leaf pte for huge page, bottom two bits != 00
	 */
	return ((pud_val(pud) & 0x3) != 0x0);
}

int pgd_huge(pgd_t pgd)
{
	/*
	 * leaf pte for huge page, bottom two bits != 00
	 */
	return ((pgd_val(pgd) & 0x3) != 0x0);
}
#else
int pmd_huge(pmd_t pmd)
{
	return 0;
}

int pud_huge(pud_t pud)
{
	return 0;
}

int pgd_huge(pgd_t pgd)
{
	return 0;
}
#endif

/*
 * We have 4 cases for pgds and pmds:
 * (1) invalid (all zeroes)
 * (2) pointer to next table, as normal; bottom 6 bits == 0
 * (3) leaf pte for huge page, bottom two bits != 00
 * (4) hugepd pointer, bottom two bits == 00, next 4 bits indicate size of table
 */
110 111 112 113 114
pte_t *find_linux_pte_or_hugepte(pgd_t *pgdir, unsigned long ea, unsigned *shift)
{
	pgd_t *pg;
	pud_t *pu;
	pmd_t *pm;
115
	pte_t *ret_pte;
116 117 118 119 120 121 122
	hugepd_t *hpdp = NULL;
	unsigned pdshift = PGDIR_SHIFT;

	if (shift)
		*shift = 0;

	pg = pgdir + pgd_index(ea);
123 124 125 126 127

	if (pgd_huge(*pg)) {
		ret_pte = (pte_t *) pg;
		goto out;
	} else if (is_hugepd(pg))
128
		hpdp = (hugepd_t *)pg;
129
	else if (!pgd_none(*pg)) {
130 131
		pdshift = PUD_SHIFT;
		pu = pud_offset(pg, ea);
132 133 134 135 136

		if (pud_huge(*pu)) {
			ret_pte = (pte_t *) pu;
			goto out;
		} else if (is_hugepd(pu))
137 138 139 140
			hpdp = (hugepd_t *)pu;
		else if (!pud_none(*pu)) {
			pdshift = PMD_SHIFT;
			pm = pmd_offset(pu, ea);
141 142 143 144 145

			if (pmd_huge(*pm)) {
				ret_pte = (pte_t *) pm;
				goto out;
			} else if (is_hugepd(pm))
146
				hpdp = (hugepd_t *)pm;
147
			else if (!pmd_none(*pm))
B
Becky Bruce 已提交
148
				return pte_offset_kernel(pm, ea);
149 150 151 152 153
		}
	}
	if (!hpdp)
		return NULL;

154 155 156
	ret_pte = hugepte_offset(hpdp, ea, pdshift);
	pdshift = hugepd_shift(*hpdp);
out:
157
	if (shift)
158 159
		*shift = pdshift;
	return ret_pte;
160
}
161
EXPORT_SYMBOL_GPL(find_linux_pte_or_hugepte);
162 163 164 165 166 167

pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
{
	return find_linux_pte_or_hugepte(mm->pgd, addr, NULL);
}

168
static int __hugepte_alloc(struct mm_struct *mm, hugepd_t *hpdp,
169
			   unsigned long address, unsigned pdshift, unsigned pshift)
170
{
B
Becky Bruce 已提交
171 172 173
	struct kmem_cache *cachep;
	pte_t *new;

174
#ifdef CONFIG_PPC_FSL_BOOK3E
B
Becky Bruce 已提交
175 176 177
	int i;
	int num_hugepd = 1 << (pshift - pdshift);
	cachep = hugepte_cache;
178 179
#else
	cachep = PGT_CACHE(pdshift - pshift);
B
Becky Bruce 已提交
180 181 182
#endif

	new = kmem_cache_zalloc(cachep, GFP_KERNEL|__GFP_REPEAT);
183

184 185 186
	BUG_ON(pshift > HUGEPD_SHIFT_MASK);
	BUG_ON((unsigned long)new & HUGEPD_SHIFT_MASK);

187 188 189 190
	if (! new)
		return -ENOMEM;

	spin_lock(&mm->page_table_lock);
191
#ifdef CONFIG_PPC_FSL_BOOK3E
B
Becky Bruce 已提交
192 193 194 195 196 197 198 199 200 201
	/*
	 * We have multiple higher-level entries that point to the same
	 * actual pte location.  Fill in each as we go and backtrack on error.
	 * We need all of these so the DTLB pgtable walk code can find the
	 * right higher-level entry without knowing if it's a hugepage or not.
	 */
	for (i = 0; i < num_hugepd; i++, hpdp++) {
		if (unlikely(!hugepd_none(*hpdp)))
			break;
		else
202
			/* We use the old format for PPC_FSL_BOOK3E */
B
Becky Bruce 已提交
203 204 205 206 207 208 209 210
			hpdp->pd = ((unsigned long)new & ~PD_HUGE) | pshift;
	}
	/* If we bailed from the for loop early, an error occurred, clean up */
	if (i < num_hugepd) {
		for (i = i - 1 ; i >= 0; i--, hpdp--)
			hpdp->pd = 0;
		kmem_cache_free(cachep, new);
	}
211 212 213
#else
	if (!hugepd_none(*hpdp))
		kmem_cache_free(cachep, new);
214 215 216 217 218
	else {
#ifdef CONFIG_PPC_BOOK3S_64
		hpdp->pd = (unsigned long)new |
			    (shift_to_mmu_psize(pshift) << 2);
#else
219
		hpdp->pd = ((unsigned long)new & ~PD_HUGE) | pshift;
220 221
#endif
	}
B
Becky Bruce 已提交
222
#endif
223 224 225 226
	spin_unlock(&mm->page_table_lock);
	return 0;
}

227 228 229 230 231 232 233 234 235 236 237 238
/*
 * These macros define how to determine which level of the page table holds
 * the hpdp.
 */
#ifdef CONFIG_PPC_FSL_BOOK3E
#define HUGEPD_PGD_SHIFT PGDIR_SHIFT
#define HUGEPD_PUD_SHIFT PUD_SHIFT
#else
#define HUGEPD_PGD_SHIFT PUD_SHIFT
#define HUGEPD_PUD_SHIFT PMD_SHIFT
#endif

239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293
#ifdef CONFIG_PPC_BOOK3S_64
/*
 * At this point we do the placement change only for BOOK3S 64. This would
 * possibly work on other subarchs.
 */
pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr, unsigned long sz)
{
	pgd_t *pg;
	pud_t *pu;
	pmd_t *pm;
	hugepd_t *hpdp = NULL;
	unsigned pshift = __ffs(sz);
	unsigned pdshift = PGDIR_SHIFT;

	addr &= ~(sz-1);
	pg = pgd_offset(mm, addr);

	if (pshift == PGDIR_SHIFT)
		/* 16GB huge page */
		return (pte_t *) pg;
	else if (pshift > PUD_SHIFT)
		/*
		 * We need to use hugepd table
		 */
		hpdp = (hugepd_t *)pg;
	else {
		pdshift = PUD_SHIFT;
		pu = pud_alloc(mm, pg, addr);
		if (pshift == PUD_SHIFT)
			return (pte_t *)pu;
		else if (pshift > PMD_SHIFT)
			hpdp = (hugepd_t *)pu;
		else {
			pdshift = PMD_SHIFT;
			pm = pmd_alloc(mm, pu, addr);
			if (pshift == PMD_SHIFT)
				/* 16MB hugepage */
				return (pte_t *)pm;
			else
				hpdp = (hugepd_t *)pm;
		}
	}
	if (!hpdp)
		return NULL;

	BUG_ON(!hugepd_none(*hpdp) && !hugepd_ok(*hpdp));

	if (hugepd_none(*hpdp) && __hugepte_alloc(mm, hpdp, addr, pdshift, pshift))
		return NULL;

	return hugepte_offset(hpdp, addr, pdshift);
}

#else

294
pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr, unsigned long sz)
295
{
296 297 298 299 300 301 302 303 304 305
	pgd_t *pg;
	pud_t *pu;
	pmd_t *pm;
	hugepd_t *hpdp = NULL;
	unsigned pshift = __ffs(sz);
	unsigned pdshift = PGDIR_SHIFT;

	addr &= ~(sz-1);

	pg = pgd_offset(mm, addr);
306 307

	if (pshift >= HUGEPD_PGD_SHIFT) {
308 309 310 311
		hpdp = (hugepd_t *)pg;
	} else {
		pdshift = PUD_SHIFT;
		pu = pud_alloc(mm, pg, addr);
312
		if (pshift >= HUGEPD_PUD_SHIFT) {
313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329
			hpdp = (hugepd_t *)pu;
		} else {
			pdshift = PMD_SHIFT;
			pm = pmd_alloc(mm, pu, addr);
			hpdp = (hugepd_t *)pm;
		}
	}

	if (!hpdp)
		return NULL;

	BUG_ON(!hugepd_none(*hpdp) && !hugepd_ok(*hpdp));

	if (hugepd_none(*hpdp) && __hugepte_alloc(mm, hpdp, addr, pdshift, pshift))
		return NULL;

	return hugepte_offset(hpdp, addr, pdshift);
330
}
331
#endif
332

333
#ifdef CONFIG_PPC_FSL_BOOK3E
334 335 336
/* Build list of addresses of gigantic pages.  This function is used in early
 * boot before the buddy or bootmem allocator is setup.
 */
B
Becky Bruce 已提交
337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390
void add_gpage(u64 addr, u64 page_size, unsigned long number_of_pages)
{
	unsigned int idx = shift_to_mmu_psize(__ffs(page_size));
	int i;

	if (addr == 0)
		return;

	gpage_freearray[idx].nr_gpages = number_of_pages;

	for (i = 0; i < number_of_pages; i++) {
		gpage_freearray[idx].gpage_list[i] = addr;
		addr += page_size;
	}
}

/*
 * Moves the gigantic page addresses from the temporary list to the
 * huge_boot_pages list.
 */
int alloc_bootmem_huge_page(struct hstate *hstate)
{
	struct huge_bootmem_page *m;
	int idx = shift_to_mmu_psize(hstate->order + PAGE_SHIFT);
	int nr_gpages = gpage_freearray[idx].nr_gpages;

	if (nr_gpages == 0)
		return 0;

#ifdef CONFIG_HIGHMEM
	/*
	 * If gpages can be in highmem we can't use the trick of storing the
	 * data structure in the page; allocate space for this
	 */
	m = alloc_bootmem(sizeof(struct huge_bootmem_page));
	m->phys = gpage_freearray[idx].gpage_list[--nr_gpages];
#else
	m = phys_to_virt(gpage_freearray[idx].gpage_list[--nr_gpages]);
#endif

	list_add(&m->list, &huge_boot_pages);
	gpage_freearray[idx].nr_gpages = nr_gpages;
	gpage_freearray[idx].gpage_list[nr_gpages] = 0;
	m->hstate = hstate;

	return 1;
}
/*
 * Scan the command line hugepagesz= options for gigantic pages; store those in
 * a list that we use to allocate the memory once all options are parsed.
 */

unsigned long gpage_npages[MMU_PAGE_COUNT];

391 392
static int __init do_gpage_early_setup(char *param, char *val,
				       const char *unused)
B
Becky Bruce 已提交
393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432
{
	static phys_addr_t size;
	unsigned long npages;

	/*
	 * The hugepagesz and hugepages cmdline options are interleaved.  We
	 * use the size variable to keep track of whether or not this was done
	 * properly and skip over instances where it is incorrect.  Other
	 * command-line parsing code will issue warnings, so we don't need to.
	 *
	 */
	if ((strcmp(param, "default_hugepagesz") == 0) ||
	    (strcmp(param, "hugepagesz") == 0)) {
		size = memparse(val, NULL);
	} else if (strcmp(param, "hugepages") == 0) {
		if (size != 0) {
			if (sscanf(val, "%lu", &npages) <= 0)
				npages = 0;
			gpage_npages[shift_to_mmu_psize(__ffs(size))] = npages;
			size = 0;
		}
	}
	return 0;
}


/*
 * This function allocates physical space for pages that are larger than the
 * buddy allocator can handle.  We want to allocate these in highmem because
 * the amount of lowmem is limited.  This means that this function MUST be
 * called before lowmem_end_addr is set up in MMU_init() in order for the lmb
 * allocate to grab highmem.
 */
void __init reserve_hugetlb_gpages(void)
{
	static __initdata char cmdline[COMMAND_LINE_SIZE];
	phys_addr_t size, base;
	int i;

	strlcpy(cmdline, boot_command_line, COMMAND_LINE_SIZE);
433 434
	parse_args("hugetlb gpages", cmdline, NULL, 0, 0, 0,
			&do_gpage_early_setup);
B
Becky Bruce 已提交
435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454

	/*
	 * Walk gpage list in reverse, allocating larger page sizes first.
	 * Skip over unsupported sizes, or sizes that have 0 gpages allocated.
	 * When we reach the point in the list where pages are no longer
	 * considered gpages, we're done.
	 */
	for (i = MMU_PAGE_COUNT-1; i >= 0; i--) {
		if (mmu_psize_defs[i].shift == 0 || gpage_npages[i] == 0)
			continue;
		else if (mmu_psize_to_shift(i) < (MAX_ORDER + PAGE_SHIFT))
			break;

		size = (phys_addr_t)(1ULL << mmu_psize_to_shift(i));
		base = memblock_alloc_base(size * gpage_npages[i], size,
					   MEMBLOCK_ALLOC_ANYWHERE);
		add_gpage(base, size, gpage_npages[i]);
	}
}

455
#else /* !PPC_FSL_BOOK3E */
B
Becky Bruce 已提交
456 457 458 459 460

/* Build list of addresses of gigantic pages.  This function is used in early
 * boot before the buddy or bootmem allocator is setup.
 */
void add_gpage(u64 addr, u64 page_size, unsigned long number_of_pages)
461 462 463 464 465 466 467 468 469 470 471
{
	if (!addr)
		return;
	while (number_of_pages > 0) {
		gpage_freearray[nr_gpages] = addr;
		nr_gpages++;
		number_of_pages--;
		addr += page_size;
	}
}

472
/* Moves the gigantic page addresses from the temporary list to the
473 474 475
 * huge_boot_pages list.
 */
int alloc_bootmem_huge_page(struct hstate *hstate)
476 477 478 479 480 481 482
{
	struct huge_bootmem_page *m;
	if (nr_gpages == 0)
		return 0;
	m = phys_to_virt(gpage_freearray[--nr_gpages]);
	gpage_freearray[nr_gpages] = 0;
	list_add(&m->list, &huge_boot_pages);
483
	m->hstate = hstate;
484 485
	return 1;
}
B
Becky Bruce 已提交
486
#endif
487

488 489 490 491 492
int huge_pmd_unshare(struct mm_struct *mm, unsigned long *addr, pte_t *ptep)
{
	return 0;
}

493
#ifdef CONFIG_PPC_FSL_BOOK3E
B
Becky Bruce 已提交
494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542
#define HUGEPD_FREELIST_SIZE \
	((PAGE_SIZE - sizeof(struct hugepd_freelist)) / sizeof(pte_t))

struct hugepd_freelist {
	struct rcu_head	rcu;
	unsigned int index;
	void *ptes[0];
};

static DEFINE_PER_CPU(struct hugepd_freelist *, hugepd_freelist_cur);

static void hugepd_free_rcu_callback(struct rcu_head *head)
{
	struct hugepd_freelist *batch =
		container_of(head, struct hugepd_freelist, rcu);
	unsigned int i;

	for (i = 0; i < batch->index; i++)
		kmem_cache_free(hugepte_cache, batch->ptes[i]);

	free_page((unsigned long)batch);
}

static void hugepd_free(struct mmu_gather *tlb, void *hugepte)
{
	struct hugepd_freelist **batchp;

	batchp = &__get_cpu_var(hugepd_freelist_cur);

	if (atomic_read(&tlb->mm->mm_users) < 2 ||
	    cpumask_equal(mm_cpumask(tlb->mm),
			  cpumask_of(smp_processor_id()))) {
		kmem_cache_free(hugepte_cache, hugepte);
		return;
	}

	if (*batchp == NULL) {
		*batchp = (struct hugepd_freelist *)__get_free_page(GFP_ATOMIC);
		(*batchp)->index = 0;
	}

	(*batchp)->ptes[(*batchp)->index++] = hugepte;
	if ((*batchp)->index == HUGEPD_FREELIST_SIZE) {
		call_rcu_sched(&(*batchp)->rcu, hugepd_free_rcu_callback);
		*batchp = NULL;
	}
}
#endif

543 544 545
static void free_hugepd_range(struct mmu_gather *tlb, hugepd_t *hpdp, int pdshift,
			      unsigned long start, unsigned long end,
			      unsigned long floor, unsigned long ceiling)
546 547
{
	pte_t *hugepte = hugepd_page(*hpdp);
B
Becky Bruce 已提交
548 549
	int i;

550
	unsigned long pdmask = ~((1UL << pdshift) - 1);
B
Becky Bruce 已提交
551 552
	unsigned int num_hugepd = 1;

553 554
#ifdef CONFIG_PPC_FSL_BOOK3E
	/* Note: On fsl the hpdp may be the first of several */
B
Becky Bruce 已提交
555
	num_hugepd = (1 << (hugepd_shift(*hpdp) - pdshift));
556 557
#else
	unsigned int shift = hugepd_shift(*hpdp);
B
Becky Bruce 已提交
558
#endif
559 560 561 562 563 564 565 566 567 568 569

	start &= pdmask;
	if (start < floor)
		return;
	if (ceiling) {
		ceiling &= pdmask;
		if (! ceiling)
			return;
	}
	if (end - 1 > ceiling - 1)
		return;
570

B
Becky Bruce 已提交
571 572 573
	for (i = 0; i < num_hugepd; i++, hpdp++)
		hpdp->pd = 0;

574
	tlb->need_flush = 1;
575 576

#ifdef CONFIG_PPC_FSL_BOOK3E
B
Becky Bruce 已提交
577
	hugepd_free(tlb, hugepte);
578 579
#else
	pgtable_free_tlb(tlb, hugepte, pdshift - shift);
B
Becky Bruce 已提交
580
#endif
581 582 583 584
}

static void hugetlb_free_pmd_range(struct mmu_gather *tlb, pud_t *pud,
				   unsigned long addr, unsigned long end,
585
				   unsigned long floor, unsigned long ceiling)
586 587 588 589 590 591 592
{
	pmd_t *pmd;
	unsigned long next;
	unsigned long start;

	start = addr;
	do {
593
		pmd = pmd_offset(pud, addr);
594
		next = pmd_addr_end(addr, end);
595
		if (pmd_none_or_clear_bad(pmd))
596
			continue;
597 598 599 600 601 602 603 604 605
#ifdef CONFIG_PPC_FSL_BOOK3E
		/*
		 * Increment next by the size of the huge mapping since
		 * there may be more than one entry at this level for a
		 * single hugepage, but all of them point to
		 * the same kmem cache that holds the hugepte.
		 */
		next = addr + (1 << hugepd_shift(*(hugepd_t *)pmd));
#endif
606 607
		free_hugepd_range(tlb, (hugepd_t *)pmd, PMD_SHIFT,
				  addr, next, floor, ceiling);
608
	} while (addr = next, addr != end);
609 610 611 612 613 614 615 616

	start &= PUD_MASK;
	if (start < floor)
		return;
	if (ceiling) {
		ceiling &= PUD_MASK;
		if (!ceiling)
			return;
L
Linus Torvalds 已提交
617
	}
618 619
	if (end - 1 > ceiling - 1)
		return;
L
Linus Torvalds 已提交
620

621 622
	pmd = pmd_offset(pud, start);
	pud_clear(pud);
623
	pmd_free_tlb(tlb, pmd, start);
624 625 626 627 628 629 630 631 632 633 634 635
}

static void hugetlb_free_pud_range(struct mmu_gather *tlb, pgd_t *pgd,
				   unsigned long addr, unsigned long end,
				   unsigned long floor, unsigned long ceiling)
{
	pud_t *pud;
	unsigned long next;
	unsigned long start;

	start = addr;
	do {
636
		pud = pud_offset(pgd, addr);
637
		next = pud_addr_end(addr, end);
638
		if (!is_hugepd(pud)) {
639 640
			if (pud_none_or_clear_bad(pud))
				continue;
641
			hugetlb_free_pmd_range(tlb, pud, addr, next, floor,
642
					       ceiling);
643
		} else {
644 645 646 647 648 649 650 651 652
#ifdef CONFIG_PPC_FSL_BOOK3E
			/*
			 * Increment next by the size of the huge mapping since
			 * there may be more than one entry at this level for a
			 * single hugepage, but all of them point to
			 * the same kmem cache that holds the hugepte.
			 */
			next = addr + (1 << hugepd_shift(*(hugepd_t *)pud));
#endif
653 654
			free_hugepd_range(tlb, (hugepd_t *)pud, PUD_SHIFT,
					  addr, next, floor, ceiling);
655
		}
656
	} while (addr = next, addr != end);
657 658 659 660 661 662 663 664 665 666 667 668 669 670

	start &= PGDIR_MASK;
	if (start < floor)
		return;
	if (ceiling) {
		ceiling &= PGDIR_MASK;
		if (!ceiling)
			return;
	}
	if (end - 1 > ceiling - 1)
		return;

	pud = pud_offset(pgd, start);
	pgd_clear(pgd);
671
	pud_free_tlb(tlb, pud, start);
672 673 674 675 676 677 678
}

/*
 * This function frees user-level page tables of a process.
 *
 * Must be called with pagetable lock held.
 */
679
void hugetlb_free_pgd_range(struct mmu_gather *tlb,
680 681 682 683 684 685 686
			    unsigned long addr, unsigned long end,
			    unsigned long floor, unsigned long ceiling)
{
	pgd_t *pgd;
	unsigned long next;

	/*
687 688 689 690 691 692 693 694 695 696
	 * Because there are a number of different possible pagetable
	 * layouts for hugepage ranges, we limit knowledge of how
	 * things should be laid out to the allocation path
	 * (huge_pte_alloc(), above).  Everything else works out the
	 * structure as it goes from information in the hugepd
	 * pointers.  That means that we can't here use the
	 * optimization used in the normal page free_pgd_range(), of
	 * checking whether we're actually covering a large enough
	 * range to have to do anything at the top level of the walk
	 * instead of at the bottom.
697
	 *
698 699 700
	 * To make sense of this, you should probably go read the big
	 * block comment at the top of the normal free_pgd_range(),
	 * too.
701 702 703 704
	 */

	do {
		next = pgd_addr_end(addr, end);
B
Becky Bruce 已提交
705
		pgd = pgd_offset(tlb->mm, addr);
706
		if (!is_hugepd(pgd)) {
707 708 709 710
			if (pgd_none_or_clear_bad(pgd))
				continue;
			hugetlb_free_pud_range(tlb, pgd, addr, next, floor, ceiling);
		} else {
711
#ifdef CONFIG_PPC_FSL_BOOK3E
B
Becky Bruce 已提交
712 713
			/*
			 * Increment next by the size of the huge mapping since
714 715 716
			 * there may be more than one entry at the pgd level
			 * for a single hugepage, but all of them point to the
			 * same kmem cache that holds the hugepte.
B
Becky Bruce 已提交
717 718 719
			 */
			next = addr + (1 << hugepd_shift(*(hugepd_t *)pgd));
#endif
720 721
			free_hugepd_range(tlb, (hugepd_t *)pgd, PGDIR_SHIFT,
					  addr, next, floor, ceiling);
722
		}
B
Becky Bruce 已提交
723
	} while (addr = next, addr != end);
L
Linus Torvalds 已提交
724 725 726 727 728 729 730
}

struct page *
follow_huge_addr(struct mm_struct *mm, unsigned long address, int write)
{
	pte_t *ptep;
	struct page *page;
731 732 733 734
	unsigned shift;
	unsigned long mask;

	ptep = find_linux_pte_or_hugepte(mm->pgd, address, &shift);
L
Linus Torvalds 已提交
735

736
	/* Verify it is a huge page else bail. */
737
	if (!ptep || !shift)
L
Linus Torvalds 已提交
738 739
		return ERR_PTR(-EINVAL);

740
	mask = (1UL << shift) - 1;
L
Linus Torvalds 已提交
741
	page = pte_page(*ptep);
742 743
	if (page)
		page += (address & mask) / PAGE_SIZE;
L
Linus Torvalds 已提交
744 745 746 747 748 749 750 751 752 753 754 755

	return page;
}

struct page *
follow_huge_pmd(struct mm_struct *mm, unsigned long address,
		pmd_t *pmd, int write)
{
	BUG();
	return NULL;
}

756 757
int gup_hugepte(pte_t *ptep, unsigned long sz, unsigned long addr,
		unsigned long end, int write, struct page **pages, int *nr)
758 759 760
{
	unsigned long mask;
	unsigned long pte_end;
761
	struct page *head, *page, *tail;
762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783
	pte_t pte;
	int refs;

	pte_end = (addr + sz) & ~(sz-1);
	if (pte_end < end)
		end = pte_end;

	pte = *ptep;
	mask = _PAGE_PRESENT | _PAGE_USER;
	if (write)
		mask |= _PAGE_RW;

	if ((pte_val(pte) & mask) != mask)
		return 0;

	/* hugepages are never "special" */
	VM_BUG_ON(!pfn_valid(pte_pfn(pte)));

	refs = 0;
	head = pte_page(pte);

	page = head + ((addr & (sz-1)) >> PAGE_SHIFT);
784
	tail = page;
785 786 787 788 789 790 791 792 793 794 795 796 797 798 799
	do {
		VM_BUG_ON(compound_head(page) != head);
		pages[*nr] = page;
		(*nr)++;
		page++;
		refs++;
	} while (addr += PAGE_SIZE, addr != end);

	if (!page_cache_add_speculative(head, refs)) {
		*nr -= refs;
		return 0;
	}

	if (unlikely(pte_val(pte) != pte_val(*ptep))) {
		/* Could be optimized better */
800 801
		*nr -= refs;
		while (refs--)
802
			put_page(head);
803 804 805 806 807 808 809 810 811 812 813
		return 0;
	}

	/*
	 * Any tail page need their mapcount reference taken before we
	 * return.
	 */
	while (refs--) {
		if (PageTail(tail))
			get_huge_page_tail(tail);
		tail++;
814 815 816 817 818
	}

	return 1;
}

D
David Gibson 已提交
819 820 821 822 823 824 825
static unsigned long hugepte_addr_end(unsigned long addr, unsigned long end,
				      unsigned long sz)
{
	unsigned long __boundary = (addr + sz) & ~(sz-1);
	return (__boundary - 1 < end - 1) ? __boundary : end;
}

826 827 828 829 830 831
int gup_hugepd(hugepd_t *hugepd, unsigned pdshift,
	       unsigned long addr, unsigned long end,
	       int write, struct page **pages, int *nr)
{
	pte_t *ptep;
	unsigned long sz = 1UL << hugepd_shift(*hugepd);
D
David Gibson 已提交
832
	unsigned long next;
833 834 835

	ptep = hugepte_offset(hugepd, addr, pdshift);
	do {
D
David Gibson 已提交
836
		next = hugepte_addr_end(addr, end, sz);
837 838
		if (!gup_hugepte(ptep, sz, addr, end, write, pages, nr))
			return 0;
D
David Gibson 已提交
839
	} while (ptep++, addr = next, addr != end);
840 841 842

	return 1;
}
L
Linus Torvalds 已提交
843

844
#ifdef CONFIG_PPC_MM_SLICES
L
Linus Torvalds 已提交
845 846 847 848
unsigned long hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
					unsigned long len, unsigned long pgoff,
					unsigned long flags)
{
849 850
	struct hstate *hstate = hstate_file(file);
	int mmu_psize = shift_to_mmu_psize(huge_page_shift(hstate));
851

852
	return slice_get_unmapped_area(addr, len, flags, mmu_psize, 1);
L
Linus Torvalds 已提交
853
}
854
#endif
L
Linus Torvalds 已提交
855

856 857
unsigned long vma_mmu_pagesize(struct vm_area_struct *vma)
{
858
#ifdef CONFIG_PPC_MM_SLICES
859 860 861
	unsigned int psize = get_slice_psize(vma->vm_mm, vma->vm_start);

	return 1UL << mmu_psize_to_shift(psize);
B
Becky Bruce 已提交
862 863 864 865 866 867 868 869 870 871 872 873 874
#else
	if (!is_vm_hugetlb_page(vma))
		return PAGE_SIZE;

	return huge_page_size(hstate_vma(vma));
#endif
}

static inline bool is_power_of_4(unsigned long x)
{
	if (is_power_of_2(x))
		return (__ilog2(x) % 2) ? false : true;
	return false;
875 876
}

877
static int __init add_huge_page_size(unsigned long long size)
878
{
879 880
	int shift = __ffs(size);
	int mmu_psize;
881

882
	/* Check that it is a page size supported by the hardware and
883
	 * that it fits within pagetable and slice limits. */
B
Becky Bruce 已提交
884 885 886 887
#ifdef CONFIG_PPC_FSL_BOOK3E
	if ((size < PAGE_SIZE) || !is_power_of_4(size))
		return -EINVAL;
#else
888 889 890
	if (!is_power_of_2(size)
	    || (shift > SLICE_HIGH_SHIFT) || (shift <= PAGE_SHIFT))
		return -EINVAL;
B
Becky Bruce 已提交
891
#endif
892

893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912
	if ((mmu_psize = shift_to_mmu_psize(shift)) < 0)
		return -EINVAL;

#ifdef CONFIG_SPU_FS_64K_LS
	/* Disable support for 64K huge pages when 64K SPU local store
	 * support is enabled as the current implementation conflicts.
	 */
	if (shift == PAGE_SHIFT_64K)
		return -EINVAL;
#endif /* CONFIG_SPU_FS_64K_LS */

	BUG_ON(mmu_psize_defs[mmu_psize].shift != shift);

	/* Return if huge page size has already been setup */
	if (size_to_hstate(size))
		return 0;

	hugetlb_add_hstate(shift - PAGE_SHIFT);

	return 0;
913 914 915 916 917 918 919 920
}

static int __init hugepage_setup_sz(char *str)
{
	unsigned long long size;

	size = memparse(str, &str);

921
	if (add_huge_page_size(size) != 0)
922 923 924 925 926 927
		printk(KERN_WARNING "Invalid huge page size specified(%llu)\n", size);

	return 1;
}
__setup("hugepagesz=", hugepage_setup_sz);

928
#ifdef CONFIG_PPC_FSL_BOOK3E
B
Becky Bruce 已提交
929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967
struct kmem_cache *hugepte_cache;
static int __init hugetlbpage_init(void)
{
	int psize;

	for (psize = 0; psize < MMU_PAGE_COUNT; ++psize) {
		unsigned shift;

		if (!mmu_psize_defs[psize].shift)
			continue;

		shift = mmu_psize_to_shift(psize);

		/* Don't treat normal page sizes as huge... */
		if (shift != PAGE_SHIFT)
			if (add_huge_page_size(1ULL << shift) < 0)
				continue;
	}

	/*
	 * Create a kmem cache for hugeptes.  The bottom bits in the pte have
	 * size information encoded in them, so align them to allow this
	 */
	hugepte_cache =  kmem_cache_create("hugepte-cache", sizeof(pte_t),
					   HUGEPD_SHIFT_MASK + 1, 0, NULL);
	if (hugepte_cache == NULL)
		panic("%s: Unable to create kmem cache for hugeptes\n",
		      __func__);

	/* Default hpage size = 4M */
	if (mmu_psize_defs[MMU_PAGE_4M].shift)
		HPAGE_SHIFT = mmu_psize_defs[MMU_PAGE_4M].shift;
	else
		panic("%s: Unable to set default huge page size\n", __func__);


	return 0;
}
#else
968 969
static int __init hugetlbpage_init(void)
{
970
	int psize;
971

972
	if (!mmu_has_feature(MMU_FTR_16M_PAGE))
973
		return -ENODEV;
974

975 976 977
	for (psize = 0; psize < MMU_PAGE_COUNT; ++psize) {
		unsigned shift;
		unsigned pdshift;
978

979 980
		if (!mmu_psize_defs[psize].shift)
			continue;
981

982 983 984 985 986 987 988 989 990 991 992
		shift = mmu_psize_to_shift(psize);

		if (add_huge_page_size(1ULL << shift) < 0)
			continue;

		if (shift < PMD_SHIFT)
			pdshift = PMD_SHIFT;
		else if (shift < PUD_SHIFT)
			pdshift = PUD_SHIFT;
		else
			pdshift = PGDIR_SHIFT;
993 994 995 996 997 998 999 1000 1001 1002
		/*
		 * if we have pdshift and shift value same, we don't
		 * use pgt cache for hugepd.
		 */
		if (pdshift != shift) {
			pgtable_cache_add(pdshift - shift, NULL);
			if (!PGT_CACHE(pdshift - shift))
				panic("hugetlbpage_init(): could not create "
				      "pgtable cache for %d bit pagesize\n", shift);
		}
1003
	}
1004

1005 1006 1007 1008 1009 1010 1011 1012
	/* Set default large page size. Currently, we pick 16M or 1M
	 * depending on what is available
	 */
	if (mmu_psize_defs[MMU_PAGE_16M].shift)
		HPAGE_SHIFT = mmu_psize_defs[MMU_PAGE_16M].shift;
	else if (mmu_psize_defs[MMU_PAGE_1M].shift)
		HPAGE_SHIFT = mmu_psize_defs[MMU_PAGE_1M].shift;

1013 1014
	return 0;
}
B
Becky Bruce 已提交
1015
#endif
1016
module_init(hugetlbpage_init);
1017 1018 1019 1020

void flush_dcache_icache_hugepage(struct page *page)
{
	int i;
B
Becky Bruce 已提交
1021
	void *start;
1022 1023 1024

	BUG_ON(!PageCompound(page));

B
Becky Bruce 已提交
1025 1026 1027 1028
	for (i = 0; i < (1UL << compound_order(page)); i++) {
		if (!PageHighMem(page)) {
			__flush_dcache_icache(page_address(page+i));
		} else {
1029
			start = kmap_atomic(page+i);
B
Becky Bruce 已提交
1030
			__flush_dcache_icache(start);
1031
			kunmap_atomic(start);
B
Becky Bruce 已提交
1032 1033
		}
	}
1034
}