Blame view
mm/mincore.c
7.79 KB
1da177e4c
|
1 2 3 |
/* * linux/mm/mincore.c * |
2f77d1070
|
4 |
* Copyright (C) 1994-2006 Linus Torvalds |
1da177e4c
|
5 6 7 8 9 |
*/ /* * The mincore() system call. */ |
1da177e4c
|
10 |
#include <linux/pagemap.h> |
5a0e3ad6a
|
11 |
#include <linux/gfp.h> |
1da177e4c
|
12 13 14 |
#include <linux/mm.h> #include <linux/mman.h> #include <linux/syscalls.h> |
42da9cbd3
|
15 16 |
#include <linux/swap.h> #include <linux/swapops.h> |
4f16fc107
|
17 |
#include <linux/hugetlb.h> |
1da177e4c
|
18 19 20 |
#include <asm/uaccess.h> #include <asm/pgtable.h> |
f48840107
|
21 |
static void mincore_hugetlb_page_range(struct vm_area_struct *vma, |
25ef0e50c
|
22 |
unsigned long addr, unsigned long end, |
f48840107
|
23 24 25 26 |
unsigned char *vec) { #ifdef CONFIG_HUGETLB_PAGE struct hstate *h; |
f48840107
|
27 |
|
f48840107
|
28 29 30 31 32 33 34 35 36 37 38 39 |
h = hstate_vma(vma); while (1) { unsigned char present; pte_t *ptep; /* * Huge pages are always in RAM for now, but * theoretically it needs to be checked. */ ptep = huge_pte_offset(current->mm, addr & huge_page_mask(h)); present = ptep && !huge_pte_none(huge_ptep_get(ptep)); while (1) { |
25ef0e50c
|
40 41 |
*vec = present; vec++; |
f48840107
|
42 |
addr += PAGE_SIZE; |
25ef0e50c
|
43 |
if (addr == end) |
f48840107
|
44 45 46 47 48 49 50 51 52 53 |
return; /* check hugepage border */ if (!(addr & ~huge_page_mask(h))) break; } } #else BUG(); #endif } |
1da177e4c
|
54 55 56 57 58 59 |
/* * Later we can get more picky about what "in core" means precisely. * For now, simply check to see if the page is in the page cache, * and is up to date; i.e. that no page-in operation would be required * at this time if an application were to map and access this page. */ |
42da9cbd3
|
60 |
static unsigned char mincore_page(struct address_space *mapping, pgoff_t pgoff) |
1da177e4c
|
61 62 |
{ unsigned char present = 0; |
42da9cbd3
|
63 |
struct page *page; |
1da177e4c
|
64 |
|
42da9cbd3
|
65 66 67 68 |
/* * When tmpfs swaps out a page from a file, any process mapping that * file will not get a swp_entry_t in its pte, but rather it is like * any other file mapping (ie. marked !present and faulted in with |
3c18ddd16
|
69 |
* tmpfs's .fault). So swapped out tmpfs mappings are tested here. |
42da9cbd3
|
70 71 |
*/ page = find_get_page(mapping, pgoff); |
31475dd61
|
72 |
#ifdef CONFIG_SWAP |
8079b1c85
|
73 |
/* shmem/tmpfs may return swap: account for swapcache page too. */ |
31475dd61
|
74 75 76 77 78 |
if (radix_tree_exceptional_entry(page)) { swp_entry_t swap = radix_to_swp_entry(page); page = find_get_page(&swapper_space, swap.val); } #endif |
1da177e4c
|
79 80 81 82 83 84 85 |
if (page) { present = PageUptodate(page); page_cache_release(page); } return present; } |
f48840107
|
86 |
static void mincore_unmapped_range(struct vm_area_struct *vma, |
25ef0e50c
|
87 |
unsigned long addr, unsigned long end, |
f48840107
|
88 89 |
unsigned char *vec) { |
25ef0e50c
|
90 |
unsigned long nr = (end - addr) >> PAGE_SHIFT; |
f48840107
|
91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 |
int i; if (vma->vm_file) { pgoff_t pgoff; pgoff = linear_page_index(vma, addr); for (i = 0; i < nr; i++, pgoff++) vec[i] = mincore_page(vma->vm_file->f_mapping, pgoff); } else { for (i = 0; i < nr; i++) vec[i] = 0; } } static void mincore_pte_range(struct vm_area_struct *vma, pmd_t *pmd, |
25ef0e50c
|
106 |
unsigned long addr, unsigned long end, |
f48840107
|
107 108 |
unsigned char *vec) { |
25ef0e50c
|
109 |
unsigned long next; |
f48840107
|
110 111 |
spinlock_t *ptl; pte_t *ptep; |
f48840107
|
112 113 |
ptep = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); |
25ef0e50c
|
114 |
do { |
f48840107
|
115 116 |
pte_t pte = *ptep; pgoff_t pgoff; |
25ef0e50c
|
117 |
next = addr + PAGE_SIZE; |
f48840107
|
118 |
if (pte_none(pte)) |
25ef0e50c
|
119 |
mincore_unmapped_range(vma, addr, next, vec); |
f48840107
|
120 |
else if (pte_present(pte)) |
25ef0e50c
|
121 |
*vec = 1; |
f48840107
|
122 123 |
else if (pte_file(pte)) { pgoff = pte_to_pgoff(pte); |
25ef0e50c
|
124 |
*vec = mincore_page(vma->vm_file->f_mapping, pgoff); |
f48840107
|
125 126 127 128 129 |
} else { /* pte is a swap entry */ swp_entry_t entry = pte_to_swp_entry(pte); if (is_migration_entry(entry)) { /* migration entries are always uptodate */ |
25ef0e50c
|
130 |
*vec = 1; |
f48840107
|
131 132 133 |
} else { #ifdef CONFIG_SWAP pgoff = entry.val; |
25ef0e50c
|
134 |
*vec = mincore_page(&swapper_space, pgoff); |
f48840107
|
135 136 |
#else WARN_ON(1); |
25ef0e50c
|
137 |
*vec = 1; |
f48840107
|
138 139 140 |
#endif } } |
25ef0e50c
|
141 142 |
vec++; } while (ptep++, addr = next, addr != end); |
f48840107
|
143 144 |
pte_unmap_unlock(ptep - 1, ptl); } |
e48293fd7
|
145 146 147 148 149 150 151 152 153 154 |
static void mincore_pmd_range(struct vm_area_struct *vma, pud_t *pud, unsigned long addr, unsigned long end, unsigned char *vec) { unsigned long next; pmd_t *pmd; pmd = pmd_offset(pud, addr); do { next = pmd_addr_end(addr, end); |
0ca1634d4
|
155 156 157 158 159 160 161 |
if (pmd_trans_huge(*pmd)) { if (mincore_huge_pmd(vma, pmd, addr, next, vec)) { vec += (next - addr) >> PAGE_SHIFT; continue; } /* fall through */ } |
e48293fd7
|
162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 |
if (pmd_none_or_clear_bad(pmd)) mincore_unmapped_range(vma, addr, next, vec); else mincore_pte_range(vma, pmd, addr, next, vec); vec += (next - addr) >> PAGE_SHIFT; } while (pmd++, addr = next, addr != end); } static void mincore_pud_range(struct vm_area_struct *vma, pgd_t *pgd, unsigned long addr, unsigned long end, unsigned char *vec) { unsigned long next; pud_t *pud; pud = pud_offset(pgd, addr); do { next = pud_addr_end(addr, end); if (pud_none_or_clear_bad(pud)) mincore_unmapped_range(vma, addr, next, vec); else mincore_pmd_range(vma, pud, addr, next, vec); vec += (next - addr) >> PAGE_SHIFT; } while (pud++, addr = next, addr != end); } static void mincore_page_range(struct vm_area_struct *vma, unsigned long addr, unsigned long end, unsigned char *vec) { unsigned long next; pgd_t *pgd; pgd = pgd_offset(vma->vm_mm, addr); do { next = pgd_addr_end(addr, end); if (pgd_none_or_clear_bad(pgd)) mincore_unmapped_range(vma, addr, next, vec); else mincore_pud_range(vma, pgd, addr, next, vec); vec += (next - addr) >> PAGE_SHIFT; } while (pgd++, addr = next, addr != end); } |
2f77d1070
|
205 206 207 208 209 |
/* * Do a chunk of "sys_mincore()". We've already checked * all the arguments, we hold the mmap semaphore: we should * just return the amount of info we're asked for. */ |
6a60f1b35
|
210 |
static long do_mincore(unsigned long addr, unsigned long pages, unsigned char *vec) |
1da177e4c
|
211 |
{ |
6a60f1b35
|
212 |
struct vm_area_struct *vma; |
25ef0e50c
|
213 |
unsigned long end; |
1da177e4c
|
214 |
|
6a60f1b35
|
215 |
vma = find_vma(current->mm, addr); |
4fb23e439
|
216 217 |
if (!vma || addr < vma->vm_start) return -ENOMEM; |
1da177e4c
|
218 |
|
25ef0e50c
|
219 |
end = min(vma->vm_end, addr + (pages << PAGE_SHIFT)); |
6a60f1b35
|
220 |
|
4f16fc107
|
221 |
if (is_vm_hugetlb_page(vma)) { |
25ef0e50c
|
222 223 |
mincore_hugetlb_page_range(vma, addr, end, vec); return (end - addr) >> PAGE_SHIFT; |
4f16fc107
|
224 |
} |
4f16fc107
|
225 |
|
25ef0e50c
|
226 |
end = pmd_addr_end(addr, end); |
1da177e4c
|
227 |
|
e48293fd7
|
228 229 230 231 |
if (is_vm_hugetlb_page(vma)) mincore_hugetlb_page_range(vma, addr, end, vec); else mincore_page_range(vma, addr, end, vec); |
42da9cbd3
|
232 |
|
25ef0e50c
|
233 |
return (end - addr) >> PAGE_SHIFT; |
1da177e4c
|
234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 |
} /* * The mincore(2) system call. * * mincore() returns the memory residency status of the pages in the * current process's address space specified by [addr, addr + len). * The status is returned in a vector of bytes. The least significant * bit of each byte is 1 if the referenced page is in memory, otherwise * it is zero. * * Because the status of a page can change after mincore() checks it * but before it returns to the application, the returned vector may * contain stale information. Only locked pages are guaranteed to * remain in memory. * * return values: * zero - success * -EFAULT - vec points to an illegal address * -EINVAL - addr is not a multiple of PAGE_CACHE_SIZE * -ENOMEM - Addresses in the range [addr, addr + len] are * invalid for the address space of this process, or * specify one or more pages which are not currently * mapped * -EAGAIN - A kernel resource was temporarily unavailable. */ |
3480b2574
|
260 261 |
SYSCALL_DEFINE3(mincore, unsigned long, start, size_t, len, unsigned char __user *, vec) |
1da177e4c
|
262 |
{ |
2f77d1070
|
263 264 265 |
long retval; unsigned long pages; unsigned char *tmp; |
1da177e4c
|
266 |
|
2f77d1070
|
267 268 269 |
/* Check the start address: needs to be page-aligned.. */ if (start & ~PAGE_CACHE_MASK) return -EINVAL; |
1da177e4c
|
270 |
|
2f77d1070
|
271 272 273 |
/* ..and we need to be passed a valid user-space range */ if (!access_ok(VERIFY_READ, (void __user *) start, len)) return -ENOMEM; |
1da177e4c
|
274 |
|
2f77d1070
|
275 276 277 |
/* This also avoids any overflows on PAGE_CACHE_ALIGN */ pages = len >> PAGE_SHIFT; pages += (len & ~PAGE_MASK) != 0; |
1da177e4c
|
278 |
|
2f77d1070
|
279 280 |
if (!access_ok(VERIFY_WRITE, vec, pages)) return -EFAULT; |
1da177e4c
|
281 |
|
2f77d1070
|
282 283 |
tmp = (void *) __get_free_page(GFP_USER); if (!tmp) |
4fb23e439
|
284 |
return -EAGAIN; |
2f77d1070
|
285 286 287 288 289 290 291 292 |
retval = 0; while (pages) { /* * Do at most PAGE_SIZE entries per iteration, due to * the temporary buffer size. */ down_read(¤t->mm->mmap_sem); |
6a60f1b35
|
293 |
retval = do_mincore(start, min(pages, PAGE_SIZE), tmp); |
2f77d1070
|
294 295 296 297 298 299 300 |
up_read(¤t->mm->mmap_sem); if (retval <= 0) break; if (copy_to_user(vec, tmp, retval)) { retval = -EFAULT; break; |
1da177e4c
|
301 |
} |
2f77d1070
|
302 303 304 305 |
pages -= retval; vec += retval; start += retval << PAGE_SHIFT; retval = 0; |
1da177e4c
|
306 |
} |
2f77d1070
|
307 308 |
free_page((unsigned long) tmp); return retval; |
1da177e4c
|
309 |
} |