nommu.c 50.3 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) 2007-2010 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
#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>
A
Al Viro 已提交
32
#include <linux/audit.h>
L
Linus Torvalds 已提交
33 34 35 36

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

#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
55 56 57 58 59

void *high_memory;
struct page *mem_map;
unsigned long max_mapnr;
unsigned long num_physpages;
H
Hugh Dickins 已提交
60
unsigned long highest_memmap_pfn;
61
struct percpu_counter vm_committed_as;
L
Linus Torvalds 已提交
62 63 64
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;
65
int sysctl_nr_trim_pages = CONFIG_NOMMU_INITIAL_TRIM_EXCESS;
L
Linus Torvalds 已提交
66 67
int heap_stack_gap = 0;

68
atomic_long_t mmap_pages_allocated;
69

L
Linus Torvalds 已提交
70
EXPORT_SYMBOL(mem_map);
71
EXPORT_SYMBOL(num_physpages);
L
Linus Torvalds 已提交
72

73 74 75 76
/* 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
77

78
const struct vm_operations_struct generic_file_vm_ops = {
L
Linus Torvalds 已提交
79 80 81 82 83 84 85 86 87 88 89 90
};

/*
 * 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;

91 92 93 94
	/*
	 * If the object we have should not have ksize performed on it,
	 * return size of 0
	 */
95
	if (!objp || !virt_addr_valid(objp))
96 97 98 99 100 101 102 103
		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
104 105 106
	if (PageSlab(page))
		return ksize(objp);

107 108 109 110 111 112 113 114 115 116 117 118 119 120
	/*
	 * If it's not a compound page, see if we have a matching VMA
	 * region. This test is intentionally done in reverse order,
	 * so if there's no VMA, we still fall through and hand back
	 * PAGE_SIZE for 0-order pages.
	 */
	if (!PageCompound(page)) {
		struct vm_area_struct *vma;

		vma = find_vma(current->mm, (unsigned long)objp);
		if (vma)
			return vma->vm_end - vma->vm_start;
	}

121 122
	/*
	 * The ksize() function is only guaranteed to work for pointers
123
	 * returned by kmalloc(). So handle arbitrary pointers here.
124
	 */
125
	return PAGE_SIZE << compound_order(page);
L
Linus Torvalds 已提交
126 127
}

N
Nick Piggin 已提交
128
int __get_user_pages(struct task_struct *tsk, struct mm_struct *mm,
H
Hugh Dickins 已提交
129
		     unsigned long start, int nr_pages, unsigned int foll_flags,
130 131
		     struct page **pages, struct vm_area_struct **vmas,
		     int *retry)
L
Linus Torvalds 已提交
132
{
133
	struct vm_area_struct *vma;
134 135 136 137
	unsigned long vm_flags;
	int i;

	/* calculate required read or write permissions.
H
Hugh Dickins 已提交
138
	 * If FOLL_FORCE is set, we only require the "MAY" flags.
139
	 */
H
Hugh Dickins 已提交
140 141 142 143
	vm_flags  = (foll_flags & FOLL_WRITE) ?
			(VM_WRITE | VM_MAYWRITE) : (VM_READ | VM_MAYREAD);
	vm_flags &= (foll_flags & FOLL_FORCE) ?
			(VM_MAYREAD | VM_MAYWRITE) : (VM_READ | VM_WRITE);
L
Linus Torvalds 已提交
144

145
	for (i = 0; i < nr_pages; i++) {
146
		vma = find_vma(mm, start);
147 148 149 150
		if (!vma)
			goto finish_or_fault;

		/* protect what we can, including chardevs */
H
Hugh Dickins 已提交
151 152
		if ((vma->vm_flags & (VM_IO | VM_PFNMAP)) ||
		    !(vm_flags & vma->vm_flags))
153
			goto finish_or_fault;
154

L
Linus Torvalds 已提交
155 156 157 158 159 160
		if (pages) {
			pages[i] = virt_to_page(start);
			if (pages[i])
				page_cache_get(pages[i]);
		}
		if (vmas)
161
			vmas[i] = vma;
162
		start = (start + PAGE_SIZE) & PAGE_MASK;
L
Linus Torvalds 已提交
163
	}
164 165 166 167 168

	return i;

finish_or_fault:
	return i ? : -EFAULT;
L
Linus Torvalds 已提交
169
}
N
Nick Piggin 已提交
170 171 172 173 174 175 176 177 178

/*
 * 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,
179
	unsigned long start, int nr_pages, int write, int force,
N
Nick Piggin 已提交
180 181 182 183 184
	struct page **pages, struct vm_area_struct **vmas)
{
	int flags = 0;

	if (write)
H
Hugh Dickins 已提交
185
		flags |= FOLL_WRITE;
N
Nick Piggin 已提交
186
	if (force)
H
Hugh Dickins 已提交
187
		flags |= FOLL_FORCE;
N
Nick Piggin 已提交
188

189 190
	return __get_user_pages(tsk, mm, start, nr_pages, flags, pages, vmas,
				NULL);
N
Nick Piggin 已提交
191
}
192 193
EXPORT_SYMBOL(get_user_pages);

P
Paul Mundt 已提交
194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214
/**
 * follow_pfn - look up PFN at a user virtual address
 * @vma: memory mapping
 * @address: user virtual address
 * @pfn: location to store found PFN
 *
 * Only IO mappings and raw PFN mappings are allowed.
 *
 * Returns zero and the pfn at @pfn on success, -ve otherwise.
 */
int follow_pfn(struct vm_area_struct *vma, unsigned long address,
	unsigned long *pfn)
{
	if (!(vma->vm_flags & (VM_IO | VM_PFNMAP)))
		return -EINVAL;

	*pfn = address >> PAGE_SHIFT;
	return 0;
}
EXPORT_SYMBOL(follow_pfn);

L
Linus Torvalds 已提交
215 216 217
DEFINE_RWLOCK(vmlist_lock);
struct vm_struct *vmlist;

218
void vfree(const void *addr)
L
Linus Torvalds 已提交
219 220 221
{
	kfree(addr);
}
222
EXPORT_SYMBOL(vfree);
L
Linus Torvalds 已提交
223

A
Al Viro 已提交
224
void *__vmalloc(unsigned long size, gfp_t gfp_mask, pgprot_t prot)
L
Linus Torvalds 已提交
225 226
{
	/*
227 228
	 *  You can't specify __GFP_HIGHMEM with kmalloc() since kmalloc()
	 * returns only a logical address.
L
Linus Torvalds 已提交
229
	 */
230
	return kmalloc(size, (gfp_mask | __GFP_COMP) & ~__GFP_HIGHMEM);
L
Linus Torvalds 已提交
231
}
232
EXPORT_SYMBOL(__vmalloc);
L
Linus Torvalds 已提交
233

234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253
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);

254
struct page *vmalloc_to_page(const void *addr)
L
Linus Torvalds 已提交
255 256 257
{
	return virt_to_page(addr);
}
258
EXPORT_SYMBOL(vmalloc_to_page);
L
Linus Torvalds 已提交
259

260
unsigned long vmalloc_to_pfn(const void *addr)
L
Linus Torvalds 已提交
261 262 263
{
	return page_to_pfn(virt_to_page(addr));
}
264
EXPORT_SYMBOL(vmalloc_to_pfn);
L
Linus Torvalds 已提交
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

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.
 *
290
 *	For tight control over page level allocator and protection flags
L
Linus Torvalds 已提交
291 292 293 294 295 296
 *	use __vmalloc() instead.
 */
void *vmalloc(unsigned long size)
{
       return __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL);
}
297 298
EXPORT_SYMBOL(vmalloc);

299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328
/*
 *	vzalloc - allocate virtually continguos memory with zero fill
 *
 *	@size:		allocation size
 *
 *	Allocate enough pages to cover @size from the page level
 *	allocator and map them into continguos kernel virtual space.
 *	The memory allocated is set to zero.
 *
 *	For tight control over page level allocator and protection flags
 *	use __vmalloc() instead.
 */
void *vzalloc(unsigned long size)
{
	return __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO,
			PAGE_KERNEL);
}
EXPORT_SYMBOL(vzalloc);

/**
 * vmalloc_node - allocate memory on a specific node
 * @size:	allocation size
 * @node:	numa node
 *
 * Allocate enough pages to cover @size from the page level
 * allocator and map them into contiguous kernel virtual space.
 *
 * For tight control over page level allocator and protection flags
 * use __vmalloc() instead.
 */
329 330 331 332
void *vmalloc_node(unsigned long size, int node)
{
	return vmalloc(size);
}
333
EXPORT_SYMBOL(vmalloc_node);
334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351

/**
 * vzalloc_node - allocate memory on a specific node with zero fill
 * @size:	allocation size
 * @node:	numa node
 *
 * Allocate enough pages to cover @size from the page level
 * allocator and map them into contiguous kernel virtual space.
 * The memory allocated is set to zero.
 *
 * For tight control over page level allocator and protection flags
 * use __vmalloc() instead.
 */
void *vzalloc_node(unsigned long size, int node)
{
	return vzalloc(size);
}
EXPORT_SYMBOL(vzalloc_node);
L
Linus Torvalds 已提交
352

P
Paul Mundt 已提交
353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373
#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);
}

374 375
/**
 * vmalloc_32  -  allocate virtually contiguous memory (32bit addressable)
L
Linus Torvalds 已提交
376 377 378 379 380 381 382 383 384
 *	@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);
}
385 386 387 388 389 390 391 392
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.
393 394 395
 *
 * VM_USERMAP is set on the corresponding VMA so that subsequent calls to
 * remap_vmalloc_range() are permissible.
396 397 398
 */
void *vmalloc_32_user(unsigned long size)
{
399 400 401 402 403
	/*
	 * 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);
404 405
}
EXPORT_SYMBOL(vmalloc_32_user);
L
Linus Torvalds 已提交
406 407 408 409 410 411

void *vmap(struct page **pages, unsigned int count, unsigned long flags, pgprot_t prot)
{
	BUG();
	return NULL;
}
412
EXPORT_SYMBOL(vmap);
L
Linus Torvalds 已提交
413

414
void vunmap(const void *addr)
L
Linus Torvalds 已提交
415 416 417
{
	BUG();
}
418
EXPORT_SYMBOL(vunmap);
L
Linus Torvalds 已提交
419

420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437
void *vm_map_ram(struct page **pages, unsigned int count, int node, pgprot_t prot)
{
	BUG();
	return NULL;
}
EXPORT_SYMBOL(vm_map_ram);

void vm_unmap_ram(const void *mem, unsigned int count)
{
	BUG();
}
EXPORT_SYMBOL(vm_unmap_ram);

void vm_unmap_aliases(void)
{
}
EXPORT_SYMBOL_GPL(vm_unmap_aliases);

438 439 440 441 442 443 444 445
/*
 * Implement a stub for vmalloc_sync_all() if the architecture chose not to
 * have one.
 */
void  __attribute__((weak)) vmalloc_sync_all(void)
{
}

446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470
/**
 *	alloc_vm_area - allocate a range of kernel address space
 *	@size:		size of the area
 *
 *	Returns:	NULL on failure, vm_struct on success
 *
 *	This function reserves a range of kernel address space, and
 *	allocates pagetables to map that range.  No actual mappings
 *	are created.  If the kernel address space is not shared
 *	between processes, it syncs the pagetable across all
 *	processes.
 */
struct vm_struct *alloc_vm_area(size_t size)
{
	BUG();
	return NULL;
}
EXPORT_SYMBOL_GPL(alloc_vm_area);

void free_vm_area(struct vm_struct *area)
{
	BUG();
}
EXPORT_SYMBOL_GPL(free_vm_area);

471 472 473 474 475 476 477
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
478 479 480 481 482 483 484
/*
 *  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.
 */
485
SYSCALL_DEFINE1(brk, unsigned long, brk)
L
Linus Torvalds 已提交
486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505
{
	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.
	 */
506
	flush_icache_range(mm->brk, brk);
L
Linus Torvalds 已提交
507 508 509
	return mm->brk = brk;
}

510 511 512 513
/*
 * initialise the VMA and region record slabs
 */
void __init mmap_init(void)
L
Linus Torvalds 已提交
514
{
515 516 517 518
	int ret;

	ret = percpu_counter_init(&vm_committed_as, 0);
	VM_BUG_ON(ret);
519
	vm_region_jar = KMEM_CACHE(vm_region, SLAB_PANIC);
L
Linus Torvalds 已提交
520 521
}

522
/*
523 524
 * validate the region tree
 * - the caller must hold the region lock
525
 */
526 527
#ifdef CONFIG_DEBUG_NOMMU_REGIONS
static noinline void validate_nommu_regions(void)
528
{
529 530
	struct vm_region *region, *last;
	struct rb_node *p, *lastp;
531

532 533 534 535 536
	lastp = rb_first(&nommu_region_tree);
	if (!lastp)
		return;

	last = rb_entry(lastp, struct vm_region, vm_rb);
537 538
	BUG_ON(unlikely(last->vm_end <= last->vm_start));
	BUG_ON(unlikely(last->vm_top < last->vm_end));
539 540 541 542 543

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

544 545 546
		BUG_ON(unlikely(region->vm_end <= region->vm_start));
		BUG_ON(unlikely(region->vm_top < region->vm_end));
		BUG_ON(unlikely(region->vm_start < last->vm_top));
547

548 549
		lastp = p;
	}
550
}
551
#else
552 553 554
static void validate_nommu_regions(void)
{
}
555
#endif
556 557

/*
558
 * add a region into the global tree
559
 */
560
static void add_nommu_region(struct vm_region *region)
561
{
562 563
	struct vm_region *pregion;
	struct rb_node **p, *parent;
564

565 566 567 568 569 570 571 572 573 574 575 576 577 578 579
	validate_nommu_regions();

	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();
580 581
	}

582 583
	rb_link_node(&region->vm_rb, parent, p);
	rb_insert_color(&region->vm_rb, &nommu_region_tree);
584

585
	validate_nommu_regions();
586 587
}

588
/*
589
 * delete a region from the global tree
590
 */
591
static void delete_nommu_region(struct vm_region *region)
592
{
593
	BUG_ON(!nommu_region_tree.rb_node);
594

595 596 597
	validate_nommu_regions();
	rb_erase(&region->vm_rb, &nommu_region_tree);
	validate_nommu_regions();
598 599
}

600
/*
601
 * free a contiguous series of pages
602
 */
603
static void free_page_series(unsigned long from, unsigned long to)
604
{
605 606 607 608
	for (; from < to; from += PAGE_SIZE) {
		struct page *page = virt_to_page(from);

		kdebug("- free %lx", from);
609
		atomic_long_dec(&mmap_pages_allocated);
610
		if (page_count(page) != 1)
611 612
			kdebug("free page %p: refcount not one: %d",
			       page, page_count(page));
613
		put_page(page);
614 615 616
	}
}

617
/*
618
 * release a reference to a region
619
 * - the caller must hold the region semaphore for writing, which this releases
620
 * - the region may not have been added to the tree yet, in which case vm_top
621
 *   will equal vm_start
622
 */
623 624
static void __put_nommu_region(struct vm_region *region)
	__releases(nommu_region_sem)
L
Linus Torvalds 已提交
625
{
626
	kenter("%p{%d}", region, region->vm_usage);
L
Linus Torvalds 已提交
627

628
	BUG_ON(!nommu_region_tree.rb_node);
L
Linus Torvalds 已提交
629

630
	if (--region->vm_usage == 0) {
631
		if (region->vm_top > region->vm_start)
632 633 634 635 636 637 638 639 640 641
			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");
642
			free_page_series(region->vm_start, region->vm_top);
643 644 645 646
		}
		kmem_cache_free(vm_region_jar, region);
	} else {
		up_write(&nommu_region_sem);
L
Linus Torvalds 已提交
647
	}
648
}
L
Linus Torvalds 已提交
649

650 651 652 653 654 655 656
/*
 * 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
657 658
}

659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674
/*
 * update protection on a vma
 */
static void protect_vma(struct vm_area_struct *vma, unsigned long flags)
{
#ifdef CONFIG_MPU
	struct mm_struct *mm = vma->vm_mm;
	long start = vma->vm_start & PAGE_MASK;
	while (start < vma->vm_end) {
		protect_page(mm, start, flags);
		start += PAGE_SIZE;
	}
	update_protections(mm);
#endif
}

675
/*
676 677 678 679
 * 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
680
 */
681
static void add_vma_to_mm(struct mm_struct *mm, struct vm_area_struct *vma)
L
Linus Torvalds 已提交
682
{
683
	struct vm_area_struct *pvma, **pp, *next;
L
Linus Torvalds 已提交
684
	struct address_space *mapping;
685 686 687 688 689 690 691 692
	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
693

694 695
	protect_vma(vma, vma->vm_flags);

L
Linus Torvalds 已提交
696 697 698 699 700 701 702 703 704
	/* 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);
	}

705 706 707
	/* add the VMA to the tree */
	parent = NULL;
	p = &mm->mm_rb.rb_node;
L
Linus Torvalds 已提交
708 709 710 711
	while (*p) {
		parent = *p;
		pvma = rb_entry(parent, struct vm_area_struct, vm_rb);

712 713 714
		/* 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
715
			p = &(*p)->rb_left;
716
		else if (vma->vm_start > pvma->vm_start)
L
Linus Torvalds 已提交
717
			p = &(*p)->rb_right;
718 719 720 721 722 723 724 725 726 727
		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
728 729 730
	}

	rb_link_node(&vma->vm_rb, parent, p);
731 732 733 734 735 736 737 738 739 740 741 742
	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;
	}

743
	next = *pp;
744
	*pp = vma;
745 746 747
	vma->vm_next = next;
	if (next)
		next->vm_prev = vma;
L
Linus Torvalds 已提交
748 749
}

750
/*
751
 * delete a VMA from its owning mm_struct and address space
752
 */
753
static void delete_vma_from_mm(struct vm_area_struct *vma)
L
Linus Torvalds 已提交
754
{
755
	struct vm_area_struct **pp;
L
Linus Torvalds 已提交
756
	struct address_space *mapping;
757 758 759 760
	struct mm_struct *mm = vma->vm_mm;

	kenter("%p", vma);

761 762
	protect_vma(vma, 0);

763 764 765
	mm->map_count--;
	if (mm->mmap_cache == vma)
		mm->mmap_cache = NULL;
L
Linus Torvalds 已提交
766 767 768 769 770 771 772 773 774 775

	/* 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);
	}

776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840
	/* 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)
{
841
	return find_vma(mm, addr);
842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884
}

/*
 * 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
885 886 887 888 889 890 891 892 893 894 895 896 897 898
}

/*
 * 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)
{
899
	unsigned long capabilities, rlen;
L
Linus Torvalds 已提交
900 901 902 903
	unsigned long reqprot = prot;
	int ret;

	/* do the simple checks first */
904
	if (flags & MAP_FIXED) {
L
Linus Torvalds 已提交
905 906 907 908 909 910 911 912 913 914
		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;

915
	if (!len)
L
Linus Torvalds 已提交
916 917
		return -EINVAL;

918
	/* Careful about overflows.. */
919 920
	rlen = PAGE_ALIGN(len);
	if (!rlen || rlen > TASK_SIZE)
921 922
		return -ENOMEM;

L
Linus Torvalds 已提交
923
	/* offset overflow? */
924
	if ((pgoff + (rlen >> PAGE_SHIFT)) < pgoff)
925
		return -EOVERFLOW;
L
Linus Torvalds 已提交
926 927 928 929 930 931 932 933 934 935 936 937 938 939 940

	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
941
			mapping = file->f_path.dentry->d_inode->i_mapping;
L
Linus Torvalds 已提交
942 943 944 945 946 947 948 949

		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
950
			switch (file->f_path.dentry->d_inode->i_mode & S_IFMT) {
L
Linus Torvalds 已提交
951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974
			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;

975 976 977 978
		/* The file shall have been opened with read permission. */
		if (!(file->f_mode & FMODE_READ))
			return -EACCES;

L
Linus Torvalds 已提交
979 980 981 982 983 984
		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
985
			if (IS_APPEND(file->f_path.dentry->d_inode) &&
L
Linus Torvalds 已提交
986 987 988
			    (file->f_mode & FMODE_WRITE))
				return -EACCES;

J
Josef Sipek 已提交
989
			if (locks_verify_locked(file->f_path.dentry->d_inode))
L
Linus Torvalds 已提交
990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009
				return -EAGAIN;

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

			/* 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;
		}

1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023
		if (capabilities & BDI_CAP_MAP_DIRECT) {
			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))
			    ) {
				capabilities &= ~BDI_CAP_MAP_DIRECT;
				if (flags & MAP_SHARED) {
					printk(KERN_WARNING
					       "MAP_SHARED not completely supported on !MMU\n");
					return -EINVAL;
				}
			}
		}

L
Linus Torvalds 已提交
1024 1025
		/* handle executable mappings and implied executable
		 * mappings */
J
Josef Sipek 已提交
1026
		if (file->f_path.mnt->mnt_flags & MNT_NOEXEC) {
L
Linus Torvalds 已提交
1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057
			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 */
1058
	ret = security_file_mmap(file, reqprot, prot, flags, addr, 0);
L
Linus Torvalds 已提交
1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082
	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 |= mm->def_flags; */

	if (!(capabilities & BDI_CAP_MAP_DIRECT)) {
		/* attempt to share read-only copies of mapped file chunks */
1083
		vm_flags |= VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC;
L
Linus Torvalds 已提交
1084 1085
		if (file && !(prot & PROT_WRITE))
			vm_flags |= VM_MAYSHARE;
1086
	} else {
L
Linus Torvalds 已提交
1087 1088 1089
		/* overlay a shareable mapping on the backing device or inode
		 * if possible - used for chardevs, ramfs/tmpfs/shmfs and
		 * romfs/cramfs */
1090
		vm_flags |= VM_MAYSHARE | (capabilities & BDI_CAP_VMFLAGS);
L
Linus Torvalds 已提交
1091
		if (flags & MAP_SHARED)
1092
			vm_flags |= VM_SHARED;
L
Linus Torvalds 已提交
1093 1094 1095 1096 1097 1098
	}

	/* 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
	 */
1099
	if ((flags & MAP_PRIVATE) && tracehook_expect_breakpoints(current))
L
Linus Torvalds 已提交
1100 1101 1102 1103 1104 1105
		vm_flags &= ~VM_MAYSHARE;

	return vm_flags;
}

/*
1106 1107
 * set up a shared mapping on a file (the driver or filesystem provides and
 * pins the storage)
L
Linus Torvalds 已提交
1108
 */
1109
static int do_mmap_shared_file(struct vm_area_struct *vma)
L
Linus Torvalds 已提交
1110 1111 1112 1113
{
	int ret;

	ret = vma->vm_file->f_op->mmap(vma->vm_file, vma);
1114 1115
	if (ret == 0) {
		vma->vm_region->vm_top = vma->vm_region->vm_end;
1116
		return 0;
1117
	}
L
Linus Torvalds 已提交
1118 1119 1120
	if (ret != -ENOSYS)
		return ret;

1121 1122 1123
	/* getting -ENOSYS indicates that direct mmap isn't possible (as
	 * opposed to tried but failed) so we can only give a suitable error as
	 * it's not possible to make a private copy if MAP_SHARED was given */
L
Linus Torvalds 已提交
1124 1125 1126 1127 1128 1129
	return -ENODEV;
}

/*
 * set up a private mapping or an anonymous shared mapping
 */
1130 1131
static int do_mmap_private(struct vm_area_struct *vma,
			   struct vm_region *region,
1132 1133
			   unsigned long len,
			   unsigned long capabilities)
L
Linus Torvalds 已提交
1134
{
1135 1136
	struct page *pages;
	unsigned long total, point, n, rlen;
L
Linus Torvalds 已提交
1137
	void *base;
1138
	int ret, order;
L
Linus Torvalds 已提交
1139 1140 1141 1142 1143

	/* 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
	 */
1144
	if (capabilities & BDI_CAP_MAP_DIRECT) {
L
Linus Torvalds 已提交
1145
		ret = vma->vm_file->f_op->mmap(vma->vm_file, vma);
1146
		if (ret == 0) {
L
Linus Torvalds 已提交
1147
			/* shouldn't return success if we're not sharing */
1148 1149
			BUG_ON(!(vma->vm_flags & VM_MAYSHARE));
			vma->vm_region->vm_top = vma->vm_region->vm_end;
1150
			return 0;
L
Linus Torvalds 已提交
1151
		}
1152 1153
		if (ret != -ENOSYS)
			return ret;
L
Linus Torvalds 已提交
1154 1155 1156 1157 1158 1159

		/* 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 */
	}

1160 1161
	rlen = PAGE_ALIGN(len);

L
Linus Torvalds 已提交
1162 1163 1164 1165
	/* 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
	 */
1166 1167 1168 1169 1170
	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
1171 1172
		goto enomem;

1173
	total = 1 << order;
1174
	atomic_long_add(total, &mmap_pages_allocated);
1175 1176

	point = rlen >> PAGE_SHIFT;
1177 1178 1179 1180 1181 1182 1183 1184

	/* 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);
1185
			atomic_long_sub(n, &mmap_pages_allocated);
1186 1187 1188 1189
			total -= n;
			set_page_refcounted(pages + total);
			__free_pages(pages + total, order);
		}
1190 1191 1192 1193
	}

	for (point = 1; point < total; point++)
		set_page_refcounted(&pages[point]);
L
Linus Torvalds 已提交
1194

1195 1196 1197 1198
	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;
1199
	region->vm_top   = region->vm_start + (total << PAGE_SHIFT);
1200 1201 1202

	vma->vm_start = region->vm_start;
	vma->vm_end   = region->vm_start + len;
L
Linus Torvalds 已提交
1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213

	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);
1214
		ret = vma->vm_file->f_op->read(vma->vm_file, base, rlen, &fpos);
L
Linus Torvalds 已提交
1215 1216 1217 1218 1219 1220
		set_fs(old_fs);

		if (ret < 0)
			goto error_free;

		/* clear the last little bit */
1221 1222
		if (ret < rlen)
			memset(base + ret, 0, rlen - ret);
L
Linus Torvalds 已提交
1223 1224 1225 1226 1227 1228

	}

	return 0;

error_free:
1229 1230 1231
	free_page_series(region->vm_start, region->vm_end);
	region->vm_start = vma->vm_start = 0;
	region->vm_end   = vma->vm_end = 0;
1232
	region->vm_top   = 0;
L
Linus Torvalds 已提交
1233 1234 1235
	return ret;

enomem:
1236 1237
	printk("Allocation of length %lu from process %d (%s) failed\n",
	       len, current->pid, current->comm);
L
Linus Torvalds 已提交
1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251
	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)
{
1252 1253
	struct vm_area_struct *vma;
	struct vm_region *region;
L
Linus Torvalds 已提交
1254
	struct rb_node *rb;
1255
	unsigned long capabilities, vm_flags, result;
L
Linus Torvalds 已提交
1256 1257
	int ret;

1258 1259
	kenter(",%lx,%lx,%lx,%lx,%lx", addr, len, prot, flags, pgoff);

L
Linus Torvalds 已提交
1260 1261 1262 1263
	/* 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);
1264 1265
	if (ret < 0) {
		kleave(" = %d [val]", ret);
L
Linus Torvalds 已提交
1266
		return ret;
1267
	}
L
Linus Torvalds 已提交
1268

1269 1270 1271
	/* we ignore the address hint */
	addr = 0;

L
Linus Torvalds 已提交
1272 1273 1274 1275
	/* 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);

1276 1277 1278 1279 1280 1281 1282 1283
	/* 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
1284

1285
	region->vm_usage = 1;
1286 1287 1288
	region->vm_flags = vm_flags;
	region->vm_pgoff = pgoff;

1289
	INIT_LIST_HEAD(&vma->anon_vma_chain);
1290 1291
	vma->vm_flags = vm_flags;
	vma->vm_pgoff = pgoff;
L
Linus Torvalds 已提交
1292

1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306
	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
1307
	 * mmap() calls that overlap with our proposed mapping
1308
	 * - we can only share with a superset match on most regular files
L
Linus Torvalds 已提交
1309 1310 1311 1312 1313 1314
	 * - 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) {
1315 1316
		struct vm_region *pregion;
		unsigned long pglen, rpglen, pgend, rpgend, start;
L
Linus Torvalds 已提交
1317

1318 1319
		pglen = (len + PAGE_SIZE - 1) >> PAGE_SHIFT;
		pgend = pgoff + pglen;
1320

1321 1322
		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
1323

1324
			if (!(pregion->vm_flags & VM_MAYSHARE))
L
Linus Torvalds 已提交
1325 1326 1327
				continue;

			/* search for overlapping mappings on the same file */
1328 1329
			if (pregion->vm_file->f_path.dentry->d_inode !=
			    file->f_path.dentry->d_inode)
L
Linus Torvalds 已提交
1330 1331
				continue;

1332
			if (pregion->vm_pgoff >= pgend)
L
Linus Torvalds 已提交
1333 1334
				continue;

1335 1336 1337 1338
			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
1339 1340
				continue;

1341 1342 1343 1344 1345
			/* 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
1346 1347 1348 1349 1350
				if (!(capabilities & BDI_CAP_MAP_DIRECT))
					goto sharing_violation;
				continue;
			}

1351
			/* we've found a region we can share */
1352
			pregion->vm_usage++;
1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368
			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;
1369
					pregion->vm_usage--;
1370 1371 1372 1373 1374 1375 1376 1377 1378
					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
1379 1380 1381 1382 1383 1384
		}

		/* 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
		 */
1385
		if (capabilities & BDI_CAP_MAP_DIRECT) {
L
Linus Torvalds 已提交
1386 1387 1388 1389 1390
			addr = file->f_op->get_unmapped_area(file, addr, len,
							     pgoff, flags);
			if (IS_ERR((void *) addr)) {
				ret = addr;
				if (ret != (unsigned long) -ENOSYS)
1391
					goto error_just_free;
L
Linus Torvalds 已提交
1392 1393 1394 1395 1396 1397

				/* 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))
1398
					goto error_just_free;
L
Linus Torvalds 已提交
1399 1400

				capabilities &= ~BDI_CAP_MAP_DIRECT;
1401 1402 1403
			} else {
				vma->vm_start = region->vm_start = addr;
				vma->vm_end = region->vm_end = addr + len;
L
Linus Torvalds 已提交
1404 1405 1406 1407
			}
		}
	}

1408
	vma->vm_region = region;
L
Linus Torvalds 已提交
1409

1410 1411 1412
	/* set up the mapping
	 * - the region is filled in if BDI_CAP_MAP_DIRECT is still set
	 */
L
Linus Torvalds 已提交
1413
	if (file && vma->vm_flags & VM_SHARED)
1414
		ret = do_mmap_shared_file(vma);
L
Linus Torvalds 已提交
1415
	else
1416
		ret = do_mmap_private(vma, region, len, capabilities);
L
Linus Torvalds 已提交
1417
	if (ret < 0)
1418 1419
		goto error_just_free;
	add_nommu_region(region);
1420

1421 1422 1423 1424 1425
	/* clear anonymous mappings that don't ask for uninitialized data */
	if (!vma->vm_file && !(flags & MAP_UNINITIALIZED))
		memset((void *)region->vm_start, 0,
		       region->vm_end - region->vm_start);

L
Linus Torvalds 已提交
1426
	/* okay... we have a mapping; now we have to register it */
1427
	result = vma->vm_start;
L
Linus Torvalds 已提交
1428 1429 1430

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

1431 1432
share:
	add_vma_to_mm(current->mm, vma);
L
Linus Torvalds 已提交
1433

1434 1435 1436 1437 1438 1439
	/* we flush the region from the icache only when the first executable
	 * mapping of it is made  */
	if (vma->vm_flags & VM_EXEC && !region->vm_icache_flushed) {
		flush_icache_range(region->vm_start, region->vm_end);
		region->vm_icache_flushed = true;
	}
L
Linus Torvalds 已提交
1440

1441
	up_write(&nommu_region_sem);
L
Linus Torvalds 已提交
1442

1443 1444
	kleave(" = %lx", result);
	return result;
L
Linus Torvalds 已提交
1445

1446 1447 1448
error_just_free:
	up_write(&nommu_region_sem);
error:
1449 1450
	if (region->vm_file)
		fput(region->vm_file);
1451
	kmem_cache_free(vm_region_jar, region);
1452 1453
	if (vma->vm_file)
		fput(vma->vm_file);
1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464
	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
1465

1466 1467 1468 1469
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
1470 1471 1472 1473
	       len, current->pid);
	show_free_areas();
	return -ENOMEM;

1474 1475 1476
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
1477 1478 1479 1480
	       len, current->pid);
	show_free_areas();
	return -ENOMEM;
}
1481
EXPORT_SYMBOL(do_mmap_pgoff);
L
Linus Torvalds 已提交
1482

H
Hugh Dickins 已提交
1483 1484 1485 1486 1487 1488 1489
SYSCALL_DEFINE6(mmap_pgoff, unsigned long, addr, unsigned long, len,
		unsigned long, prot, unsigned long, flags,
		unsigned long, fd, unsigned long, pgoff)
{
	struct file *file = NULL;
	unsigned long retval = -EBADF;

A
Al Viro 已提交
1490
	audit_mmap_fd(fd, flags);
H
Hugh Dickins 已提交
1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508
	if (!(flags & MAP_ANONYMOUS)) {
		file = fget(fd);
		if (!file)
			goto out;
	}

	flags &= ~(MAP_EXECUTABLE | MAP_DENYWRITE);

	down_write(&current->mm->mmap_sem);
	retval = do_mmap_pgoff(file, addr, len, prot, flags, pgoff);
	up_write(&current->mm->mmap_sem);

	if (file)
		fput(file);
out:
	return retval;
}

C
Christoph Hellwig 已提交
1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532
#ifdef __ARCH_WANT_SYS_OLD_MMAP
struct mmap_arg_struct {
	unsigned long addr;
	unsigned long len;
	unsigned long prot;
	unsigned long flags;
	unsigned long fd;
	unsigned long offset;
};

SYSCALL_DEFINE1(old_mmap, struct mmap_arg_struct __user *, arg)
{
	struct mmap_arg_struct a;

	if (copy_from_user(&a, arg, sizeof(a)))
		return -EFAULT;
	if (a.offset & ~PAGE_MASK)
		return -EINVAL;

	return sys_mmap_pgoff(a.addr, a.len, a.prot, a.flags, a.fd,
			      a.offset >> PAGE_SHIFT);
}
#endif /* __ARCH_WANT_SYS_OLD_MMAP */

L
Linus Torvalds 已提交
1533
/*
1534 1535
 * 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
1536
 */
1537 1538
int split_vma(struct mm_struct *mm, struct vm_area_struct *vma,
	      unsigned long addr, int new_below)
L
Linus Torvalds 已提交
1539
{
1540 1541 1542
	struct vm_area_struct *new;
	struct vm_region *region;
	unsigned long npages;
L
Linus Torvalds 已提交
1543

1544
	kenter("");
L
Linus Torvalds 已提交
1545

1546 1547 1548
	/* we're only permitted to split anonymous regions (these should have
	 * only a single usage on the region) */
	if (vma->vm_file)
1549
		return -ENOMEM;
L
Linus Torvalds 已提交
1550

1551 1552
	if (mm->map_count >= sysctl_max_map_count)
		return -ENOMEM;
L
Linus Torvalds 已提交
1553

1554 1555 1556
	region = kmem_cache_alloc(vm_region_jar, GFP_KERNEL);
	if (!region)
		return -ENOMEM;
L
Linus Torvalds 已提交
1557

1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571
	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) {
1572
		region->vm_top = region->vm_end = new->vm_end = addr;
1573 1574 1575
	} else {
		region->vm_start = new->vm_start = addr;
		region->vm_pgoff = new->vm_pgoff += npages;
L
Linus Torvalds 已提交
1576
	}
1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588

	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;
1589
		vma->vm_region->vm_top = addr;
1590 1591 1592 1593 1594 1595 1596
	}
	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
1597 1598
}

1599
/*
1600 1601
 * shrink a VMA by removing the specified chunk from either the beginning or
 * the end
1602
 */
1603 1604 1605
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
1606
{
1607
	struct vm_region *region;
L
Linus Torvalds 已提交
1608

1609
	kenter("");
L
Linus Torvalds 已提交
1610

1611 1612 1613 1614 1615 1616 1617 1618
	/* 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
1619

1620 1621
	/* cut the backing region down to size */
	region = vma->vm_region;
1622
	BUG_ON(region->vm_usage != 1);
1623 1624 1625

	down_write(&nommu_region_sem);
	delete_nommu_region(region);
1626 1627 1628 1629
	if (from > region->vm_start) {
		to = region->vm_top;
		region->vm_top = region->vm_end = from;
	} else {
1630
		region->vm_start = to;
1631
	}
1632 1633 1634 1635 1636 1637
	add_nommu_region(region);
	up_write(&nommu_region_sem);

	free_page_series(from, to);
	return 0;
}
L
Linus Torvalds 已提交
1638

1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649
/*
 * 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
1650

1651
	kenter(",%lx,%zx", start, len);
L
Linus Torvalds 已提交
1652

1653 1654
	if (len == 0)
		return -EINVAL;
1655

1656 1657 1658
	/* find the first potentially overlapping VMA */
	vma = find_vma(mm, start);
	if (!vma) {
1659 1660 1661 1662 1663 1664 1665 1666 1667
		static int limit = 0;
		if (limit < 5) {
			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);
			limit++;
		}
1668 1669
		return -EINVAL;
	}
L
Linus Torvalds 已提交
1670

1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707 1708 1709
	/* 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
1710

1711 1712 1713 1714
erase_whole_vma:
	delete_vma_from_mm(vma);
	delete_vma(mm, vma);
	kleave(" = 0");
L
Linus Torvalds 已提交
1715 1716
	return 0;
}
1717
EXPORT_SYMBOL(do_munmap);
L
Linus Torvalds 已提交
1718

1719
SYSCALL_DEFINE2(munmap, unsigned long, addr, size_t, len)
1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730
{
	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;
}

/*
1731
 * release all the mappings made in a process's VM space
1732
 */
1733
void exit_mmap(struct mm_struct *mm)
L
Linus Torvalds 已提交
1734
{
1735
	struct vm_area_struct *vma;
L
Linus Torvalds 已提交
1736

1737 1738
	if (!mm)
		return;
L
Linus Torvalds 已提交
1739

1740
	kenter("");
L
Linus Torvalds 已提交
1741

1742
	mm->total_vm = 0;
L
Linus Torvalds 已提交
1743

1744 1745 1746 1747
	while ((vma = mm->mmap)) {
		mm->mmap = vma->vm_next;
		delete_vma_from_mm(vma);
		delete_vma(mm, vma);
1748
		cond_resched();
L
Linus Torvalds 已提交
1749
	}
1750 1751

	kleave("");
L
Linus Torvalds 已提交
1752 1753 1754 1755 1756 1757 1758 1759
}

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

/*
1760 1761
 * 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
1762
 *
1763
 * under NOMMU conditions, we only permit changing a mapping's size, and only
1764 1765
 * 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
1766
 *
1767
 * MREMAP_FIXED is not supported under NOMMU conditions
L
Linus Torvalds 已提交
1768 1769 1770 1771 1772
 */
unsigned long do_mremap(unsigned long addr,
			unsigned long old_len, unsigned long new_len,
			unsigned long flags, unsigned long new_addr)
{
1773
	struct vm_area_struct *vma;
L
Linus Torvalds 已提交
1774 1775

	/* insanity checks first */
1776
	if (old_len == 0 || new_len == 0)
L
Linus Torvalds 已提交
1777 1778
		return (unsigned long) -EINVAL;

1779 1780 1781
	if (addr & ~PAGE_MASK)
		return -EINVAL;

L
Linus Torvalds 已提交
1782 1783 1784
	if (flags & MREMAP_FIXED && new_addr != addr)
		return (unsigned long) -EINVAL;

1785
	vma = find_vma_exact(current->mm, addr, old_len);
1786 1787
	if (!vma)
		return (unsigned long) -EINVAL;
L
Linus Torvalds 已提交
1788

1789
	if (vma->vm_end != vma->vm_start + old_len)
L
Linus Torvalds 已提交
1790 1791
		return (unsigned long) -EFAULT;

1792
	if (vma->vm_flags & VM_MAYSHARE)
L
Linus Torvalds 已提交
1793 1794
		return (unsigned long) -EPERM;

1795
	if (new_len > vma->vm_region->vm_end - vma->vm_region->vm_start)
L
Linus Torvalds 已提交
1796 1797 1798
		return (unsigned long) -ENOMEM;

	/* all checks complete - do it */
1799 1800 1801
	vma->vm_end = vma->vm_start + new_len;
	return vma->vm_start;
}
1802
EXPORT_SYMBOL(do_mremap);
1803

1804 1805 1806
SYSCALL_DEFINE5(mremap, unsigned long, addr, unsigned long, old_len,
		unsigned long, new_len, unsigned long, flags,
		unsigned long, new_addr)
1807 1808 1809 1810 1811 1812 1813
{
	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
1814 1815
}

1816
struct page *follow_page(struct vm_area_struct *vma, unsigned long address,
1817
			unsigned int foll_flags)
L
Linus Torvalds 已提交
1818 1819 1820 1821 1822 1823 1824
{
	return NULL;
}

int remap_pfn_range(struct vm_area_struct *vma, unsigned long from,
		unsigned long to, unsigned long size, pgprot_t prot)
{
1825 1826
	vma->vm_start = vma->vm_pgoff << PAGE_SHIFT;
	return 0;
L
Linus Torvalds 已提交
1827
}
1828
EXPORT_SYMBOL(remap_pfn_range);
L
Linus Torvalds 已提交
1829

1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844
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
1845 1846 1847 1848 1849 1850
unsigned long arch_get_unmapped_area(struct file *file, unsigned long addr,
	unsigned long len, unsigned long pgoff, unsigned long flags)
{
	return -ENOMEM;
}

1851
void arch_unmap_area(struct mm_struct *mm, unsigned long addr)
L
Linus Torvalds 已提交
1852 1853 1854 1855 1856 1857 1858 1859
{
}

void unmap_mapping_range(struct address_space *mapping,
			 loff_t const holebegin, loff_t const holelen,
			 int even_cows)
{
}
1860
EXPORT_SYMBOL(unmap_mapping_range);
L
Linus Torvalds 已提交
1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877

/*
 * 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.
 */
1878
int __vm_enough_memory(struct mm_struct *mm, long pages, int cap_sys_admin)
L
Linus Torvalds 已提交
1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892
{
	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;

1893
		free = global_page_state(NR_FILE_PAGES);
L
Linus Torvalds 已提交
1894 1895 1896 1897 1898 1899 1900 1901
		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
		 */
1902
		free += global_page_state(NR_SLAB_RECLAIMABLE);
L
Linus Torvalds 已提交
1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917

		/*
		 * 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();
1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929

		/*
		 * 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
1930 1931 1932 1933 1934 1935
		if (!cap_sys_admin)
			n -= n / 32;
		free += n;

		if (free > pages)
			return 0;
1936 1937

		goto error;
L
Linus Torvalds 已提交
1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949
	}

	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
1950 1951
	if (mm)
		allowed -= mm->total_vm / 32;
L
Linus Torvalds 已提交
1952

1953
	if (percpu_counter_read_positive(&vm_committed_as) < allowed)
L
Linus Torvalds 已提交
1954
		return 0;
1955

1956
error:
L
Linus Torvalds 已提交
1957 1958 1959 1960 1961
	vm_unacct_memory(pages);

	return -ENOMEM;
}

1962
int in_gate_area_no_mm(unsigned long addr)
L
Linus Torvalds 已提交
1963 1964 1965
{
	return 0;
}
1966

N
Nick Piggin 已提交
1967
int filemap_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
1968 1969
{
	BUG();
N
Nick Piggin 已提交
1970
	return 0;
1971
}
1972
EXPORT_SYMBOL(filemap_fault);
1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992

/*
 * 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 */
1993 1994
	vma = find_vma(mm, addr);
	if (vma) {
1995 1996 1997 1998 1999
		/* 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 */
2000
		if (write && vma->vm_flags & VM_MAYWRITE)
2001 2002
			copy_to_user_page(vma, NULL, addr,
					 (void *) addr, buf, len);
2003
		else if (!write && vma->vm_flags & VM_MAYREAD)
2004 2005
			copy_from_user_page(vma, NULL, addr,
					    buf, (void *) addr, len);
2006 2007 2008 2009 2010 2011 2012 2013 2014 2015
		else
			len = 0;
	} else {
		len = 0;
	}

	up_read(&mm->mmap_sem);
	mmput(mm);
	return len;
}
2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039 2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077

/**
 * nommu_shrink_inode_mappings - Shrink the shared mappings on an inode
 * @inode: The inode to check
 * @size: The current filesize of the inode
 * @newsize: The proposed filesize of the inode
 *
 * Check the shared mappings on an inode on behalf of a shrinking truncate to
 * make sure that that any outstanding VMAs aren't broken and then shrink the
 * vm_regions that extend that beyond so that do_mmap_pgoff() doesn't
 * automatically grant mappings that are too large.
 */
int nommu_shrink_inode_mappings(struct inode *inode, size_t size,
				size_t newsize)
{
	struct vm_area_struct *vma;
	struct prio_tree_iter iter;
	struct vm_region *region;
	pgoff_t low, high;
	size_t r_size, r_top;

	low = newsize >> PAGE_SHIFT;
	high = (size + PAGE_SIZE - 1) >> PAGE_SHIFT;

	down_write(&nommu_region_sem);

	/* search for VMAs that fall within the dead zone */
	vma_prio_tree_foreach(vma, &iter, &inode->i_mapping->i_mmap,
			      low, high) {
		/* found one - only interested if it's shared out of the page
		 * cache */
		if (vma->vm_flags & VM_SHARED) {
			up_write(&nommu_region_sem);
			return -ETXTBSY; /* not quite true, but near enough */
		}
	}

	/* reduce any regions that overlap the dead zone - if in existence,
	 * these will be pointed to by VMAs that don't overlap the dead zone
	 *
	 * we don't check for any regions that start beyond the EOF as there
	 * shouldn't be any
	 */
	vma_prio_tree_foreach(vma, &iter, &inode->i_mapping->i_mmap,
			      0, ULONG_MAX) {
		if (!(vma->vm_flags & VM_SHARED))
			continue;

		region = vma->vm_region;
		r_size = region->vm_top - region->vm_start;
		r_top = (region->vm_pgoff << PAGE_SHIFT) + r_size;

		if (r_top > newsize) {
			region->vm_top -= r_top - newsize;
			if (region->vm_end > region->vm_top)
				region->vm_end = region->vm_top;
		}
	}

	up_write(&nommu_region_sem);
	return 0;
}