nommu.c 45.3 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8
/*
 *  linux/mm/nommu.c
 *
 *  Replacement code for mm functions to support CPU's that don't
 *  have any form of memory management unit (thus no virtual memory).
 *
 *  See Documentation/nommu-mmap.txt
 *
9
 *  Copyright (c) 2004-2008 David Howells <dhowells@redhat.com>
L
Linus Torvalds 已提交
10 11 12
 *  Copyright (c) 2000-2003 David McCullough <davidm@snapgear.com>
 *  Copyright (c) 2000-2001 D Jeff Dionne <jeff@uClinux.org>
 *  Copyright (c) 2002      Greg Ungerer <gerg@snapgear.com>
13
 *  Copyright (c) 2007-2008 Paul Mundt <lethal@linux-sh.org>
L
Linus Torvalds 已提交
14 15
 */

16
#include <linux/module.h>
L
Linus Torvalds 已提交
17 18 19 20 21 22 23 24
#include <linux/mm.h>
#include <linux/mman.h>
#include <linux/swap.h>
#include <linux/file.h>
#include <linux/highmem.h>
#include <linux/pagemap.h>
#include <linux/slab.h>
#include <linux/vmalloc.h>
25
#include <linux/tracehook.h>
L
Linus Torvalds 已提交
26 27 28 29 30 31 32 33 34 35
#include <linux/blkdev.h>
#include <linux/backing-dev.h>
#include <linux/mount.h>
#include <linux/personality.h>
#include <linux/security.h>
#include <linux/syscalls.h>

#include <asm/uaccess.h>
#include <asm/tlb.h>
#include <asm/tlbflush.h>
36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
#include "internal.h"

static inline __attribute__((format(printf, 1, 2)))
void no_printk(const char *fmt, ...)
{
}

#if 0
#define kenter(FMT, ...) \
	printk(KERN_DEBUG "==> %s("FMT")\n", __func__, ##__VA_ARGS__)
#define kleave(FMT, ...) \
	printk(KERN_DEBUG "<== %s()"FMT"\n", __func__, ##__VA_ARGS__)
#define kdebug(FMT, ...) \
	printk(KERN_DEBUG "xxx" FMT"yyy\n", ##__VA_ARGS__)
#else
#define kenter(FMT, ...) \
	no_printk(KERN_DEBUG "==> %s("FMT")\n", __func__, ##__VA_ARGS__)
#define kleave(FMT, ...) \
	no_printk(KERN_DEBUG "<== %s()"FMT"\n", __func__, ##__VA_ARGS__)
#define kdebug(FMT, ...) \
	no_printk(KERN_DEBUG FMT"\n", ##__VA_ARGS__)
#endif
L
Linus Torvalds 已提交
58

N
Nick Piggin 已提交
59 60
#include "internal.h"

L
Linus Torvalds 已提交
61 62 63 64
void *high_memory;
struct page *mem_map;
unsigned long max_mapnr;
unsigned long num_physpages;
A
Alan Cox 已提交
65
atomic_long_t vm_committed_space = ATOMIC_LONG_INIT(0);
L
Linus Torvalds 已提交
66 67 68
int sysctl_overcommit_memory = OVERCOMMIT_GUESS; /* heuristic overcommit */
int sysctl_overcommit_ratio = 50; /* default is 50% */
int sysctl_max_map_count = DEFAULT_MAX_MAP_COUNT;
69
int sysctl_nr_trim_pages = 1; /* page trimming behaviour */
L
Linus Torvalds 已提交
70 71
int heap_stack_gap = 0;

72 73
atomic_t mmap_pages_allocated;

L
Linus Torvalds 已提交
74
EXPORT_SYMBOL(mem_map);
75
EXPORT_SYMBOL(num_physpages);
L
Linus Torvalds 已提交
76

77 78 79 80
/* list of mapped, potentially shareable regions */
static struct kmem_cache *vm_region_jar;
struct rb_root nommu_region_tree = RB_ROOT;
DECLARE_RWSEM(nommu_region_sem);
L
Linus Torvalds 已提交
81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113

struct vm_operations_struct generic_file_vm_ops = {
};

/*
 * Handle all mappings that got truncated by a "truncate()"
 * system call.
 *
 * NOTE! We have to be ready to update the memory sharing
 * between the file and the memory map for a potential last
 * incomplete page.  Ugly, but necessary.
 */
int vmtruncate(struct inode *inode, loff_t offset)
{
	struct address_space *mapping = inode->i_mapping;
	unsigned long limit;

	if (inode->i_size < offset)
		goto do_expand;
	i_size_write(inode, offset);

	truncate_inode_pages(mapping, offset);
	goto out_truncate;

do_expand:
	limit = current->signal->rlim[RLIMIT_FSIZE].rlim_cur;
	if (limit != RLIM_INFINITY && offset > limit)
		goto out_sig;
	if (offset > inode->i_sb->s_maxbytes)
		goto out;
	i_size_write(inode, offset);

out_truncate:
A
Al Viro 已提交
114
	if (inode->i_op->truncate)
L
Linus Torvalds 已提交
115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134
		inode->i_op->truncate(inode);
	return 0;
out_sig:
	send_sig(SIGXFSZ, current, 0);
out:
	return -EFBIG;
}

EXPORT_SYMBOL(vmtruncate);

/*
 * Return the total memory allocated for this pointer, not
 * just what the caller asked for.
 *
 * Doesn't have to be accurate, i.e. may have races.
 */
unsigned int kobjsize(const void *objp)
{
	struct page *page;

135 136 137 138
	/*
	 * If the object we have should not have ksize performed on it,
	 * return size of 0
	 */
139
	if (!objp || !virt_addr_valid(objp))
140 141 142 143 144 145 146 147
		return 0;

	page = virt_to_head_page(objp);

	/*
	 * If the allocator sets PageSlab, we know the pointer came from
	 * kmalloc().
	 */
L
Linus Torvalds 已提交
148 149 150
	if (PageSlab(page))
		return ksize(objp);

151 152
	/*
	 * The ksize() function is only guaranteed to work for pointers
153
	 * returned by kmalloc(). So handle arbitrary pointers here.
154
	 */
155
	return PAGE_SIZE << compound_order(page);
L
Linus Torvalds 已提交
156 157
}

N
Nick Piggin 已提交
158 159 160
int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
		     unsigned long start, int len, int flags,
		struct page **pages, struct vm_area_struct **vmas)
L
Linus Torvalds 已提交
161
{
162
	struct vm_area_struct *vma;
163 164
	unsigned long vm_flags;
	int i;
N
Nick Piggin 已提交
165 166 167
	int write = !!(flags & GUP_FLAGS_WRITE);
	int force = !!(flags & GUP_FLAGS_FORCE);
	int ignore = !!(flags & GUP_FLAGS_IGNORE_VMA_PERMISSIONS);
168 169 170 171 172 173

	/* calculate required read or write permissions.
	 * - if 'force' is set, we only require the "MAY" flags.
	 */
	vm_flags  = write ? (VM_WRITE | VM_MAYWRITE) : (VM_READ | VM_MAYREAD);
	vm_flags &= force ? (VM_MAYREAD | VM_MAYWRITE) : (VM_READ | VM_WRITE);
L
Linus Torvalds 已提交
174 175

	for (i = 0; i < len; i++) {
176
		vma = find_vma(mm, start);
177 178 179 180 181
		if (!vma)
			goto finish_or_fault;

		/* protect what we can, including chardevs */
		if (vma->vm_flags & (VM_IO | VM_PFNMAP) ||
N
Nick Piggin 已提交
182
		    (!ignore && !(vm_flags & vma->vm_flags)))
183
			goto finish_or_fault;
184

L
Linus Torvalds 已提交
185 186 187 188 189 190
		if (pages) {
			pages[i] = virt_to_page(start);
			if (pages[i])
				page_cache_get(pages[i]);
		}
		if (vmas)
191
			vmas[i] = vma;
L
Linus Torvalds 已提交
192 193
		start += PAGE_SIZE;
	}
194 195 196 197 198

	return i;

finish_or_fault:
	return i ? : -EFAULT;
L
Linus Torvalds 已提交
199
}
N
Nick Piggin 已提交
200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223


/*
 * get a list of pages in an address range belonging to the specified process
 * and indicate the VMA that covers each page
 * - this is potentially dodgy as we may end incrementing the page count of a
 *   slab page or a secondary page from a compound page
 * - don't permit access to VMAs that don't support it, such as I/O mappings
 */
int get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
	unsigned long start, int len, int write, int force,
	struct page **pages, struct vm_area_struct **vmas)
{
	int flags = 0;

	if (write)
		flags |= GUP_FLAGS_WRITE;
	if (force)
		flags |= GUP_FLAGS_FORCE;

	return __get_user_pages(tsk, mm,
				start, len, flags,
				pages, vmas);
}
224 225
EXPORT_SYMBOL(get_user_pages);

L
Linus Torvalds 已提交
226 227 228
DEFINE_RWLOCK(vmlist_lock);
struct vm_struct *vmlist;

229
void vfree(const void *addr)
L
Linus Torvalds 已提交
230 231 232
{
	kfree(addr);
}
233
EXPORT_SYMBOL(vfree);
L
Linus Torvalds 已提交
234

A
Al Viro 已提交
235
void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot)
L
Linus Torvalds 已提交
236 237
{
	/*
238 239
	 *  You can't specify __GFP_HIGHMEM with kmalloc() since kmalloc()
	 * returns only a logical address.
L
Linus Torvalds 已提交
240
	 */
241
	return kmalloc(size, (gfp_mask | __GFP_COMP) & ~__GFP_HIGHMEM);
L
Linus Torvalds 已提交
242
}
243
EXPORT_SYMBOL(__vmalloc);
L
Linus Torvalds 已提交
244

245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264
void *vmalloc_user(unsigned long size)
{
	void *ret;

	ret = __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO,
			PAGE_KERNEL);
	if (ret) {
		struct vm_area_struct *vma;

		down_write(&current->mm->mmap_sem);
		vma = find_vma(current->mm, (unsigned long)ret);
		if (vma)
			vma->vm_flags |= VM_USERMAP;
		up_write(&current->mm->mmap_sem);
	}

	return ret;
}
EXPORT_SYMBOL(vmalloc_user);

265
struct page *vmalloc_to_page(const void *addr)
L
Linus Torvalds 已提交
266 267 268
{
	return virt_to_page(addr);
}
269
EXPORT_SYMBOL(vmalloc_to_page);
L
Linus Torvalds 已提交
270

271
unsigned long vmalloc_to_pfn(const void *addr)
L
Linus Torvalds 已提交
272 273 274
{
	return page_to_pfn(virt_to_page(addr));
}
275
EXPORT_SYMBOL(vmalloc_to_pfn);
L
Linus Torvalds 已提交
276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300

long vread(char *buf, char *addr, unsigned long count)
{
	memcpy(buf, addr, count);
	return count;
}

long vwrite(char *buf, char *addr, unsigned long count)
{
	/* Don't allow overflow */
	if ((unsigned long) addr + count < count)
		count = -(unsigned long) addr;

	memcpy(addr, buf, count);
	return(count);
}

/*
 *	vmalloc  -  allocate virtually continguos memory
 *
 *	@size:		allocation size
 *
 *	Allocate enough pages to cover @size from the page level
 *	allocator and map them into continguos kernel virtual space.
 *
301
 *	For tight control over page level allocator and protection flags
L
Linus Torvalds 已提交
302 303 304 305 306 307
 *	use __vmalloc() instead.
 */
void *vmalloc(unsigned long size)
{
       return __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL);
}
308 309 310 311 312 313 314
EXPORT_SYMBOL(vmalloc);

void *vmalloc_node(unsigned long size, int node)
{
	return vmalloc(size);
}
EXPORT_SYMBOL(vmalloc_node);
L
Linus Torvalds 已提交
315

P
Paul Mundt 已提交
316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336
#ifndef PAGE_KERNEL_EXEC
# define PAGE_KERNEL_EXEC PAGE_KERNEL
#endif

/**
 *	vmalloc_exec  -  allocate virtually contiguous, executable memory
 *	@size:		allocation size
 *
 *	Kernel-internal function to allocate enough pages to cover @size
 *	the page level allocator and map them into contiguous and
 *	executable kernel virtual space.
 *
 *	For tight control over page level allocator and protection flags
 *	use __vmalloc() instead.
 */

void *vmalloc_exec(unsigned long size)
{
	return __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL_EXEC);
}

337 338
/**
 * vmalloc_32  -  allocate virtually contiguous memory (32bit addressable)
L
Linus Torvalds 已提交
339 340 341 342 343 344 345 346 347
 *	@size:		allocation size
 *
 *	Allocate enough 32bit PA addressable pages to cover @size from the
 *	page level allocator and map them into continguos kernel virtual space.
 */
void *vmalloc_32(unsigned long size)
{
	return __vmalloc(size, GFP_KERNEL, PAGE_KERNEL);
}
348 349 350 351 352 353 354 355
EXPORT_SYMBOL(vmalloc_32);

/**
 * vmalloc_32_user - allocate zeroed virtually contiguous 32bit memory
 *	@size:		allocation size
 *
 * The resulting memory area is 32bit addressable and zeroed so it can be
 * mapped to userspace without leaking data.
356 357 358
 *
 * VM_USERMAP is set on the corresponding VMA so that subsequent calls to
 * remap_vmalloc_range() are permissible.
359 360 361
 */
void *vmalloc_32_user(unsigned long size)
{
362 363 364 365 366
	/*
	 * We'll have to sort out the ZONE_DMA bits for 64-bit,
	 * but for now this can simply use vmalloc_user() directly.
	 */
	return vmalloc_user(size);
367 368
}
EXPORT_SYMBOL(vmalloc_32_user);
L
Linus Torvalds 已提交
369 370 371 372 373 374

void *vmap(struct page **pages, unsigned int count, unsigned long flags, pgprot_t prot)
{
	BUG();
	return NULL;
}
375
EXPORT_SYMBOL(vmap);
L
Linus Torvalds 已提交
376

377
void vunmap(const void *addr)
L
Linus Torvalds 已提交
378 379 380
{
	BUG();
}
381
EXPORT_SYMBOL(vunmap);
L
Linus Torvalds 已提交
382

383 384 385 386 387 388 389 390
/*
 * Implement a stub for vmalloc_sync_all() if the architecture chose not to
 * have one.
 */
void  __attribute__((weak)) vmalloc_sync_all(void)
{
}

391 392 393 394 395 396 397
int vm_insert_page(struct vm_area_struct *vma, unsigned long addr,
		   struct page *page)
{
	return -EINVAL;
}
EXPORT_SYMBOL(vm_insert_page);

L
Linus Torvalds 已提交
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
/*
 *  sys_brk() for the most part doesn't need the global kernel
 *  lock, except when an application is doing something nasty
 *  like trying to un-brk an area that has already been mapped
 *  to a regular file.  in this case, the unmapping will need
 *  to invoke file system routines that need the global lock.
 */
asmlinkage unsigned long sys_brk(unsigned long brk)
{
	struct mm_struct *mm = current->mm;

	if (brk < mm->start_brk || brk > mm->context.end_brk)
		return mm->brk;

	if (mm->brk == brk)
		return mm->brk;

	/*
	 * Always allow shrinking brk
	 */
	if (brk <= mm->brk) {
		mm->brk = brk;
		return brk;
	}

	/*
	 * Ok, looks good - let it rip.
	 */
	return mm->brk = brk;
}

429 430 431 432
/*
 * initialise the VMA and region record slabs
 */
void __init mmap_init(void)
L
Linus Torvalds 已提交
433
{
434 435 436 437 438 439
	vm_region_jar = kmem_cache_create("vm_region_jar",
					  sizeof(struct vm_region), 0,
					  SLAB_PANIC, NULL);
	vm_area_cachep = kmem_cache_create("vm_area_struct",
					   sizeof(struct vm_area_struct), 0,
					   SLAB_PANIC, NULL);
L
Linus Torvalds 已提交
440 441
}

442
/*
443 444
 * validate the region tree
 * - the caller must hold the region lock
445
 */
446 447
#ifdef CONFIG_DEBUG_NOMMU_REGIONS
static noinline void validate_nommu_regions(void)
448
{
449 450
	struct vm_region *region, *last;
	struct rb_node *p, *lastp;
451

452 453 454 455 456 457 458
	lastp = rb_first(&nommu_region_tree);
	if (!lastp)
		return;

	last = rb_entry(lastp, struct vm_region, vm_rb);
	if (unlikely(last->vm_end <= last->vm_start))
		BUG();
459 460
	if (unlikely(last->vm_top < last->vm_end))
		BUG();
461 462 463 464 465 466 467

	while ((p = rb_next(lastp))) {
		region = rb_entry(p, struct vm_region, vm_rb);
		last = rb_entry(lastp, struct vm_region, vm_rb);

		if (unlikely(region->vm_end <= region->vm_start))
			BUG();
468 469 470
		if (unlikely(region->vm_top < region->vm_end))
			BUG();
		if (unlikely(region->vm_start < last->vm_top))
471
			BUG();
472

473 474
		lastp = p;
	}
475
}
476 477 478
#else
#define validate_nommu_regions() do {} while(0)
#endif
479 480

/*
481
 * add a region into the global tree
482
 */
483
static void add_nommu_region(struct vm_region *region)
484
{
485 486
	struct vm_region *pregion;
	struct rb_node **p, *parent;
487

488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504
	validate_nommu_regions();

	BUG_ON(region->vm_start & ~PAGE_MASK);

	parent = NULL;
	p = &nommu_region_tree.rb_node;
	while (*p) {
		parent = *p;
		pregion = rb_entry(parent, struct vm_region, vm_rb);
		if (region->vm_start < pregion->vm_start)
			p = &(*p)->rb_left;
		else if (region->vm_start > pregion->vm_start)
			p = &(*p)->rb_right;
		else if (pregion == region)
			return;
		else
			BUG();
505 506
	}

507 508
	rb_link_node(&region->vm_rb, parent, p);
	rb_insert_color(&region->vm_rb, &nommu_region_tree);
509

510
	validate_nommu_regions();
511 512
}

513
/*
514
 * delete a region from the global tree
515
 */
516
static void delete_nommu_region(struct vm_region *region)
517
{
518
	BUG_ON(!nommu_region_tree.rb_node);
519

520 521 522
	validate_nommu_regions();
	rb_erase(&region->vm_rb, &nommu_region_tree);
	validate_nommu_regions();
523 524
}

525
/*
526
 * free a contiguous series of pages
527
 */
528
static void free_page_series(unsigned long from, unsigned long to)
529
{
530 531 532 533 534 535 536 537
	for (; from < to; from += PAGE_SIZE) {
		struct page *page = virt_to_page(from);

		kdebug("- free %lx", from);
		atomic_dec(&mmap_pages_allocated);
		if (page_count(page) != 1)
			kdebug("free page %p [%d]", page, page_count(page));
		put_page(page);
538 539 540
	}
}

541
/*
542 543
 * release a reference to a region
 * - the caller must hold the region semaphore, which this releases
544
 * - the region may not have been added to the tree yet, in which case vm_top
545
 *   will equal vm_start
546
 */
547 548
static void __put_nommu_region(struct vm_region *region)
	__releases(nommu_region_sem)
L
Linus Torvalds 已提交
549
{
550
	kenter("%p{%d}", region, atomic_read(&region->vm_usage));
L
Linus Torvalds 已提交
551

552
	BUG_ON(!nommu_region_tree.rb_node);
L
Linus Torvalds 已提交
553

554
	if (atomic_dec_and_test(&region->vm_usage)) {
555
		if (region->vm_top > region->vm_start)
556 557 558 559 560 561 562 563 564 565
			delete_nommu_region(region);
		up_write(&nommu_region_sem);

		if (region->vm_file)
			fput(region->vm_file);

		/* IO memory and memory shared directly out of the pagecache
		 * from ramfs/tmpfs mustn't be released here */
		if (region->vm_flags & VM_MAPPED_COPY) {
			kdebug("free series");
566
			free_page_series(region->vm_start, region->vm_top);
567 568 569 570
		}
		kmem_cache_free(vm_region_jar, region);
	} else {
		up_write(&nommu_region_sem);
L
Linus Torvalds 已提交
571
	}
572
}
L
Linus Torvalds 已提交
573

574 575 576 577 578 579 580
/*
 * release a reference to a region
 */
static void put_nommu_region(struct vm_region *region)
{
	down_write(&nommu_region_sem);
	__put_nommu_region(region);
L
Linus Torvalds 已提交
581 582
}

583
/*
584 585 586 587
 * add a VMA into a process's mm_struct in the appropriate place in the list
 * and tree and add to the address space's page tree also if not an anonymous
 * page
 * - should be called with mm->mmap_sem held writelocked
588
 */
589
static void add_vma_to_mm(struct mm_struct *mm, struct vm_area_struct *vma)
L
Linus Torvalds 已提交
590
{
591
	struct vm_area_struct *pvma, **pp;
L
Linus Torvalds 已提交
592
	struct address_space *mapping;
593 594 595 596 597 598 599 600
	struct rb_node **p, *parent;

	kenter(",%p", vma);

	BUG_ON(!vma->vm_region);

	mm->map_count++;
	vma->vm_mm = mm;
L
Linus Torvalds 已提交
601 602 603 604 605 606 607 608 609 610

	/* add the VMA to the mapping */
	if (vma->vm_file) {
		mapping = vma->vm_file->f_mapping;

		flush_dcache_mmap_lock(mapping);
		vma_prio_tree_insert(vma, &mapping->i_mmap);
		flush_dcache_mmap_unlock(mapping);
	}

611 612 613
	/* add the VMA to the tree */
	parent = NULL;
	p = &mm->mm_rb.rb_node;
L
Linus Torvalds 已提交
614 615 616 617
	while (*p) {
		parent = *p;
		pvma = rb_entry(parent, struct vm_area_struct, vm_rb);

618 619 620
		/* sort by: start addr, end addr, VMA struct addr in that order
		 * (the latter is necessary as we may get identical VMAs) */
		if (vma->vm_start < pvma->vm_start)
L
Linus Torvalds 已提交
621
			p = &(*p)->rb_left;
622
		else if (vma->vm_start > pvma->vm_start)
L
Linus Torvalds 已提交
623
			p = &(*p)->rb_right;
624 625 626 627 628 629 630 631 632 633
		else if (vma->vm_end < pvma->vm_end)
			p = &(*p)->rb_left;
		else if (vma->vm_end > pvma->vm_end)
			p = &(*p)->rb_right;
		else if (vma < pvma)
			p = &(*p)->rb_left;
		else if (vma > pvma)
			p = &(*p)->rb_right;
		else
			BUG();
L
Linus Torvalds 已提交
634 635 636
	}

	rb_link_node(&vma->vm_rb, parent, p);
637 638 639 640 641 642 643 644 645 646 647 648 649 650
	rb_insert_color(&vma->vm_rb, &mm->mm_rb);

	/* add VMA to the VMA list also */
	for (pp = &mm->mmap; (pvma = *pp); pp = &(*pp)->vm_next) {
		if (pvma->vm_start > vma->vm_start)
			break;
		if (pvma->vm_start < vma->vm_start)
			continue;
		if (pvma->vm_end < vma->vm_end)
			break;
	}

	vma->vm_next = *pp;
	*pp = vma;
L
Linus Torvalds 已提交
651 652
}

653
/*
654
 * delete a VMA from its owning mm_struct and address space
655
 */
656
static void delete_vma_from_mm(struct vm_area_struct *vma)
L
Linus Torvalds 已提交
657
{
658
	struct vm_area_struct **pp;
L
Linus Torvalds 已提交
659
	struct address_space *mapping;
660 661 662 663 664 665 666
	struct mm_struct *mm = vma->vm_mm;

	kenter("%p", vma);

	mm->map_count--;
	if (mm->mmap_cache == vma)
		mm->mmap_cache = NULL;
L
Linus Torvalds 已提交
667 668 669 670 671 672 673 674 675 676

	/* remove the VMA from the mapping */
	if (vma->vm_file) {
		mapping = vma->vm_file->f_mapping;

		flush_dcache_mmap_lock(mapping);
		vma_prio_tree_remove(vma, &mapping->i_mmap);
		flush_dcache_mmap_unlock(mapping);
	}

677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785
	/* remove from the MM's tree and list */
	rb_erase(&vma->vm_rb, &mm->mm_rb);
	for (pp = &mm->mmap; *pp; pp = &(*pp)->vm_next) {
		if (*pp == vma) {
			*pp = vma->vm_next;
			break;
		}
	}

	vma->vm_mm = NULL;
}

/*
 * destroy a VMA record
 */
static void delete_vma(struct mm_struct *mm, struct vm_area_struct *vma)
{
	kenter("%p", vma);
	if (vma->vm_ops && vma->vm_ops->close)
		vma->vm_ops->close(vma);
	if (vma->vm_file) {
		fput(vma->vm_file);
		if (vma->vm_flags & VM_EXECUTABLE)
			removed_exe_file_vma(mm);
	}
	put_nommu_region(vma->vm_region);
	kmem_cache_free(vm_area_cachep, vma);
}

/*
 * look up the first VMA in which addr resides, NULL if none
 * - should be called with mm->mmap_sem at least held readlocked
 */
struct vm_area_struct *find_vma(struct mm_struct *mm, unsigned long addr)
{
	struct vm_area_struct *vma;
	struct rb_node *n = mm->mm_rb.rb_node;

	/* check the cache first */
	vma = mm->mmap_cache;
	if (vma && vma->vm_start <= addr && vma->vm_end > addr)
		return vma;

	/* trawl the tree (there may be multiple mappings in which addr
	 * resides) */
	for (n = rb_first(&mm->mm_rb); n; n = rb_next(n)) {
		vma = rb_entry(n, struct vm_area_struct, vm_rb);
		if (vma->vm_start > addr)
			return NULL;
		if (vma->vm_end > addr) {
			mm->mmap_cache = vma;
			return vma;
		}
	}

	return NULL;
}
EXPORT_SYMBOL(find_vma);

/*
 * find a VMA
 * - we don't extend stack VMAs under NOMMU conditions
 */
struct vm_area_struct *find_extend_vma(struct mm_struct *mm, unsigned long addr)
{
	return find_vma(mm, addr);
}

/*
 * expand a stack to a given address
 * - not supported under NOMMU conditions
 */
int expand_stack(struct vm_area_struct *vma, unsigned long address)
{
	return -ENOMEM;
}

/*
 * look up the first VMA exactly that exactly matches addr
 * - should be called with mm->mmap_sem at least held readlocked
 */
static struct vm_area_struct *find_vma_exact(struct mm_struct *mm,
					     unsigned long addr,
					     unsigned long len)
{
	struct vm_area_struct *vma;
	struct rb_node *n = mm->mm_rb.rb_node;
	unsigned long end = addr + len;

	/* check the cache first */
	vma = mm->mmap_cache;
	if (vma && vma->vm_start == addr && vma->vm_end == end)
		return vma;

	/* trawl the tree (there may be multiple mappings in which addr
	 * resides) */
	for (n = rb_first(&mm->mm_rb); n; n = rb_next(n)) {
		vma = rb_entry(n, struct vm_area_struct, vm_rb);
		if (vma->vm_start < addr)
			continue;
		if (vma->vm_start > addr)
			return NULL;
		if (vma->vm_end == end) {
			mm->mmap_cache = vma;
			return vma;
		}
	}

	return NULL;
L
Linus Torvalds 已提交
786 787 788 789 790 791 792 793 794 795 796 797 798 799
}

/*
 * determine whether a mapping should be permitted and, if so, what sort of
 * mapping we're capable of supporting
 */
static int validate_mmap_request(struct file *file,
				 unsigned long addr,
				 unsigned long len,
				 unsigned long prot,
				 unsigned long flags,
				 unsigned long pgoff,
				 unsigned long *_capabilities)
{
800
	unsigned long capabilities, rlen;
L
Linus Torvalds 已提交
801 802 803 804 805 806 807 808 809 810 811 812 813 814 815
	unsigned long reqprot = prot;
	int ret;

	/* do the simple checks first */
	if (flags & MAP_FIXED || addr) {
		printk(KERN_DEBUG
		       "%d: Can't do fixed-address/overlay mmap of RAM\n",
		       current->pid);
		return -EINVAL;
	}

	if ((flags & MAP_TYPE) != MAP_PRIVATE &&
	    (flags & MAP_TYPE) != MAP_SHARED)
		return -EINVAL;

816
	if (!len)
L
Linus Torvalds 已提交
817 818
		return -EINVAL;

819
	/* Careful about overflows.. */
820 821
	rlen = PAGE_ALIGN(len);
	if (!rlen || rlen > TASK_SIZE)
822 823
		return -ENOMEM;

L
Linus Torvalds 已提交
824
	/* offset overflow? */
825
	if ((pgoff + (rlen >> PAGE_SHIFT)) < pgoff)
826
		return -EOVERFLOW;
L
Linus Torvalds 已提交
827 828 829 830 831 832 833 834 835 836 837 838 839 840 841

	if (file) {
		/* validate file mapping requests */
		struct address_space *mapping;

		/* files must support mmap */
		if (!file->f_op || !file->f_op->mmap)
			return -ENODEV;

		/* work out if what we've got could possibly be shared
		 * - we support chardevs that provide their own "memory"
		 * - we support files/blockdevs that are memory backed
		 */
		mapping = file->f_mapping;
		if (!mapping)
J
Josef Sipek 已提交
842
			mapping = file->f_path.dentry->d_inode->i_mapping;
L
Linus Torvalds 已提交
843 844 845 846 847 848 849 850

		capabilities = 0;
		if (mapping && mapping->backing_dev_info)
			capabilities = mapping->backing_dev_info->capabilities;

		if (!capabilities) {
			/* no explicit capabilities set, so assume some
			 * defaults */
J
Josef Sipek 已提交
851
			switch (file->f_path.dentry->d_inode->i_mode & S_IFMT) {
L
Linus Torvalds 已提交
852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881
			case S_IFREG:
			case S_IFBLK:
				capabilities = BDI_CAP_MAP_COPY;
				break;

			case S_IFCHR:
				capabilities =
					BDI_CAP_MAP_DIRECT |
					BDI_CAP_READ_MAP |
					BDI_CAP_WRITE_MAP;
				break;

			default:
				return -EINVAL;
			}
		}

		/* eliminate any capabilities that we can't support on this
		 * device */
		if (!file->f_op->get_unmapped_area)
			capabilities &= ~BDI_CAP_MAP_DIRECT;
		if (!file->f_op->read)
			capabilities &= ~BDI_CAP_MAP_COPY;

		if (flags & MAP_SHARED) {
			/* do checks for writing, appending and locking */
			if ((prot & PROT_WRITE) &&
			    !(file->f_mode & FMODE_WRITE))
				return -EACCES;

J
Josef Sipek 已提交
882
			if (IS_APPEND(file->f_path.dentry->d_inode) &&
L
Linus Torvalds 已提交
883 884 885
			    (file->f_mode & FMODE_WRITE))
				return -EACCES;

J
Josef Sipek 已提交
886
			if (locks_verify_locked(file->f_path.dentry->d_inode))
L
Linus Torvalds 已提交
887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916
				return -EAGAIN;

			if (!(capabilities & BDI_CAP_MAP_DIRECT))
				return -ENODEV;

			if (((prot & PROT_READ)  && !(capabilities & BDI_CAP_READ_MAP))  ||
			    ((prot & PROT_WRITE) && !(capabilities & BDI_CAP_WRITE_MAP)) ||
			    ((prot & PROT_EXEC)  && !(capabilities & BDI_CAP_EXEC_MAP))
			    ) {
				printk("MAP_SHARED not completely supported on !MMU\n");
				return -EINVAL;
			}

			/* we mustn't privatise shared mappings */
			capabilities &= ~BDI_CAP_MAP_COPY;
		}
		else {
			/* we're going to read the file into private memory we
			 * allocate */
			if (!(capabilities & BDI_CAP_MAP_COPY))
				return -ENODEV;

			/* we don't permit a private writable mapping to be
			 * shared with the backing device */
			if (prot & PROT_WRITE)
				capabilities &= ~BDI_CAP_MAP_DIRECT;
		}

		/* handle executable mappings and implied executable
		 * mappings */
J
Josef Sipek 已提交
917
		if (file->f_path.mnt->mnt_flags & MNT_NOEXEC) {
L
Linus Torvalds 已提交
918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948
			if (prot & PROT_EXEC)
				return -EPERM;
		}
		else if ((prot & PROT_READ) && !(prot & PROT_EXEC)) {
			/* handle implication of PROT_EXEC by PROT_READ */
			if (current->personality & READ_IMPLIES_EXEC) {
				if (capabilities & BDI_CAP_EXEC_MAP)
					prot |= PROT_EXEC;
			}
		}
		else if ((prot & PROT_READ) &&
			 (prot & PROT_EXEC) &&
			 !(capabilities & BDI_CAP_EXEC_MAP)
			 ) {
			/* backing file is not executable, try to copy */
			capabilities &= ~BDI_CAP_MAP_DIRECT;
		}
	}
	else {
		/* anonymous mappings are always memory backed and can be
		 * privately mapped
		 */
		capabilities = BDI_CAP_MAP_COPY;

		/* handle PROT_EXEC implication by PROT_READ */
		if ((prot & PROT_READ) &&
		    (current->personality & READ_IMPLIES_EXEC))
			prot |= PROT_EXEC;
	}

	/* allow the security API to have its say */
949
	ret = security_file_mmap(file, reqprot, prot, flags, addr, 0);
L
Linus Torvalds 已提交
950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991
	if (ret < 0)
		return ret;

	/* looks okay */
	*_capabilities = capabilities;
	return 0;
}

/*
 * we've determined that we can make the mapping, now translate what we
 * now know into VMA flags
 */
static unsigned long determine_vm_flags(struct file *file,
					unsigned long prot,
					unsigned long flags,
					unsigned long capabilities)
{
	unsigned long vm_flags;

	vm_flags = calc_vm_prot_bits(prot) | calc_vm_flag_bits(flags);
	vm_flags |= VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC;
	/* vm_flags |= mm->def_flags; */

	if (!(capabilities & BDI_CAP_MAP_DIRECT)) {
		/* attempt to share read-only copies of mapped file chunks */
		if (file && !(prot & PROT_WRITE))
			vm_flags |= VM_MAYSHARE;
	}
	else {
		/* overlay a shareable mapping on the backing device or inode
		 * if possible - used for chardevs, ramfs/tmpfs/shmfs and
		 * romfs/cramfs */
		if (flags & MAP_SHARED)
			vm_flags |= VM_MAYSHARE | VM_SHARED;
		else if ((((vm_flags & capabilities) ^ vm_flags) & BDI_CAP_VMFLAGS) == 0)
			vm_flags |= VM_MAYSHARE;
	}

	/* refuse to let anyone share private mappings with this process if
	 * it's being traced - otherwise breakpoints set in it may interfere
	 * with another untraced process
	 */
992
	if ((flags & MAP_PRIVATE) && tracehook_expect_breakpoints(current))
L
Linus Torvalds 已提交
993 994 995 996 997 998
		vm_flags &= ~VM_MAYSHARE;

	return vm_flags;
}

/*
999 1000
 * set up a shared mapping on a file (the driver or filesystem provides and
 * pins the storage)
L
Linus Torvalds 已提交
1001
 */
1002
static int do_mmap_shared_file(struct vm_area_struct *vma)
L
Linus Torvalds 已提交
1003 1004 1005 1006
{
	int ret;

	ret = vma->vm_file->f_op->mmap(vma->vm_file, vma);
1007 1008 1009 1010
	if (ret == 0) {
		vma->vm_region->vm_top = vma->vm_region->vm_end;
		return ret;
	}
L
Linus Torvalds 已提交
1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023
	if (ret != -ENOSYS)
		return ret;

	/* getting an ENOSYS error indicates that direct mmap isn't
	 * possible (as opposed to tried but failed) so we'll fall
	 * through to making a private copy of the data and mapping
	 * that if we can */
	return -ENODEV;
}

/*
 * set up a private mapping or an anonymous shared mapping
 */
1024 1025 1026
static int do_mmap_private(struct vm_area_struct *vma,
			   struct vm_region *region,
			   unsigned long len)
L
Linus Torvalds 已提交
1027
{
1028 1029
	struct page *pages;
	unsigned long total, point, n, rlen;
L
Linus Torvalds 已提交
1030
	void *base;
1031
	int ret, order;
L
Linus Torvalds 已提交
1032 1033 1034 1035 1036 1037 1038

	/* invoke the file's mapping function so that it can keep track of
	 * shared mappings on devices or memory
	 * - VM_MAYSHARE will be set if it may attempt to share
	 */
	if (vma->vm_file) {
		ret = vma->vm_file->f_op->mmap(vma->vm_file, vma);
1039
		if (ret == 0) {
L
Linus Torvalds 已提交
1040
			/* shouldn't return success if we're not sharing */
1041 1042 1043
			BUG_ON(!(vma->vm_flags & VM_MAYSHARE));
			vma->vm_region->vm_top = vma->vm_region->vm_end;
			return ret;
L
Linus Torvalds 已提交
1044
		}
1045 1046
		if (ret != -ENOSYS)
			return ret;
L
Linus Torvalds 已提交
1047 1048 1049 1050 1051 1052

		/* getting an ENOSYS error indicates that direct mmap isn't
		 * possible (as opposed to tried but failed) so we'll try to
		 * make a private copy of the data and map that instead */
	}

1053 1054
	rlen = PAGE_ALIGN(len);

L
Linus Torvalds 已提交
1055 1056 1057 1058
	/* allocate some memory to hold the mapping
	 * - note that this may not return a page-aligned address if the object
	 *   we're allocating is smaller than a page
	 */
1059 1060 1061 1062 1063
	order = get_order(rlen);
	kdebug("alloc order %d for %lx", order, len);

	pages = alloc_pages(GFP_KERNEL, order);
	if (!pages)
L
Linus Torvalds 已提交
1064 1065
		goto enomem;

1066 1067 1068 1069
	total = 1 << order;
	atomic_add(total, &mmap_pages_allocated);

	point = rlen >> PAGE_SHIFT;
1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082

	/* we allocated a power-of-2 sized page set, so we may want to trim off
	 * the excess */
	if (sysctl_nr_trim_pages && total - point >= sysctl_nr_trim_pages) {
		while (total > point) {
			order = ilog2(total - point);
			n = 1 << order;
			kdebug("shave %lu/%lu @%lu", n, total - point, total);
			atomic_sub(n, &mmap_pages_allocated);
			total -= n;
			set_page_refcounted(pages + total);
			__free_pages(pages + total, order);
		}
1083 1084 1085 1086
	}

	for (point = 1; point < total; point++)
		set_page_refcounted(&pages[point]);
L
Linus Torvalds 已提交
1087

1088 1089 1090 1091
	base = page_address(pages);
	region->vm_flags = vma->vm_flags |= VM_MAPPED_COPY;
	region->vm_start = (unsigned long) base;
	region->vm_end   = region->vm_start + rlen;
1092
	region->vm_top   = region->vm_start + (total << PAGE_SHIFT);
1093 1094 1095

	vma->vm_start = region->vm_start;
	vma->vm_end   = region->vm_start + len;
L
Linus Torvalds 已提交
1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106

	if (vma->vm_file) {
		/* read the contents of a file into the copy */
		mm_segment_t old_fs;
		loff_t fpos;

		fpos = vma->vm_pgoff;
		fpos <<= PAGE_SHIFT;

		old_fs = get_fs();
		set_fs(KERNEL_DS);
1107
		ret = vma->vm_file->f_op->read(vma->vm_file, base, rlen, &fpos);
L
Linus Torvalds 已提交
1108 1109 1110 1111 1112 1113
		set_fs(old_fs);

		if (ret < 0)
			goto error_free;

		/* clear the last little bit */
1114 1115
		if (ret < rlen)
			memset(base + ret, 0, rlen - ret);
L
Linus Torvalds 已提交
1116 1117 1118

	} else {
		/* if it's an anonymous mapping, then just clear it */
1119
		memset(base, 0, rlen);
L
Linus Torvalds 已提交
1120 1121 1122 1123 1124
	}

	return 0;

error_free:
1125 1126 1127
	free_page_series(region->vm_start, region->vm_end);
	region->vm_start = vma->vm_start = 0;
	region->vm_end   = vma->vm_end = 0;
1128
	region->vm_top   = 0;
L
Linus Torvalds 已提交
1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147
	return ret;

enomem:
	printk("Allocation of length %lu from process %d failed\n",
	       len, current->pid);
	show_free_areas();
	return -ENOMEM;
}

/*
 * handle mapping creation for uClinux
 */
unsigned long do_mmap_pgoff(struct file *file,
			    unsigned long addr,
			    unsigned long len,
			    unsigned long prot,
			    unsigned long flags,
			    unsigned long pgoff)
{
1148 1149
	struct vm_area_struct *vma;
	struct vm_region *region;
L
Linus Torvalds 已提交
1150
	struct rb_node *rb;
1151
	unsigned long capabilities, vm_flags, result;
L
Linus Torvalds 已提交
1152 1153
	int ret;

1154 1155
	kenter(",%lx,%lx,%lx,%lx,%lx", addr, len, prot, flags, pgoff);

1156 1157 1158
	if (!(flags & MAP_FIXED))
		addr = round_hint_to_min(addr);

L
Linus Torvalds 已提交
1159 1160 1161 1162
	/* decide whether we should attempt the mapping, and if so what sort of
	 * mapping */
	ret = validate_mmap_request(file, addr, len, prot, flags, pgoff,
				    &capabilities);
1163 1164
	if (ret < 0) {
		kleave(" = %d [val]", ret);
L
Linus Torvalds 已提交
1165
		return ret;
1166
	}
L
Linus Torvalds 已提交
1167 1168 1169 1170 1171

	/* we've determined that we can make the mapping, now translate what we
	 * now know into VMA flags */
	vm_flags = determine_vm_flags(file, prot, flags, capabilities);

1172 1173 1174 1175 1176 1177 1178 1179
	/* we're going to need to record the mapping */
	region = kmem_cache_zalloc(vm_region_jar, GFP_KERNEL);
	if (!region)
		goto error_getting_region;

	vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL);
	if (!vma)
		goto error_getting_vma;
L
Linus Torvalds 已提交
1180

1181 1182 1183 1184 1185 1186 1187
	atomic_set(&region->vm_usage, 1);
	region->vm_flags = vm_flags;
	region->vm_pgoff = pgoff;

	INIT_LIST_HEAD(&vma->anon_vma_node);
	vma->vm_flags = vm_flags;
	vma->vm_pgoff = pgoff;
L
Linus Torvalds 已提交
1188

1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202
	if (file) {
		region->vm_file = file;
		get_file(file);
		vma->vm_file = file;
		get_file(file);
		if (vm_flags & VM_EXECUTABLE) {
			added_exe_file_vma(current->mm);
			vma->vm_mm = current->mm;
		}
	}

	down_write(&nommu_region_sem);

	/* if we want to share, we need to check for regions created by other
L
Linus Torvalds 已提交
1203
	 * mmap() calls that overlap with our proposed mapping
1204
	 * - we can only share with a superset match on most regular files
L
Linus Torvalds 已提交
1205 1206 1207 1208 1209 1210
	 * - shared mappings on character devices and memory backed files are
	 *   permitted to overlap inexactly as far as we are concerned for in
	 *   these cases, sharing is handled in the driver or filesystem rather
	 *   than here
	 */
	if (vm_flags & VM_MAYSHARE) {
1211 1212
		struct vm_region *pregion;
		unsigned long pglen, rpglen, pgend, rpgend, start;
L
Linus Torvalds 已提交
1213

1214 1215
		pglen = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
		pgend = pgoff + pglen;
1216

1217 1218
		for (rb = rb_first(&nommu_region_tree); rb; rb = rb_next(rb)) {
			pregion = rb_entry(rb, struct vm_region, vm_rb);
L
Linus Torvalds 已提交
1219

1220
			if (!(pregion->vm_flags & VM_MAYSHARE))
L
Linus Torvalds 已提交
1221 1222 1223
				continue;

			/* search for overlapping mappings on the same file */
1224 1225
			if (pregion->vm_file->f_path.dentry->d_inode !=
			    file->f_path.dentry->d_inode)
L
Linus Torvalds 已提交
1226 1227
				continue;

1228
			if (pregion->vm_pgoff >= pgend)
L
Linus Torvalds 已提交
1229 1230
				continue;

1231 1232 1233 1234
			rpglen = pregion->vm_end - pregion->vm_start;
			rpglen = (rpglen + PAGE_SIZE - 1) >> PAGE_SHIFT;
			rpgend = pregion->vm_pgoff + rpglen;
			if (pgoff >= rpgend)
L
Linus Torvalds 已提交
1235 1236
				continue;

1237 1238 1239 1240 1241
			/* handle inexactly overlapping matches between
			 * mappings */
			if ((pregion->vm_pgoff != pgoff || rpglen != pglen) &&
			    !(pgoff >= pregion->vm_pgoff && pgend <= rpgend)) {
				/* new mapping is not a subset of the region */
L
Linus Torvalds 已提交
1242 1243 1244 1245 1246
				if (!(capabilities & BDI_CAP_MAP_DIRECT))
					goto sharing_violation;
				continue;
			}

1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274
			/* we've found a region we can share */
			atomic_inc(&pregion->vm_usage);
			vma->vm_region = pregion;
			start = pregion->vm_start;
			start += (pgoff - pregion->vm_pgoff) << PAGE_SHIFT;
			vma->vm_start = start;
			vma->vm_end = start + len;

			if (pregion->vm_flags & VM_MAPPED_COPY) {
				kdebug("share copy");
				vma->vm_flags |= VM_MAPPED_COPY;
			} else {
				kdebug("share mmap");
				ret = do_mmap_shared_file(vma);
				if (ret < 0) {
					vma->vm_region = NULL;
					vma->vm_start = 0;
					vma->vm_end = 0;
					atomic_dec(&pregion->vm_usage);
					pregion = NULL;
					goto error_just_free;
				}
			}
			fput(region->vm_file);
			kmem_cache_free(vm_region_jar, region);
			region = pregion;
			result = start;
			goto share;
L
Linus Torvalds 已提交
1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286
		}

		/* obtain the address at which to make a shared mapping
		 * - this is the hook for quasi-memory character devices to
		 *   tell us the location of a shared mapping
		 */
		if (file && file->f_op->get_unmapped_area) {
			addr = file->f_op->get_unmapped_area(file, addr, len,
							     pgoff, flags);
			if (IS_ERR((void *) addr)) {
				ret = addr;
				if (ret != (unsigned long) -ENOSYS)
1287
					goto error_just_free;
L
Linus Torvalds 已提交
1288 1289 1290 1291 1292 1293

				/* the driver refused to tell us where to site
				 * the mapping so we'll have to attempt to copy
				 * it */
				ret = (unsigned long) -ENODEV;
				if (!(capabilities & BDI_CAP_MAP_COPY))
1294
					goto error_just_free;
L
Linus Torvalds 已提交
1295 1296

				capabilities &= ~BDI_CAP_MAP_DIRECT;
1297 1298 1299
			} else {
				vma->vm_start = region->vm_start = addr;
				vma->vm_end = region->vm_end = addr + len;
L
Linus Torvalds 已提交
1300 1301 1302 1303
			}
		}
	}

1304
	vma->vm_region = region;
L
Linus Torvalds 已提交
1305 1306 1307

	/* set up the mapping */
	if (file && vma->vm_flags & VM_SHARED)
1308
		ret = do_mmap_shared_file(vma);
L
Linus Torvalds 已提交
1309
	else
1310
		ret = do_mmap_private(vma, region, len);
L
Linus Torvalds 已提交
1311
	if (ret < 0)
1312 1313 1314
		goto error_put_region;

	add_nommu_region(region);
L
Linus Torvalds 已提交
1315 1316

	/* okay... we have a mapping; now we have to register it */
1317
	result = vma->vm_start;
L
Linus Torvalds 已提交
1318 1319 1320

	current->mm->total_vm += len >> PAGE_SHIFT;

1321 1322
share:
	add_vma_to_mm(current->mm, vma);
L
Linus Torvalds 已提交
1323

1324
	up_write(&nommu_region_sem);
L
Linus Torvalds 已提交
1325 1326

	if (prot & PROT_EXEC)
1327
		flush_icache_range(result, result + len);
L
Linus Torvalds 已提交
1328

1329 1330
	kleave(" = %lx", result);
	return result;
L
Linus Torvalds 已提交
1331

1332 1333
error_put_region:
	__put_nommu_region(region);
L
Linus Torvalds 已提交
1334
	if (vma) {
M
Matt Helsley 已提交
1335
		if (vma->vm_file) {
1336
			fput(vma->vm_file);
M
Matt Helsley 已提交
1337 1338 1339
			if (vma->vm_flags & VM_EXECUTABLE)
				removed_exe_file_vma(vma->vm_mm);
		}
1340
		kmem_cache_free(vm_area_cachep, vma);
L
Linus Torvalds 已提交
1341
	}
1342
	kleave(" = %d [pr]", ret);
L
Linus Torvalds 已提交
1343 1344
	return ret;

1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361
error_just_free:
	up_write(&nommu_region_sem);
error:
	fput(region->vm_file);
	kmem_cache_free(vm_region_jar, region);
	fput(vma->vm_file);
	if (vma->vm_flags & VM_EXECUTABLE)
		removed_exe_file_vma(vma->vm_mm);
	kmem_cache_free(vm_area_cachep, vma);
	kleave(" = %d", ret);
	return ret;

sharing_violation:
	up_write(&nommu_region_sem);
	printk(KERN_WARNING "Attempt to share mismatched mappings\n");
	ret = -EINVAL;
	goto error;
L
Linus Torvalds 已提交
1362

1363 1364 1365 1366
error_getting_vma:
	kmem_cache_free(vm_region_jar, region);
	printk(KERN_WARNING "Allocation of vma for %lu byte allocation"
	       " from process %d failed\n",
L
Linus Torvalds 已提交
1367 1368 1369 1370
	       len, current->pid);
	show_free_areas();
	return -ENOMEM;

1371 1372 1373
error_getting_region:
	printk(KERN_WARNING "Allocation of vm region for %lu byte allocation"
	       " from process %d failed\n",
L
Linus Torvalds 已提交
1374 1375 1376 1377
	       len, current->pid);
	show_free_areas();
	return -ENOMEM;
}
1378
EXPORT_SYMBOL(do_mmap_pgoff);
L
Linus Torvalds 已提交
1379 1380

/*
1381 1382
 * split a vma into two pieces at address 'addr', a new vma is allocated either
 * for the first part or the tail.
L
Linus Torvalds 已提交
1383
 */
1384 1385
int split_vma(struct mm_struct *mm, struct vm_area_struct *vma,
	      unsigned long addr, int new_below)
L
Linus Torvalds 已提交
1386
{
1387 1388 1389
	struct vm_area_struct *new;
	struct vm_region *region;
	unsigned long npages;
L
Linus Torvalds 已提交
1390

1391
	kenter("");
L
Linus Torvalds 已提交
1392

1393 1394 1395 1396 1397
	/* we're only permitted to split anonymous regions that have a single
	 * owner */
	if (vma->vm_file ||
	    atomic_read(&vma->vm_region->vm_usage) != 1)
		return -ENOMEM;
L
Linus Torvalds 已提交
1398

1399 1400
	if (mm->map_count >= sysctl_max_map_count)
		return -ENOMEM;
L
Linus Torvalds 已提交
1401

1402 1403 1404
	region = kmem_cache_alloc(vm_region_jar, GFP_KERNEL);
	if (!region)
		return -ENOMEM;
L
Linus Torvalds 已提交
1405

1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419
	new = kmem_cache_alloc(vm_area_cachep, GFP_KERNEL);
	if (!new) {
		kmem_cache_free(vm_region_jar, region);
		return -ENOMEM;
	}

	/* most fields are the same, copy all, and then fixup */
	*new = *vma;
	*region = *vma->vm_region;
	new->vm_region = region;

	npages = (addr - vma->vm_start) >> PAGE_SHIFT;

	if (new_below) {
1420
		region->vm_top = region->vm_end = new->vm_end = addr;
1421 1422 1423
	} else {
		region->vm_start = new->vm_start = addr;
		region->vm_pgoff = new->vm_pgoff += npages;
L
Linus Torvalds 已提交
1424
	}
1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436

	if (new->vm_ops && new->vm_ops->open)
		new->vm_ops->open(new);

	delete_vma_from_mm(vma);
	down_write(&nommu_region_sem);
	delete_nommu_region(vma->vm_region);
	if (new_below) {
		vma->vm_region->vm_start = vma->vm_start = addr;
		vma->vm_region->vm_pgoff = vma->vm_pgoff += npages;
	} else {
		vma->vm_region->vm_end = vma->vm_end = addr;
1437
		vma->vm_region->vm_top = addr;
1438 1439 1440 1441 1442 1443 1444
	}
	add_nommu_region(vma->vm_region);
	add_nommu_region(new->vm_region);
	up_write(&nommu_region_sem);
	add_vma_to_mm(mm, vma);
	add_vma_to_mm(mm, new);
	return 0;
L
Linus Torvalds 已提交
1445 1446
}

1447
/*
1448 1449
 * shrink a VMA by removing the specified chunk from either the beginning or
 * the end
1450
 */
1451 1452 1453
static int shrink_vma(struct mm_struct *mm,
		      struct vm_area_struct *vma,
		      unsigned long from, unsigned long to)
L
Linus Torvalds 已提交
1454
{
1455
	struct vm_region *region;
L
Linus Torvalds 已提交
1456

1457
	kenter("");
L
Linus Torvalds 已提交
1458

1459 1460 1461 1462 1463 1464 1465 1466
	/* adjust the VMA's pointers, which may reposition it in the MM's tree
	 * and list */
	delete_vma_from_mm(vma);
	if (from > vma->vm_start)
		vma->vm_end = from;
	else
		vma->vm_start = to;
	add_vma_to_mm(mm, vma);
L
Linus Torvalds 已提交
1467

1468 1469 1470 1471 1472 1473
	/* cut the backing region down to size */
	region = vma->vm_region;
	BUG_ON(atomic_read(&region->vm_usage) != 1);

	down_write(&nommu_region_sem);
	delete_nommu_region(region);
1474 1475 1476 1477
	if (from > region->vm_start) {
		to = region->vm_top;
		region->vm_top = region->vm_end = from;
	} else {
1478
		region->vm_start = to;
1479
	}
1480 1481 1482 1483 1484 1485
	add_nommu_region(region);
	up_write(&nommu_region_sem);

	free_page_series(from, to);
	return 0;
}
L
Linus Torvalds 已提交
1486

1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497
/*
 * release a mapping
 * - under NOMMU conditions the chunk to be unmapped must be backed by a single
 *   VMA, though it need not cover the whole VMA
 */
int do_munmap(struct mm_struct *mm, unsigned long start, size_t len)
{
	struct vm_area_struct *vma;
	struct rb_node *rb;
	unsigned long end = start + len;
	int ret;
L
Linus Torvalds 已提交
1498

1499
	kenter(",%lx,%zx", start, len);
L
Linus Torvalds 已提交
1500

1501 1502
	if (len == 0)
		return -EINVAL;
1503

1504 1505 1506 1507 1508 1509 1510 1511 1512
	/* find the first potentially overlapping VMA */
	vma = find_vma(mm, start);
	if (!vma) {
		printk(KERN_WARNING
		       "munmap of memory not mmapped by process %d (%s):"
		       " 0x%lx-0x%lx\n",
		       current->pid, current->comm, start, start + len - 1);
		return -EINVAL;
	}
L
Linus Torvalds 已提交
1513

1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552
	/* we're allowed to split an anonymous VMA but not a file-backed one */
	if (vma->vm_file) {
		do {
			if (start > vma->vm_start) {
				kleave(" = -EINVAL [miss]");
				return -EINVAL;
			}
			if (end == vma->vm_end)
				goto erase_whole_vma;
			rb = rb_next(&vma->vm_rb);
			vma = rb_entry(rb, struct vm_area_struct, vm_rb);
		} while (rb);
		kleave(" = -EINVAL [split file]");
		return -EINVAL;
	} else {
		/* the chunk must be a subset of the VMA found */
		if (start == vma->vm_start && end == vma->vm_end)
			goto erase_whole_vma;
		if (start < vma->vm_start || end > vma->vm_end) {
			kleave(" = -EINVAL [superset]");
			return -EINVAL;
		}
		if (start & ~PAGE_MASK) {
			kleave(" = -EINVAL [unaligned start]");
			return -EINVAL;
		}
		if (end != vma->vm_end && end & ~PAGE_MASK) {
			kleave(" = -EINVAL [unaligned split]");
			return -EINVAL;
		}
		if (start != vma->vm_start && end != vma->vm_end) {
			ret = split_vma(mm, vma, start, 1);
			if (ret < 0) {
				kleave(" = %d [split]", ret);
				return ret;
			}
		}
		return shrink_vma(mm, vma, start, end);
	}
L
Linus Torvalds 已提交
1553

1554 1555 1556 1557
erase_whole_vma:
	delete_vma_from_mm(vma);
	delete_vma(mm, vma);
	kleave(" = 0");
L
Linus Torvalds 已提交
1558 1559
	return 0;
}
1560
EXPORT_SYMBOL(do_munmap);
L
Linus Torvalds 已提交
1561

1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573
asmlinkage long sys_munmap(unsigned long addr, size_t len)
{
	int ret;
	struct mm_struct *mm = current->mm;

	down_write(&mm->mmap_sem);
	ret = do_munmap(mm, addr, len);
	up_write(&mm->mmap_sem);
	return ret;
}

/*
1574
 * release all the mappings made in a process's VM space
1575
 */
1576
void exit_mmap(struct mm_struct *mm)
L
Linus Torvalds 已提交
1577
{
1578
	struct vm_area_struct *vma;
L
Linus Torvalds 已提交
1579

1580 1581
	if (!mm)
		return;
L
Linus Torvalds 已提交
1582

1583
	kenter("");
L
Linus Torvalds 已提交
1584

1585
	mm->total_vm = 0;
L
Linus Torvalds 已提交
1586

1587 1588 1589 1590
	while ((vma = mm->mmap)) {
		mm->mmap = vma->vm_next;
		delete_vma_from_mm(vma);
		delete_vma(mm, vma);
L
Linus Torvalds 已提交
1591
	}
1592 1593

	kleave("");
L
Linus Torvalds 已提交
1594 1595 1596 1597 1598 1599 1600 1601
}

unsigned long do_brk(unsigned long addr, unsigned long len)
{
	return -ENOMEM;
}

/*
1602 1603
 * expand (or shrink) an existing mapping, potentially moving it at the same
 * time (controlled by the MREMAP_MAYMOVE flag and available VM space)
L
Linus Torvalds 已提交
1604
 *
1605
 * under NOMMU conditions, we only permit changing a mapping's size, and only
1606 1607
 * as long as it stays within the region allocated by do_mmap_private() and the
 * block is not shareable
L
Linus Torvalds 已提交
1608
 *
1609
 * MREMAP_FIXED is not supported under NOMMU conditions
L
Linus Torvalds 已提交
1610 1611 1612 1613 1614
 */
unsigned long do_mremap(unsigned long addr,
			unsigned long old_len, unsigned long new_len,
			unsigned long flags, unsigned long new_addr)
{
1615
	struct vm_area_struct *vma;
L
Linus Torvalds 已提交
1616 1617

	/* insanity checks first */
1618
	if (old_len == 0 || new_len == 0)
L
Linus Torvalds 已提交
1619 1620
		return (unsigned long) -EINVAL;

1621 1622 1623
	if (addr & ~PAGE_MASK)
		return -EINVAL;

L
Linus Torvalds 已提交
1624 1625 1626
	if (flags & MREMAP_FIXED && new_addr != addr)
		return (unsigned long) -EINVAL;

1627
	vma = find_vma_exact(current->mm, addr, old_len);
1628 1629
	if (!vma)
		return (unsigned long) -EINVAL;
L
Linus Torvalds 已提交
1630

1631
	if (vma->vm_end != vma->vm_start + old_len)
L
Linus Torvalds 已提交
1632 1633
		return (unsigned long) -EFAULT;

1634
	if (vma->vm_flags & VM_MAYSHARE)
L
Linus Torvalds 已提交
1635 1636
		return (unsigned long) -EPERM;

1637
	if (new_len > vma->vm_region->vm_end - vma->vm_region->vm_start)
L
Linus Torvalds 已提交
1638 1639 1640
		return (unsigned long) -ENOMEM;

	/* all checks complete - do it */
1641 1642 1643
	vma->vm_end = vma->vm_start + new_len;
	return vma->vm_start;
}
1644
EXPORT_SYMBOL(do_mremap);
1645

1646 1647 1648 1649
asmlinkage
unsigned long sys_mremap(unsigned long addr,
			 unsigned long old_len, unsigned long new_len,
			 unsigned long flags, unsigned long new_addr)
1650 1651 1652 1653 1654 1655 1656
{
	unsigned long ret;

	down_write(&current->mm->mmap_sem);
	ret = do_mremap(addr, old_len, new_len, flags, new_addr);
	up_write(&current->mm->mmap_sem);
	return ret;
L
Linus Torvalds 已提交
1657 1658
}

1659
struct page *follow_page(struct vm_area_struct *vma, unsigned long address,
1660
			unsigned int foll_flags)
L
Linus Torvalds 已提交
1661 1662 1663 1664 1665 1666 1667
{
	return NULL;
}

int remap_pfn_range(struct vm_area_struct *vma, unsigned long from,
		unsigned long to, unsigned long size, pgprot_t prot)
{
1668 1669
	vma->vm_start = vma->vm_pgoff << PAGE_SHIFT;
	return 0;
L
Linus Torvalds 已提交
1670
}
1671
EXPORT_SYMBOL(remap_pfn_range);
L
Linus Torvalds 已提交
1672

1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687
int remap_vmalloc_range(struct vm_area_struct *vma, void *addr,
			unsigned long pgoff)
{
	unsigned int size = vma->vm_end - vma->vm_start;

	if (!(vma->vm_flags & VM_USERMAP))
		return -EINVAL;

	vma->vm_start = (unsigned long)(addr + (pgoff << PAGE_SHIFT));
	vma->vm_end = vma->vm_start + size;

	return 0;
}
EXPORT_SYMBOL(remap_vmalloc_range);

L
Linus Torvalds 已提交
1688 1689 1690 1691 1692 1693 1694 1695 1696 1697
void swap_unplug_io_fn(struct backing_dev_info *bdi, struct page *page)
{
}

unsigned long arch_get_unmapped_area(struct file *file, unsigned long addr,
	unsigned long len, unsigned long pgoff, unsigned long flags)
{
	return -ENOMEM;
}

1698
void arch_unmap_area(struct mm_struct *mm, unsigned long addr)
L
Linus Torvalds 已提交
1699 1700 1701 1702 1703 1704 1705 1706
{
}

void unmap_mapping_range(struct address_space *mapping,
			 loff_t const holebegin, loff_t const holelen,
			 int even_cows)
{
}
1707
EXPORT_SYMBOL(unmap_mapping_range);
L
Linus Torvalds 已提交
1708

1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729
/*
 * ask for an unmapped area at which to create a mapping on a file
 */
unsigned long get_unmapped_area(struct file *file, unsigned long addr,
				unsigned long len, unsigned long pgoff,
				unsigned long flags)
{
	unsigned long (*get_area)(struct file *, unsigned long, unsigned long,
				  unsigned long, unsigned long);

	get_area = current->mm->get_unmapped_area;
	if (file && file->f_op && file->f_op->get_unmapped_area)
		get_area = file->f_op->get_unmapped_area;

	if (!get_area)
		return -ENOSYS;

	return get_area(file, addr, len, pgoff, flags);
}
EXPORT_SYMBOL(get_unmapped_area);

L
Linus Torvalds 已提交
1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745
/*
 * Check that a process has enough memory to allocate a new virtual
 * mapping. 0 means there is enough memory for the allocation to
 * succeed and -ENOMEM implies there is not.
 *
 * We currently support three overcommit policies, which are set via the
 * vm.overcommit_memory sysctl.  See Documentation/vm/overcommit-accounting
 *
 * Strict overcommit modes added 2002 Feb 26 by Alan Cox.
 * Additional code 2002 Jul 20 by Robert Love.
 *
 * cap_sys_admin is 1 if the process has admin privileges, 0 otherwise.
 *
 * Note this is a helper function intended to be used by LSMs which
 * wish to use this logic.
 */
1746
int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin)
L
Linus Torvalds 已提交
1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760
{
	unsigned long free, allowed;

	vm_acct_memory(pages);

	/*
	 * Sometimes we want to use more memory than we have
	 */
	if (sysctl_overcommit_memory == OVERCOMMIT_ALWAYS)
		return 0;

	if (sysctl_overcommit_memory == OVERCOMMIT_GUESS) {
		unsigned long n;

1761
		free = global_page_state(NR_FILE_PAGES);
L
Linus Torvalds 已提交
1762 1763 1764 1765 1766 1767 1768 1769
		free += nr_swap_pages;

		/*
		 * Any slabs which are created with the
		 * SLAB_RECLAIM_ACCOUNT flag claim to have contents
		 * which are reclaimable, under pressure.  The dentry
		 * cache and most inode caches should fall into this
		 */
1770
		free += global_page_state(NR_SLAB_RECLAIMABLE);
L
Linus Torvalds 已提交
1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785

		/*
		 * Leave the last 3% for root
		 */
		if (!cap_sys_admin)
			free -= free / 32;

		if (free > pages)
			return 0;

		/*
		 * nr_free_pages() is very expensive on large systems,
		 * only call if we're about to fail.
		 */
		n = nr_free_pages();
1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797

		/*
		 * Leave reserved pages. The pages are not for anonymous pages.
		 */
		if (n <= totalreserve_pages)
			goto error;
		else
			n -= totalreserve_pages;

		/*
		 * Leave the last 3% for root
		 */
L
Linus Torvalds 已提交
1798 1799 1800 1801 1802 1803
		if (!cap_sys_admin)
			n -= n / 32;
		free += n;

		if (free > pages)
			return 0;
1804 1805

		goto error;
L
Linus Torvalds 已提交
1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817
	}

	allowed = totalram_pages * sysctl_overcommit_ratio / 100;
	/*
	 * Leave the last 3% for root
	 */
	if (!cap_sys_admin)
		allowed -= allowed / 32;
	allowed += total_swap_pages;

	/* Don't let a single process grow too big:
	   leave 3% of the size of this process for other processes */
A
Alan Cox 已提交
1818 1819
	if (mm)
		allowed -= mm->total_vm / 32;
L
Linus Torvalds 已提交
1820

1821 1822 1823 1824
	/*
	 * cast `allowed' as a signed long because vm_committed_space
	 * sometimes has a negative value
	 */
A
Alan Cox 已提交
1825
	if (atomic_long_read(&vm_committed_space) < (long)allowed)
L
Linus Torvalds 已提交
1826
		return 0;
1827
error:
L
Linus Torvalds 已提交
1828 1829 1830 1831 1832 1833 1834 1835 1836
	vm_unacct_memory(pages);

	return -ENOMEM;
}

int in_gate_area_no_task(unsigned long addr)
{
	return 0;
}
1837

N
Nick Piggin 已提交
1838
int filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
1839 1840
{
	BUG();
N
Nick Piggin 已提交
1841
	return 0;
1842
}
1843
EXPORT_SYMBOL(filemap_fault);
1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863

/*
 * Access another process' address space.
 * - source/target buffer must be kernel space
 */
int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write)
{
	struct vm_area_struct *vma;
	struct mm_struct *mm;

	if (addr + len < addr)
		return 0;

	mm = get_task_mm(tsk);
	if (!mm)
		return 0;

	down_read(&mm->mmap_sem);

	/* the access must start within one of the target process's mappings */
1864 1865
	vma = find_vma(mm, addr);
	if (vma) {
1866 1867 1868 1869 1870
		/* don't overrun this mapping */
		if (addr + len >= vma->vm_end)
			len = vma->vm_end - addr;

		/* only read or write mappings where it is permitted */
1871
		if (write && vma->vm_flags & VM_MAYWRITE)
1872
			len -= copy_to_user((void *) addr, buf, len);
1873
		else if (!write && vma->vm_flags & VM_MAYREAD)
1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884
			len -= copy_from_user(buf, (void *) addr, len);
		else
			len = 0;
	} else {
		len = 0;
	}

	up_read(&mm->mmap_sem);
	mmput(mm);
	return len;
}