mprotect.c 7.7 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6
/*
 *  mm/mprotect.c
 *
 *  (C) Copyright 1994 Linus Torvalds
 *  (C) Copyright 2002 Christoph Hellwig
 *
A
Alan Cox 已提交
7
 *  Address space accounting code	<alan@lxorguk.ukuu.org.uk>
L
Linus Torvalds 已提交
8 9 10 11 12 13 14 15 16 17 18 19 20 21
 *  (C) Copyright 2002 Red Hat Inc, All Rights Reserved
 */

#include <linux/mm.h>
#include <linux/hugetlb.h>
#include <linux/slab.h>
#include <linux/shm.h>
#include <linux/mman.h>
#include <linux/fs.h>
#include <linux/highmem.h>
#include <linux/security.h>
#include <linux/mempolicy.h>
#include <linux/personality.h>
#include <linux/syscalls.h>
22 23
#include <linux/swap.h>
#include <linux/swapops.h>
A
Andrea Arcangeli 已提交
24
#include <linux/mmu_notifier.h>
25
#include <linux/migrate.h>
26
#include <linux/perf_counter.h>
L
Linus Torvalds 已提交
27 28 29 30 31
#include <asm/uaccess.h>
#include <asm/pgtable.h>
#include <asm/cacheflush.h>
#include <asm/tlbflush.h>

32 33 34 35 36 37 38
#ifndef pgprot_modify
static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot)
{
	return newprot;
}
#endif

L
Linus Torvalds 已提交
39
static void change_pte_range(struct mm_struct *mm, pmd_t *pmd,
40 41
		unsigned long addr, unsigned long end, pgprot_t newprot,
		int dirty_accountable)
L
Linus Torvalds 已提交
42
{
43
	pte_t *pte, oldpte;
44
	spinlock_t *ptl;
L
Linus Torvalds 已提交
45

46
	pte = pte_offset_map_lock(mm, pmd, addr, &ptl);
47
	arch_enter_lazy_mmu_mode();
L
Linus Torvalds 已提交
48
	do {
49 50
		oldpte = *pte;
		if (pte_present(oldpte)) {
L
Linus Torvalds 已提交
51 52
			pte_t ptent;

53
			ptent = ptep_modify_prot_start(mm, addr, pte);
54
			ptent = pte_modify(ptent, newprot);
55

56 57 58 59 60 61
			/*
			 * Avoid taking write faults for pages we know to be
			 * dirty.
			 */
			if (dirty_accountable && pte_dirty(ptent))
				ptent = pte_mkwrite(ptent);
62 63

			ptep_modify_prot_commit(mm, addr, pte, ptent);
64
		} else if (PAGE_MIGRATION && !pte_file(oldpte)) {
65 66 67 68 69 70 71 72 73 74 75
			swp_entry_t entry = pte_to_swp_entry(oldpte);

			if (is_write_migration_entry(entry)) {
				/*
				 * A protection check is difficult so
				 * just be safe and disable write
				 */
				make_migration_entry_read(&entry);
				set_pte_at(mm, addr, pte,
					swp_entry_to_pte(entry));
			}
L
Linus Torvalds 已提交
76 77
		}
	} while (pte++, addr += PAGE_SIZE, addr != end);
78
	arch_leave_lazy_mmu_mode();
79
	pte_unmap_unlock(pte - 1, ptl);
L
Linus Torvalds 已提交
80 81 82
}

static inline void change_pmd_range(struct mm_struct *mm, pud_t *pud,
83 84
		unsigned long addr, unsigned long end, pgprot_t newprot,
		int dirty_accountable)
L
Linus Torvalds 已提交
85 86 87 88 89 90 91 92 93
{
	pmd_t *pmd;
	unsigned long next;

	pmd = pmd_offset(pud, addr);
	do {
		next = pmd_addr_end(addr, end);
		if (pmd_none_or_clear_bad(pmd))
			continue;
94
		change_pte_range(mm, pmd, addr, next, newprot, dirty_accountable);
L
Linus Torvalds 已提交
95 96 97 98
	} while (pmd++, addr = next, addr != end);
}

static inline void change_pud_range(struct mm_struct *mm, pgd_t *pgd,
99 100
		unsigned long addr, unsigned long end, pgprot_t newprot,
		int dirty_accountable)
L
Linus Torvalds 已提交
101 102 103 104 105 106 107 108 109
{
	pud_t *pud;
	unsigned long next;

	pud = pud_offset(pgd, addr);
	do {
		next = pud_addr_end(addr, end);
		if (pud_none_or_clear_bad(pud))
			continue;
110
		change_pmd_range(mm, pud, addr, next, newprot, dirty_accountable);
L
Linus Torvalds 已提交
111 112 113 114
	} while (pud++, addr = next, addr != end);
}

static void change_protection(struct vm_area_struct *vma,
115 116
		unsigned long addr, unsigned long end, pgprot_t newprot,
		int dirty_accountable)
L
Linus Torvalds 已提交
117 118 119 120 121 122 123 124 125 126 127 128 129
{
	struct mm_struct *mm = vma->vm_mm;
	pgd_t *pgd;
	unsigned long next;
	unsigned long start = addr;

	BUG_ON(addr >= end);
	pgd = pgd_offset(mm, addr);
	flush_cache_range(vma, addr, end);
	do {
		next = pgd_addr_end(addr, end);
		if (pgd_none_or_clear_bad(pgd))
			continue;
130
		change_pud_range(mm, pgd, addr, next, newprot, dirty_accountable);
L
Linus Torvalds 已提交
131 132 133 134
	} while (pgd++, addr = next, addr != end);
	flush_tlb_range(vma, start, end);
}

135
int
L
Linus Torvalds 已提交
136 137 138 139 140 141 142 143 144
mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev,
	unsigned long start, unsigned long end, unsigned long newflags)
{
	struct mm_struct *mm = vma->vm_mm;
	unsigned long oldflags = vma->vm_flags;
	long nrpages = (end - start) >> PAGE_SHIFT;
	unsigned long charged = 0;
	pgoff_t pgoff;
	int error;
145
	int dirty_accountable = 0;
L
Linus Torvalds 已提交
146 147 148 149 150 151 152 153 154

	if (newflags == oldflags) {
		*pprev = vma;
		return 0;
	}

	/*
	 * If we make a private mapping writable we increase our commit;
	 * but (without finer accounting) cannot reduce our commit if we
155 156
	 * make it unwritable again. hugetlb mapping were accounted for
	 * even if read-only so there is no need to account for them here
L
Linus Torvalds 已提交
157 158
	 */
	if (newflags & VM_WRITE) {
159
		if (!(oldflags & (VM_ACCOUNT|VM_WRITE|VM_HUGETLB|
160
						VM_SHARED|VM_NORESERVE))) {
L
Linus Torvalds 已提交
161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198
			charged = nrpages;
			if (security_vm_enough_memory(charged))
				return -ENOMEM;
			newflags |= VM_ACCOUNT;
		}
	}

	/*
	 * First try to merge with previous and/or next vma.
	 */
	pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT);
	*pprev = vma_merge(mm, *pprev, start, end, newflags,
			vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma));
	if (*pprev) {
		vma = *pprev;
		goto success;
	}

	*pprev = vma;

	if (start != vma->vm_start) {
		error = split_vma(mm, vma, start, 1);
		if (error)
			goto fail;
	}

	if (end != vma->vm_end) {
		error = split_vma(mm, vma, end, 0);
		if (error)
			goto fail;
	}

success:
	/*
	 * vm_flags and vm_page_prot are protected by the mmap_sem
	 * held in write mode.
	 */
	vma->vm_flags = newflags;
199 200 201
	vma->vm_page_prot = pgprot_modify(vma->vm_page_prot,
					  vm_get_page_prot(newflags));

202
	if (vma_wants_writenotify(vma)) {
203
		vma->vm_page_prot = vm_get_page_prot(newflags & ~VM_SHARED);
204 205
		dirty_accountable = 1;
	}
206

A
Andrea Arcangeli 已提交
207
	mmu_notifier_invalidate_range_start(mm, start, end);
208
	if (is_vm_hugetlb_page(vma))
209
		hugetlb_change_protection(vma, start, end, vma->vm_page_prot);
210
	else
211
		change_protection(vma, start, end, vma->vm_page_prot, dirty_accountable);
A
Andrea Arcangeli 已提交
212
	mmu_notifier_invalidate_range_end(mm, start, end);
213 214
	vm_stat_account(mm, oldflags, vma->vm_file, -nrpages);
	vm_stat_account(mm, newflags, vma->vm_file, nrpages);
L
Linus Torvalds 已提交
215 216 217 218 219 220 221
	return 0;

fail:
	vm_unacct_memory(charged);
	return error;
}

222 223
SYSCALL_DEFINE3(mprotect, unsigned long, start, size_t, len,
		unsigned long, prot)
L
Linus Torvalds 已提交
224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240
{
	unsigned long vm_flags, nstart, end, tmp, reqprot;
	struct vm_area_struct *vma, *prev;
	int error = -EINVAL;
	const int grows = prot & (PROT_GROWSDOWN|PROT_GROWSUP);
	prot &= ~(PROT_GROWSDOWN|PROT_GROWSUP);
	if (grows == (PROT_GROWSDOWN|PROT_GROWSUP)) /* can't be both */
		return -EINVAL;

	if (start & ~PAGE_MASK)
		return -EINVAL;
	if (!len)
		return 0;
	len = PAGE_ALIGN(len);
	end = start + len;
	if (end <= start)
		return -ENOMEM;
241
	if (!arch_validate_prot(prot))
L
Linus Torvalds 已提交
242 243 244 245 246 247
		return -EINVAL;

	reqprot = prot;
	/*
	 * Does the application expect PROT_READ to imply PROT_EXEC:
	 */
248
	if ((prot & PROT_READ) && (current->personality & READ_IMPLIES_EXEC))
L
Linus Torvalds 已提交
249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286
		prot |= PROT_EXEC;

	vm_flags = calc_vm_prot_bits(prot);

	down_write(&current->mm->mmap_sem);

	vma = find_vma_prev(current->mm, start, &prev);
	error = -ENOMEM;
	if (!vma)
		goto out;
	if (unlikely(grows & PROT_GROWSDOWN)) {
		if (vma->vm_start >= end)
			goto out;
		start = vma->vm_start;
		error = -EINVAL;
		if (!(vma->vm_flags & VM_GROWSDOWN))
			goto out;
	}
	else {
		if (vma->vm_start > start)
			goto out;
		if (unlikely(grows & PROT_GROWSUP)) {
			end = vma->vm_end;
			error = -EINVAL;
			if (!(vma->vm_flags & VM_GROWSUP))
				goto out;
		}
	}
	if (start > vma->vm_start)
		prev = vma;

	for (nstart = start ; ; ) {
		unsigned long newflags;

		/* Here we know that  vma->vm_start <= nstart < vma->vm_end. */

		newflags = vm_flags | (vma->vm_flags & ~(VM_READ | VM_WRITE | VM_EXEC));

287 288
		/* newflags >> 4 shift VM_MAY% in place of VM_% */
		if ((newflags & ~(newflags >> 4)) & (VM_READ | VM_WRITE | VM_EXEC)) {
L
Linus Torvalds 已提交
289 290 291 292 293 294 295 296 297 298 299 300 301 302
			error = -EACCES;
			goto out;
		}

		error = security_file_mprotect(vma, reqprot, prot);
		if (error)
			goto out;

		tmp = vma->vm_end;
		if (tmp > end)
			tmp = end;
		error = mprotect_fixup(vma, &prev, nstart, tmp, newflags);
		if (error)
			goto out;
303
		perf_counter_mmap(vma);
L
Linus Torvalds 已提交
304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320
		nstart = tmp;

		if (nstart < prev->vm_end)
			nstart = prev->vm_end;
		if (nstart >= end)
			goto out;

		vma = prev->vm_next;
		if (!vma || vma->vm_start != nstart) {
			error = -ENOMEM;
			goto out;
		}
	}
out:
	up_write(&current->mm->mmap_sem);
	return error;
}