inode.c 19.5 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
/*
 * hugetlbpage-backed filesystem.  Based on ramfs.
 *
 * William Irwin, 2002
 *
 * Copyright (C) 2002 Linus Torvalds.
 */

#include <linux/module.h>
#include <linux/thread_info.h>
#include <asm/current.h>
#include <linux/sched.h>		/* remove ASAP */
#include <linux/fs.h>
#include <linux/mount.h>
#include <linux/file.h>
#include <linux/writeback.h>
#include <linux/pagemap.h>
#include <linux/highmem.h>
#include <linux/init.h>
#include <linux/string.h>
21
#include <linux/capability.h>
L
Linus Torvalds 已提交
22 23 24 25 26 27 28 29 30 31 32 33 34 35
#include <linux/backing-dev.h>
#include <linux/hugetlb.h>
#include <linux/pagevec.h>
#include <linux/quotaops.h>
#include <linux/slab.h>
#include <linux/dnotify.h>
#include <linux/statfs.h>
#include <linux/security.h>

#include <asm/uaccess.h>

/* some random number */
#define HUGETLBFS_MAGIC	0x958458f6

36
static const struct super_operations hugetlbfs_ops;
37
static const struct address_space_operations hugetlbfs_aops;
38
const struct file_operations hugetlbfs_file_operations;
39 40
static const struct inode_operations hugetlbfs_dir_inode_operations;
static const struct inode_operations hugetlbfs_inode_operations;
L
Linus Torvalds 已提交
41 42 43 44 45 46 47 48

static struct backing_dev_info hugetlbfs_backing_dev_info = {
	.ra_pages	= 0,	/* No readahead */
	.capabilities	= BDI_CAP_NO_ACCT_DIRTY | BDI_CAP_NO_WRITEBACK,
};

int sysctl_hugetlb_shm_group;

49 50 51 52 53 54 55 56 57 58
static void huge_pagevec_release(struct pagevec *pvec)
{
	int i;

	for (i = 0; i < pagevec_count(pvec); ++i)
		put_page(pvec->pages[i]);

	pagevec_reinit(pvec);
}

L
Linus Torvalds 已提交
59 60
static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma)
{
61
	struct inode *inode = file->f_path.dentry->d_inode;
L
Linus Torvalds 已提交
62 63 64
	loff_t len, vma_len;
	int ret;

65 66 67 68 69 70 71 72
	/*
	 * vma alignment has already been checked by prepare_hugepage_range.
	 * If you add any error returns here, do so after setting VM_HUGETLB,
	 * so is_vm_hugetlb_page tests below unmap_region go the right way
	 * when do_mmap_pgoff unwinds (may be important on powerpc and ia64).
	 */
	vma->vm_flags |= VM_HUGETLB | VM_RESERVED;
	vma->vm_ops = &hugetlb_vm_ops;
L
Linus Torvalds 已提交
73 74 75

	vma_len = (loff_t)(vma->vm_end - vma->vm_start);

76
	mutex_lock(&inode->i_mutex);
L
Linus Torvalds 已提交
77 78 79 80 81
	file_accessed(file);

	ret = -ENOMEM;
	len = vma_len + ((loff_t)vma->vm_pgoff << PAGE_SHIFT);

82 83 84 85
	if (vma->vm_flags & VM_MAYSHARE &&
	    hugetlb_reserve_pages(inode, vma->vm_pgoff >> (HPAGE_SHIFT-PAGE_SHIFT),
				  len >> HPAGE_SHIFT))
		goto out;
86

A
Adam Litke 已提交
87 88
	ret = 0;
	hugetlb_prefault_arch_hook(vma->vm_mm);
89
	if (vma->vm_flags & VM_WRITE && inode->i_size < len)
L
Linus Torvalds 已提交
90 91
		inode->i_size = len;
out:
92
	mutex_unlock(&inode->i_mutex);
L
Linus Torvalds 已提交
93 94 95 96 97

	return ret;
}

/*
98
 * Called under down_write(mmap_sem).
L
Linus Torvalds 已提交
99 100
 */

101
#ifndef HAVE_ARCH_HUGETLB_UNMAPPED_AREA
L
Linus Torvalds 已提交
102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
static unsigned long
hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
		unsigned long len, unsigned long pgoff, unsigned long flags)
{
	struct mm_struct *mm = current->mm;
	struct vm_area_struct *vma;
	unsigned long start_addr;

	if (len & ~HPAGE_MASK)
		return -EINVAL;
	if (len > TASK_SIZE)
		return -ENOMEM;

	if (addr) {
		addr = ALIGN(addr, HPAGE_SIZE);
		vma = find_vma(mm, addr);
		if (TASK_SIZE - len >= addr &&
		    (!vma || addr + len <= vma->vm_start))
			return addr;
	}

	start_addr = mm->free_area_cache;

125 126 127
	if (len <= mm->cached_hole_size)
		start_addr = TASK_UNMAPPED_BASE;

L
Linus Torvalds 已提交
128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175
full_search:
	addr = ALIGN(start_addr, HPAGE_SIZE);

	for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
		/* At this point:  (!vma || addr < vma->vm_end). */
		if (TASK_SIZE - len < addr) {
			/*
			 * Start a new search - just in case we missed
			 * some holes.
			 */
			if (start_addr != TASK_UNMAPPED_BASE) {
				start_addr = TASK_UNMAPPED_BASE;
				goto full_search;
			}
			return -ENOMEM;
		}

		if (!vma || addr + len <= vma->vm_start)
			return addr;
		addr = ALIGN(vma->vm_end, HPAGE_SIZE);
	}
}
#endif

/*
 * Read a page. Again trivial. If it didn't already exist
 * in the page cache, it is zero-filled.
 */
static int hugetlbfs_readpage(struct file *file, struct page * page)
{
	unlock_page(page);
	return -EINVAL;
}

static int hugetlbfs_prepare_write(struct file *file,
			struct page *page, unsigned offset, unsigned to)
{
	return -EINVAL;
}

static int hugetlbfs_commit_write(struct file *file,
			struct page *page, unsigned offset, unsigned to)
{
	return -EINVAL;
}

static void truncate_huge_page(struct page *page)
{
176
	cancel_dirty_page(page, /* No IO accounting for huge pages? */0);
L
Linus Torvalds 已提交
177 178 179 180 181
	ClearPageUptodate(page);
	remove_from_page_cache(page);
	put_page(page);
}

182
static void truncate_hugepages(struct inode *inode, loff_t lstart)
L
Linus Torvalds 已提交
183
{
184
	struct address_space *mapping = &inode->i_data;
L
Linus Torvalds 已提交
185 186 187
	const pgoff_t start = lstart >> HPAGE_SHIFT;
	struct pagevec pvec;
	pgoff_t next;
188
	int i, freed = 0;
L
Linus Torvalds 已提交
189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209

	pagevec_init(&pvec, 0);
	next = start;
	while (1) {
		if (!pagevec_lookup(&pvec, mapping, next, PAGEVEC_SIZE)) {
			if (next == start)
				break;
			next = start;
			continue;
		}

		for (i = 0; i < pagevec_count(&pvec); ++i) {
			struct page *page = pvec.pages[i];

			lock_page(page);
			if (page->index > next)
				next = page->index;
			++next;
			truncate_huge_page(page);
			unlock_page(page);
			hugetlb_put_quota(mapping);
210
			freed++;
L
Linus Torvalds 已提交
211 212 213 214
		}
		huge_pagevec_release(&pvec);
	}
	BUG_ON(!lstart && mapping->nrpages);
215
	hugetlb_unreserve_pages(inode, start, freed);
L
Linus Torvalds 已提交
216 217 218 219
}

static void hugetlbfs_delete_inode(struct inode *inode)
{
220
	truncate_hugepages(inode, 0);
221 222 223
	clear_inode(inode);
}

224
static void hugetlbfs_forget_inode(struct inode *inode) __releases(inode_lock)
L
Linus Torvalds 已提交
225
{
226 227 228 229 230 231 232 233 234 235 236
	struct super_block *sb = inode->i_sb;

	if (!hlist_unhashed(&inode->i_hash)) {
		if (!(inode->i_state & (I_DIRTY|I_LOCK)))
			list_move(&inode->i_list, &inode_unused);
		inodes_stat.nr_unused++;
		if (!sb || (sb->s_flags & MS_ACTIVE)) {
			spin_unlock(&inode_lock);
			return;
		}
		inode->i_state |= I_WILL_FREE;
L
Linus Torvalds 已提交
237
		spin_unlock(&inode_lock);
238 239 240 241 242 243 244 245 246
		/*
		 * write_inode_now is a noop as we set BDI_CAP_NO_WRITEBACK
		 * in our backing_dev_info.
		 */
		write_inode_now(inode, 1);
		spin_lock(&inode_lock);
		inode->i_state &= ~I_WILL_FREE;
		inodes_stat.nr_unused--;
		hlist_del_init(&inode->i_hash);
L
Linus Torvalds 已提交
247 248 249 250 251 252
	}
	list_del_init(&inode->i_list);
	list_del_init(&inode->i_sb_list);
	inode->i_state |= I_FREEING;
	inodes_stat.nr_inodes--;
	spin_unlock(&inode_lock);
253
	truncate_hugepages(inode, 0);
L
Linus Torvalds 已提交
254 255 256 257 258 259 260
	clear_inode(inode);
	destroy_inode(inode);
}

static void hugetlbfs_drop_inode(struct inode *inode)
{
	if (!inode->i_nlink)
261
		generic_delete_inode(inode);
L
Linus Torvalds 已提交
262 263 264 265 266
	else
		hugetlbfs_forget_inode(inode);
}

static inline void
H
Hugh Dickins 已提交
267
hugetlb_vmtruncate_list(struct prio_tree_root *root, pgoff_t pgoff)
L
Linus Torvalds 已提交
268 269 270 271
{
	struct vm_area_struct *vma;
	struct prio_tree_iter iter;

H
Hugh Dickins 已提交
272
	vma_prio_tree_foreach(vma, &iter, root, pgoff, ULONG_MAX) {
L
Linus Torvalds 已提交
273 274 275
		unsigned long v_offset;

		/*
H
Hugh Dickins 已提交
276 277 278 279
		 * Can the expression below overflow on 32-bit arches?
		 * No, because the prio_tree returns us only those vmas
		 * which overlap the truncated area starting at pgoff,
		 * and no vma on a 32-bit arch can span beyond the 4GB.
L
Linus Torvalds 已提交
280
		 */
H
Hugh Dickins 已提交
281 282 283
		if (vma->vm_pgoff < pgoff)
			v_offset = (pgoff - vma->vm_pgoff) << PAGE_SHIFT;
		else
L
Linus Torvalds 已提交
284 285
			v_offset = 0;

286
		__unmap_hugepage_range(vma,
287
				vma->vm_start + v_offset, vma->vm_end);
L
Linus Torvalds 已提交
288 289 290 291 292 293 294 295
	}
}

/*
 * Expanding truncates are not allowed.
 */
static int hugetlb_vmtruncate(struct inode *inode, loff_t offset)
{
H
Hugh Dickins 已提交
296
	pgoff_t pgoff;
L
Linus Torvalds 已提交
297 298 299 300 301 302
	struct address_space *mapping = inode->i_mapping;

	if (offset > inode->i_size)
		return -EINVAL;

	BUG_ON(offset & ~HPAGE_MASK);
H
Hugh Dickins 已提交
303
	pgoff = offset >> PAGE_SHIFT;
L
Linus Torvalds 已提交
304 305 306 307 308 309

	inode->i_size = offset;
	spin_lock(&mapping->i_mmap_lock);
	if (!prio_tree_empty(&mapping->i_mmap))
		hugetlb_vmtruncate_list(&mapping->i_mmap, pgoff);
	spin_unlock(&mapping->i_mmap_lock);
310
	truncate_hugepages(inode, offset);
L
Linus Torvalds 已提交
311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353
	return 0;
}

static int hugetlbfs_setattr(struct dentry *dentry, struct iattr *attr)
{
	struct inode *inode = dentry->d_inode;
	int error;
	unsigned int ia_valid = attr->ia_valid;

	BUG_ON(!inode);

	error = inode_change_ok(inode, attr);
	if (error)
		goto out;

	if (ia_valid & ATTR_SIZE) {
		error = -EINVAL;
		if (!(attr->ia_size & ~HPAGE_MASK))
			error = hugetlb_vmtruncate(inode, attr->ia_size);
		if (error)
			goto out;
		attr->ia_valid &= ~ATTR_SIZE;
	}
	error = inode_setattr(inode, attr);
out:
	return error;
}

static struct inode *hugetlbfs_get_inode(struct super_block *sb, uid_t uid, 
					gid_t gid, int mode, dev_t dev)
{
	struct inode *inode;

	inode = new_inode(sb);
	if (inode) {
		struct hugetlbfs_inode_info *info;
		inode->i_mode = mode;
		inode->i_uid = uid;
		inode->i_gid = gid;
		inode->i_blocks = 0;
		inode->i_mapping->a_ops = &hugetlbfs_aops;
		inode->i_mapping->backing_dev_info =&hugetlbfs_backing_dev_info;
		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
354
		INIT_LIST_HEAD(&inode->i_mapping->private_list);
L
Linus Torvalds 已提交
355
		info = HUGETLBFS_I(inode);
356
		mpol_shared_policy_init(&info->policy, MPOL_DEFAULT, NULL);
L
Linus Torvalds 已提交
357 358 359 360 361 362 363 364 365 366 367 368 369
		switch (mode & S_IFMT) {
		default:
			init_special_inode(inode, mode, dev);
			break;
		case S_IFREG:
			inode->i_op = &hugetlbfs_inode_operations;
			inode->i_fop = &hugetlbfs_file_operations;
			break;
		case S_IFDIR:
			inode->i_op = &hugetlbfs_dir_inode_operations;
			inode->i_fop = &simple_dir_operations;

			/* directory inodes start off with i_nlink == 2 (for "." entry) */
370
			inc_nlink(inode);
L
Linus Torvalds 已提交
371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410
			break;
		case S_IFLNK:
			inode->i_op = &page_symlink_inode_operations;
			break;
		}
	}
	return inode;
}

/*
 * File creation. Allocate an inode, and we're done..
 */
static int hugetlbfs_mknod(struct inode *dir,
			struct dentry *dentry, int mode, dev_t dev)
{
	struct inode *inode;
	int error = -ENOSPC;
	gid_t gid;

	if (dir->i_mode & S_ISGID) {
		gid = dir->i_gid;
		if (S_ISDIR(mode))
			mode |= S_ISGID;
	} else {
		gid = current->fsgid;
	}
	inode = hugetlbfs_get_inode(dir->i_sb, current->fsuid, gid, mode, dev);
	if (inode) {
		dir->i_ctime = dir->i_mtime = CURRENT_TIME;
		d_instantiate(dentry, inode);
		dget(dentry);	/* Extra count - pin the dentry in core */
		error = 0;
	}
	return error;
}

static int hugetlbfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
{
	int retval = hugetlbfs_mknod(dir, dentry, mode | S_IFDIR, 0);
	if (!retval)
411
		inc_nlink(dir);
L
Linus Torvalds 已提交
412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448
	return retval;
}

static int hugetlbfs_create(struct inode *dir, struct dentry *dentry, int mode, struct nameidata *nd)
{
	return hugetlbfs_mknod(dir, dentry, mode | S_IFREG, 0);
}

static int hugetlbfs_symlink(struct inode *dir,
			struct dentry *dentry, const char *symname)
{
	struct inode *inode;
	int error = -ENOSPC;
	gid_t gid;

	if (dir->i_mode & S_ISGID)
		gid = dir->i_gid;
	else
		gid = current->fsgid;

	inode = hugetlbfs_get_inode(dir->i_sb, current->fsuid,
					gid, S_IFLNK|S_IRWXUGO, 0);
	if (inode) {
		int l = strlen(symname)+1;
		error = page_symlink(inode, symname, l);
		if (!error) {
			d_instantiate(dentry, inode);
			dget(dentry);
		} else
			iput(inode);
	}
	dir->i_ctime = dir->i_mtime = CURRENT_TIME;

	return error;
}

/*
449
 * mark the head page dirty
L
Linus Torvalds 已提交
450 451 452
 */
static int hugetlbfs_set_page_dirty(struct page *page)
{
453 454 455
	struct page *head = (struct page *)page_private(page);

	SetPageDirty(head);
L
Linus Torvalds 已提交
456 457 458
	return 0;
}

459
static int hugetlbfs_statfs(struct dentry *dentry, struct kstatfs *buf)
L
Linus Torvalds 已提交
460
{
461
	struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(dentry->d_sb);
L
Linus Torvalds 已提交
462 463 464 465 466

	buf->f_type = HUGETLBFS_MAGIC;
	buf->f_bsize = HPAGE_SIZE;
	if (sbinfo) {
		spin_lock(&sbinfo->stat_lock);
467 468 469 470 471 472 473 474
		/* If no limits set, just report 0 for max/free/used
		 * blocks, like simple_statfs() */
		if (sbinfo->max_blocks >= 0) {
			buf->f_blocks = sbinfo->max_blocks;
			buf->f_bavail = buf->f_bfree = sbinfo->free_blocks;
			buf->f_files = sbinfo->max_inodes;
			buf->f_ffree = sbinfo->free_inodes;
		}
L
Linus Torvalds 已提交
475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490
		spin_unlock(&sbinfo->stat_lock);
	}
	buf->f_namelen = NAME_MAX;
	return 0;
}

static void hugetlbfs_put_super(struct super_block *sb)
{
	struct hugetlbfs_sb_info *sbi = HUGETLBFS_SB(sb);

	if (sbi) {
		sb->s_fs_info = NULL;
		kfree(sbi);
	}
}

491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515
static inline int hugetlbfs_dec_free_inodes(struct hugetlbfs_sb_info *sbinfo)
{
	if (sbinfo->free_inodes >= 0) {
		spin_lock(&sbinfo->stat_lock);
		if (unlikely(!sbinfo->free_inodes)) {
			spin_unlock(&sbinfo->stat_lock);
			return 0;
		}
		sbinfo->free_inodes--;
		spin_unlock(&sbinfo->stat_lock);
	}

	return 1;
}

static void hugetlbfs_inc_free_inodes(struct hugetlbfs_sb_info *sbinfo)
{
	if (sbinfo->free_inodes >= 0) {
		spin_lock(&sbinfo->stat_lock);
		sbinfo->free_inodes++;
		spin_unlock(&sbinfo->stat_lock);
	}
}


516
static struct kmem_cache *hugetlbfs_inode_cachep;
L
Linus Torvalds 已提交
517 518 519

static struct inode *hugetlbfs_alloc_inode(struct super_block *sb)
{
520
	struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(sb);
L
Linus Torvalds 已提交
521 522
	struct hugetlbfs_inode_info *p;

523 524
	if (unlikely(!hugetlbfs_dec_free_inodes(sbinfo)))
		return NULL;
525
	p = kmem_cache_alloc(hugetlbfs_inode_cachep, GFP_KERNEL);
526 527
	if (unlikely(!p)) {
		hugetlbfs_inc_free_inodes(sbinfo);
L
Linus Torvalds 已提交
528
		return NULL;
529
	}
L
Linus Torvalds 已提交
530 531 532 533 534
	return &p->vfs_inode;
}

static void hugetlbfs_destroy_inode(struct inode *inode)
{
535
	hugetlbfs_inc_free_inodes(HUGETLBFS_SB(inode->i_sb));
L
Linus Torvalds 已提交
536 537 538 539
	mpol_free_shared_policy(&HUGETLBFS_I(inode)->policy);
	kmem_cache_free(hugetlbfs_inode_cachep, HUGETLBFS_I(inode));
}

540
static const struct address_space_operations hugetlbfs_aops = {
L
Linus Torvalds 已提交
541 542 543 544 545 546
	.readpage	= hugetlbfs_readpage,
	.prepare_write	= hugetlbfs_prepare_write,
	.commit_write	= hugetlbfs_commit_write,
	.set_page_dirty	= hugetlbfs_set_page_dirty,
};

547

548
static void init_once(void *foo, struct kmem_cache *cachep, unsigned long flags)
549 550 551 552 553 554 555 556
{
	struct hugetlbfs_inode_info *ei = (struct hugetlbfs_inode_info *)foo;

	if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
	    SLAB_CTOR_CONSTRUCTOR)
		inode_init_once(&ei->vfs_inode);
}

557
const struct file_operations hugetlbfs_file_operations = {
L
Linus Torvalds 已提交
558 559 560 561 562
	.mmap			= hugetlbfs_file_mmap,
	.fsync			= simple_sync_file,
	.get_unmapped_area	= hugetlb_get_unmapped_area,
};

563
static const struct inode_operations hugetlbfs_dir_inode_operations = {
L
Linus Torvalds 已提交
564 565 566 567 568 569 570 571 572 573 574 575
	.create		= hugetlbfs_create,
	.lookup		= simple_lookup,
	.link		= simple_link,
	.unlink		= simple_unlink,
	.symlink	= hugetlbfs_symlink,
	.mkdir		= hugetlbfs_mkdir,
	.rmdir		= simple_rmdir,
	.mknod		= hugetlbfs_mknod,
	.rename		= simple_rename,
	.setattr	= hugetlbfs_setattr,
};

576
static const struct inode_operations hugetlbfs_inode_operations = {
L
Linus Torvalds 已提交
577 578 579
	.setattr	= hugetlbfs_setattr,
};

580
static const struct super_operations hugetlbfs_ops = {
L
Linus Torvalds 已提交
581 582 583
	.alloc_inode    = hugetlbfs_alloc_inode,
	.destroy_inode  = hugetlbfs_destroy_inode,
	.statfs		= hugetlbfs_statfs,
584
	.delete_inode	= hugetlbfs_delete_inode,
L
Linus Torvalds 已提交
585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712
	.drop_inode	= hugetlbfs_drop_inode,
	.put_super	= hugetlbfs_put_super,
};

static int
hugetlbfs_parse_options(char *options, struct hugetlbfs_config *pconfig)
{
	char *opt, *value, *rest;

	if (!options)
		return 0;
	while ((opt = strsep(&options, ",")) != NULL) {
		if (!*opt)
			continue;

		value = strchr(opt, '=');
		if (!value || !*value)
			return -EINVAL;
		else
			*value++ = '\0';

		if (!strcmp(opt, "uid"))
			pconfig->uid = simple_strtoul(value, &value, 0);
		else if (!strcmp(opt, "gid"))
			pconfig->gid = simple_strtoul(value, &value, 0);
		else if (!strcmp(opt, "mode"))
			pconfig->mode = simple_strtoul(value,&value,0) & 0777U;
		else if (!strcmp(opt, "size")) {
			unsigned long long size = memparse(value, &rest);
			if (*rest == '%') {
				size <<= HPAGE_SHIFT;
				size *= max_huge_pages;
				do_div(size, 100);
				rest++;
			}
			pconfig->nr_blocks = (size >> HPAGE_SHIFT);
			value = rest;
		} else if (!strcmp(opt,"nr_inodes")) {
			pconfig->nr_inodes = memparse(value, &rest);
			value = rest;
		} else
			return -EINVAL;

		if (*value)
			return -EINVAL;
	}
	return 0;
}

static int
hugetlbfs_fill_super(struct super_block *sb, void *data, int silent)
{
	struct inode * inode;
	struct dentry * root;
	int ret;
	struct hugetlbfs_config config;
	struct hugetlbfs_sb_info *sbinfo;

	config.nr_blocks = -1; /* No limit on size by default */
	config.nr_inodes = -1; /* No limit on number of inodes by default */
	config.uid = current->fsuid;
	config.gid = current->fsgid;
	config.mode = 0755;
	ret = hugetlbfs_parse_options(data, &config);

	if (ret)
		return ret;

	sbinfo = kmalloc(sizeof(struct hugetlbfs_sb_info), GFP_KERNEL);
	if (!sbinfo)
		return -ENOMEM;
	sb->s_fs_info = sbinfo;
	spin_lock_init(&sbinfo->stat_lock);
	sbinfo->max_blocks = config.nr_blocks;
	sbinfo->free_blocks = config.nr_blocks;
	sbinfo->max_inodes = config.nr_inodes;
	sbinfo->free_inodes = config.nr_inodes;
	sb->s_maxbytes = MAX_LFS_FILESIZE;
	sb->s_blocksize = HPAGE_SIZE;
	sb->s_blocksize_bits = HPAGE_SHIFT;
	sb->s_magic = HUGETLBFS_MAGIC;
	sb->s_op = &hugetlbfs_ops;
	sb->s_time_gran = 1;
	inode = hugetlbfs_get_inode(sb, config.uid, config.gid,
					S_IFDIR | config.mode, 0);
	if (!inode)
		goto out_free;

	root = d_alloc_root(inode);
	if (!root) {
		iput(inode);
		goto out_free;
	}
	sb->s_root = root;
	return 0;
out_free:
	kfree(sbinfo);
	return -ENOMEM;
}

int hugetlb_get_quota(struct address_space *mapping)
{
	int ret = 0;
	struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(mapping->host->i_sb);

	if (sbinfo->free_blocks > -1) {
		spin_lock(&sbinfo->stat_lock);
		if (sbinfo->free_blocks > 0)
			sbinfo->free_blocks--;
		else
			ret = -ENOMEM;
		spin_unlock(&sbinfo->stat_lock);
	}

	return ret;
}

void hugetlb_put_quota(struct address_space *mapping)
{
	struct hugetlbfs_sb_info *sbinfo = HUGETLBFS_SB(mapping->host->i_sb);

	if (sbinfo->free_blocks > -1) {
		spin_lock(&sbinfo->stat_lock);
		sbinfo->free_blocks++;
		spin_unlock(&sbinfo->stat_lock);
	}
}

713 714
static int hugetlbfs_get_sb(struct file_system_type *fs_type,
	int flags, const char *dev_name, void *data, struct vfsmount *mnt)
L
Linus Torvalds 已提交
715
{
716
	return get_sb_nodev(fs_type, flags, data, hugetlbfs_fill_super, mnt);
L
Linus Torvalds 已提交
717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741
}

static struct file_system_type hugetlbfs_fs_type = {
	.name		= "hugetlbfs",
	.get_sb		= hugetlbfs_get_sb,
	.kill_sb	= kill_litter_super,
};

static struct vfsmount *hugetlbfs_vfsmount;

static int can_do_hugetlb_shm(void)
{
	return likely(capable(CAP_IPC_LOCK) ||
			in_group_p(sysctl_hugetlb_shm_group) ||
			can_do_mlock());
}

struct file *hugetlb_zero_setup(size_t size)
{
	int error = -ENOMEM;
	struct file *file;
	struct inode *inode;
	struct dentry *dentry, *root;
	struct qstr quick_string;
	char buf[16];
742
	static atomic_t counter;
L
Linus Torvalds 已提交
743 744 745 746 747 748 749 750

	if (!can_do_hugetlb_shm())
		return ERR_PTR(-EPERM);

	if (!user_shm_lock(size, current->user))
		return ERR_PTR(-ENOMEM);

	root = hugetlbfs_vfsmount->mnt_root;
751
	snprintf(buf, 16, "%u", atomic_inc_return(&counter));
L
Linus Torvalds 已提交
752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769
	quick_string.name = buf;
	quick_string.len = strlen(quick_string.name);
	quick_string.hash = 0;
	dentry = d_alloc(root, &quick_string);
	if (!dentry)
		goto out_shm_unlock;

	error = -ENFILE;
	file = get_empty_filp();
	if (!file)
		goto out_dentry;

	error = -ENOSPC;
	inode = hugetlbfs_get_inode(root->d_sb, current->fsuid,
				current->fsgid, S_IFREG | S_IRWXUGO, 0);
	if (!inode)
		goto out_file;

770
	error = -ENOMEM;
771
	if (hugetlb_reserve_pages(inode, 0, size >> HPAGE_SHIFT))
772 773
		goto out_inode;

L
Linus Torvalds 已提交
774 775 776
	d_instantiate(dentry, inode);
	inode->i_size = size;
	inode->i_nlink = 0;
777 778
	file->f_path.mnt = mntget(hugetlbfs_vfsmount);
	file->f_path.dentry = dentry;
L
Linus Torvalds 已提交
779 780 781 782 783
	file->f_mapping = inode->i_mapping;
	file->f_op = &hugetlbfs_file_operations;
	file->f_mode = FMODE_WRITE | FMODE_READ;
	return file;

784 785
out_inode:
	iput(inode);
L
Linus Torvalds 已提交
786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834
out_file:
	put_filp(file);
out_dentry:
	dput(dentry);
out_shm_unlock:
	user_shm_unlock(size, current->user);
	return ERR_PTR(error);
}

static int __init init_hugetlbfs_fs(void)
{
	int error;
	struct vfsmount *vfsmount;

	hugetlbfs_inode_cachep = kmem_cache_create("hugetlbfs_inode_cache",
					sizeof(struct hugetlbfs_inode_info),
					0, 0, init_once, NULL);
	if (hugetlbfs_inode_cachep == NULL)
		return -ENOMEM;

	error = register_filesystem(&hugetlbfs_fs_type);
	if (error)
		goto out;

	vfsmount = kern_mount(&hugetlbfs_fs_type);

	if (!IS_ERR(vfsmount)) {
		hugetlbfs_vfsmount = vfsmount;
		return 0;
	}

	error = PTR_ERR(vfsmount);

 out:
	if (error)
		kmem_cache_destroy(hugetlbfs_inode_cachep);
	return error;
}

static void __exit exit_hugetlbfs_fs(void)
{
	kmem_cache_destroy(hugetlbfs_inode_cachep);
	unregister_filesystem(&hugetlbfs_fs_type);
}

module_init(init_hugetlbfs_fs)
module_exit(exit_hugetlbfs_fs)

MODULE_LICENSE("GPL");