Blame view
ipc/shm.c
27.1 KB
1da177e4c
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
/* * linux/ipc/shm.c * Copyright (C) 1992, 1993 Krishna Balasubramanian * Many improvements/fixes by Bruno Haible. * Replaced `struct shm_desc' by `struct vm_area_struct', July 1994. * Fixed the shm swap deallocation (shm_unuse()), August 1998 Andrea Arcangeli. * * /proc/sysvipc/shm support (c) 1999 Dragos Acostachioaie <dragos@iname.com> * BIGMEM support, Andrea Arcangeli <andrea@suse.de> * SMP thread shm, Jean-Luc Boyard <jean-luc.boyard@siemens.fr> * HIGHMEM support, Ingo Molnar <mingo@redhat.com> * Make shmmax, shmall, shmmni sysctl'able, Christoph Rohland <cr@sap.com> * Shared /dev/zero support, Kanoj Sarcar <kanoj@sgi.com> * Move the mm functionality over to mm/shmem.c, Christoph Rohland <cr@sap.com> * |
073115d6b
|
16 17 |
* support for audit of ipc object properties and permission changes * Dustin Kirkland <dustin.kirkland@us.ibm.com> |
4e9823111
|
18 19 20 21 |
* * namespaces support * OpenVZ, SWsoft Inc. * Pavel Emelianov <xemul@openvz.org> |
1da177e4c
|
22 |
*/ |
1da177e4c
|
23 24 25 26 27 28 29 |
#include <linux/slab.h> #include <linux/mm.h> #include <linux/hugetlb.h> #include <linux/shm.h> #include <linux/init.h> #include <linux/file.h> #include <linux/mman.h> |
1da177e4c
|
30 31 32 33 |
#include <linux/shmem_fs.h> #include <linux/security.h> #include <linux/syscalls.h> #include <linux/audit.h> |
c59ede7b7
|
34 |
#include <linux/capability.h> |
7d87e14c2
|
35 |
#include <linux/ptrace.h> |
19b4946ca
|
36 |
#include <linux/seq_file.h> |
3e148c799
|
37 |
#include <linux/rwsem.h> |
4e9823111
|
38 |
#include <linux/nsproxy.h> |
bc56bba8f
|
39 |
#include <linux/mount.h> |
ae5e1b22f
|
40 |
#include <linux/ipc_namespace.h> |
7d87e14c2
|
41 |
|
1da177e4c
|
42 43 44 |
#include <asm/uaccess.h> #include "util.h" |
bc56bba8f
|
45 46 47 48 49 50 51 52 |
struct shm_file_data { int id; struct ipc_namespace *ns; struct file *file; const struct vm_operations_struct *vm_ops; }; #define shm_file_data(file) (*((struct shm_file_data **)&(file)->private_data)) |
9a32144e9
|
53 |
static const struct file_operations shm_file_operations; |
1da177e4c
|
54 |
static struct vm_operations_struct shm_vm_ops; |
ed2ddbf88
|
55 |
#define shm_ids(ns) ((ns)->ids[IPC_SHM_IDS]) |
1da177e4c
|
56 |
|
4e9823111
|
57 58 |
#define shm_unlock(shp) \ ipc_unlock(&(shp)->shm_perm) |
1b531f213
|
59 |
#define shm_buildid(id, seq) ipc_buildid(id, seq) |
1da177e4c
|
60 |
|
7748dbfaa
|
61 |
static int newseg(struct ipc_namespace *, struct ipc_params *); |
bc56bba8f
|
62 63 |
static void shm_open(struct vm_area_struct *vma); static void shm_close(struct vm_area_struct *vma); |
4e9823111
|
64 |
static void shm_destroy (struct ipc_namespace *ns, struct shmid_kernel *shp); |
1da177e4c
|
65 |
#ifdef CONFIG_PROC_FS |
19b4946ca
|
66 |
static int sysvipc_shm_proc_show(struct seq_file *s, void *it); |
1da177e4c
|
67 |
#endif |
ed2ddbf88
|
68 |
void shm_init_ns(struct ipc_namespace *ns) |
4e9823111
|
69 |
{ |
4e9823111
|
70 71 72 73 |
ns->shm_ctlmax = SHMMAX; ns->shm_ctlall = SHMALL; ns->shm_ctlmni = SHMMNI; ns->shm_tot = 0; |
ed2ddbf88
|
74 |
ipc_init_ids(&ns->ids[IPC_SHM_IDS]); |
4e9823111
|
75 |
} |
f4566f048
|
76 |
/* |
3e148c799
|
77 78 |
* Called with shm_ids.rw_mutex (writer) and the shp structure locked. * Only shm_ids.rw_mutex remains locked on exit. |
f4566f048
|
79 |
*/ |
01b8b07a5
|
80 |
static void do_shm_rmid(struct ipc_namespace *ns, struct kern_ipc_perm *ipcp) |
4e9823111
|
81 |
{ |
01b8b07a5
|
82 83 |
struct shmid_kernel *shp; shp = container_of(ipcp, struct shmid_kernel, shm_perm); |
4e9823111
|
84 85 86 87 88 89 90 91 |
if (shp->shm_nattch){ shp->shm_perm.mode |= SHM_DEST; /* Do not find it any more */ shp->shm_perm.key = IPC_PRIVATE; shm_unlock(shp); } else shm_destroy(ns, shp); } |
ae5e1b22f
|
92 |
#ifdef CONFIG_IPC_NS |
4e9823111
|
93 94 |
void shm_exit_ns(struct ipc_namespace *ns) { |
01b8b07a5
|
95 |
free_ipcs(ns, &shm_ids(ns), do_shm_rmid); |
4e9823111
|
96 |
} |
ae5e1b22f
|
97 |
#endif |
1da177e4c
|
98 99 100 |
void __init shm_init (void) { |
ed2ddbf88
|
101 |
shm_init_ns(&init_ipc_ns); |
19b4946ca
|
102 103 104 |
ipc_init_proc_interface("sysvipc/shm", " key shmid perms size cpid lpid nattch uid gid cuid cgid atime dtime ctime ", |
4e9823111
|
105 |
IPC_SHM_IDS, sysvipc_shm_proc_show); |
1da177e4c
|
106 |
} |
3e148c799
|
107 108 109 110 111 112 113 114 |
/* * shm_lock_(check_)down routines are called in the paths where the rw_mutex * is held to protect access to the idr tree. */ static inline struct shmid_kernel *shm_lock_down(struct ipc_namespace *ns, int id) { struct kern_ipc_perm *ipcp = ipc_lock_down(&shm_ids(ns), id); |
b1ed88b47
|
115 116 |
if (IS_ERR(ipcp)) return (struct shmid_kernel *)ipcp; |
3e148c799
|
117 118 119 120 121 122 123 124 |
return container_of(ipcp, struct shmid_kernel, shm_perm); } static inline struct shmid_kernel *shm_lock_check_down( struct ipc_namespace *ns, int id) { struct kern_ipc_perm *ipcp = ipc_lock_check_down(&shm_ids(ns), id); |
b1ed88b47
|
125 126 |
if (IS_ERR(ipcp)) return (struct shmid_kernel *)ipcp; |
3e148c799
|
127 128 129 130 131 132 133 |
return container_of(ipcp, struct shmid_kernel, shm_perm); } /* * shm_lock_(check_) routines are called in the paths where the rw_mutex * is not held. */ |
023a53557
|
134 |
static inline struct shmid_kernel *shm_lock(struct ipc_namespace *ns, int id) |
1da177e4c
|
135 |
{ |
03f02c765
|
136 |
struct kern_ipc_perm *ipcp = ipc_lock(&shm_ids(ns), id); |
b1ed88b47
|
137 138 |
if (IS_ERR(ipcp)) return (struct shmid_kernel *)ipcp; |
03f02c765
|
139 |
return container_of(ipcp, struct shmid_kernel, shm_perm); |
023a53557
|
140 141 142 143 144 |
} static inline struct shmid_kernel *shm_lock_check(struct ipc_namespace *ns, int id) { |
03f02c765
|
145 |
struct kern_ipc_perm *ipcp = ipc_lock_check(&shm_ids(ns), id); |
b1ed88b47
|
146 147 |
if (IS_ERR(ipcp)) return (struct shmid_kernel *)ipcp; |
03f02c765
|
148 |
return container_of(ipcp, struct shmid_kernel, shm_perm); |
1da177e4c
|
149 |
} |
7ca7e564e
|
150 |
static inline void shm_rmid(struct ipc_namespace *ns, struct shmid_kernel *s) |
1da177e4c
|
151 |
{ |
7ca7e564e
|
152 |
ipc_rmid(&shm_ids(ns), &s->shm_perm); |
1da177e4c
|
153 |
} |
4e9823111
|
154 |
static inline int shm_addid(struct ipc_namespace *ns, struct shmid_kernel *shp) |
1da177e4c
|
155 |
{ |
4e9823111
|
156 |
return ipc_addid(&shm_ids(ns), &shp->shm_perm, ns->shm_ctlmni); |
1da177e4c
|
157 |
} |
bc56bba8f
|
158 159 |
/* This is called by fork, once for every shm attach. */ static void shm_open(struct vm_area_struct *vma) |
4e9823111
|
160 |
{ |
bc56bba8f
|
161 162 |
struct file *file = vma->vm_file; struct shm_file_data *sfd = shm_file_data(file); |
1da177e4c
|
163 |
struct shmid_kernel *shp; |
bc56bba8f
|
164 |
shp = shm_lock(sfd->ns, sfd->id); |
023a53557
|
165 |
BUG_ON(IS_ERR(shp)); |
1da177e4c
|
166 |
shp->shm_atim = get_seconds(); |
b488893a3
|
167 |
shp->shm_lprid = task_tgid_vnr(current); |
1da177e4c
|
168 169 170 |
shp->shm_nattch++; shm_unlock(shp); } |
1da177e4c
|
171 172 173 |
/* * shm_destroy - free the struct shmid_kernel * |
f4566f048
|
174 |
* @ns: namespace |
1da177e4c
|
175 176 |
* @shp: struct to free * |
3e148c799
|
177 |
* It has to be called with shp and shm_ids.rw_mutex (writer) locked, |
1da177e4c
|
178 179 |
* but returns with shp unlocked and freed. */ |
4e9823111
|
180 |
static void shm_destroy(struct ipc_namespace *ns, struct shmid_kernel *shp) |
1da177e4c
|
181 |
{ |
4e9823111
|
182 |
ns->shm_tot -= (shp->shm_segsz + PAGE_SIZE - 1) >> PAGE_SHIFT; |
7ca7e564e
|
183 |
shm_rmid(ns, shp); |
1da177e4c
|
184 185 186 187 |
shm_unlock(shp); if (!is_file_hugepages(shp->shm_file)) shmem_lock(shp->shm_file, 0, shp->mlock_user); else |
6d63079ad
|
188 |
user_shm_unlock(shp->shm_file->f_path.dentry->d_inode->i_size, |
1da177e4c
|
189 190 191 192 193 194 195 |
shp->mlock_user); fput (shp->shm_file); security_shm_free(shp); ipc_rcu_putref(shp); } /* |
bc56bba8f
|
196 |
* remove the attach descriptor vma. |
1da177e4c
|
197 198 199 200 |
* free memory for segment if it is marked destroyed. * The descriptor has already been removed from the current->mm->mmap list * and will later be kfree()d. */ |
bc56bba8f
|
201 |
static void shm_close(struct vm_area_struct *vma) |
1da177e4c
|
202 |
{ |
bc56bba8f
|
203 204 |
struct file * file = vma->vm_file; struct shm_file_data *sfd = shm_file_data(file); |
1da177e4c
|
205 |
struct shmid_kernel *shp; |
bc56bba8f
|
206 |
struct ipc_namespace *ns = sfd->ns; |
4e9823111
|
207 |
|
3e148c799
|
208 |
down_write(&shm_ids(ns).rw_mutex); |
1da177e4c
|
209 |
/* remove from the list of attaches of the shm segment */ |
3e148c799
|
210 |
shp = shm_lock_down(ns, sfd->id); |
023a53557
|
211 |
BUG_ON(IS_ERR(shp)); |
b488893a3
|
212 |
shp->shm_lprid = task_tgid_vnr(current); |
1da177e4c
|
213 214 215 |
shp->shm_dtim = get_seconds(); shp->shm_nattch--; if(shp->shm_nattch == 0 && |
b33291c0b
|
216 |
shp->shm_perm.mode & SHM_DEST) |
4e9823111
|
217 |
shm_destroy(ns, shp); |
1da177e4c
|
218 219 |
else shm_unlock(shp); |
3e148c799
|
220 |
up_write(&shm_ids(ns).rw_mutex); |
1da177e4c
|
221 |
} |
d0217ac04
|
222 |
static int shm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) |
bc56bba8f
|
223 224 225 |
{ struct file *file = vma->vm_file; struct shm_file_data *sfd = shm_file_data(file); |
d0217ac04
|
226 |
return sfd->vm_ops->fault(vma, vmf); |
bc56bba8f
|
227 228 229 |
} #ifdef CONFIG_NUMA |
d823e3e75
|
230 |
static int shm_set_policy(struct vm_area_struct *vma, struct mempolicy *new) |
bc56bba8f
|
231 232 233 234 235 236 237 238 |
{ struct file *file = vma->vm_file; struct shm_file_data *sfd = shm_file_data(file); int err = 0; if (sfd->vm_ops->set_policy) err = sfd->vm_ops->set_policy(vma, new); return err; } |
d823e3e75
|
239 240 |
static struct mempolicy *shm_get_policy(struct vm_area_struct *vma, unsigned long addr) |
bc56bba8f
|
241 242 243 244 245 246 247 |
{ struct file *file = vma->vm_file; struct shm_file_data *sfd = shm_file_data(file); struct mempolicy *pol = NULL; if (sfd->vm_ops->get_policy) pol = sfd->vm_ops->get_policy(vma, addr); |
22741925d
|
248 |
else if (vma->vm_policy) |
bc56bba8f
|
249 |
pol = vma->vm_policy; |
22741925d
|
250 251 |
else pol = current->mempolicy; |
bc56bba8f
|
252 253 254 |
return pol; } #endif |
1da177e4c
|
255 256 |
static int shm_mmap(struct file * file, struct vm_area_struct * vma) { |
bc56bba8f
|
257 |
struct shm_file_data *sfd = shm_file_data(file); |
b0e15190e
|
258 |
int ret; |
bc56bba8f
|
259 260 261 262 |
ret = sfd->file->f_op->mmap(sfd->file, vma); if (ret != 0) return ret; sfd->vm_ops = vma->vm_ops; |
2e92a3bae
|
263 |
#ifdef CONFIG_MMU |
54cb8821d
|
264 |
BUG_ON(!sfd->vm_ops->fault); |
2e92a3bae
|
265 |
#endif |
bc56bba8f
|
266 267 |
vma->vm_ops = &shm_vm_ops; shm_open(vma); |
b0e15190e
|
268 269 |
return ret; |
1da177e4c
|
270 |
} |
4e9823111
|
271 272 |
static int shm_release(struct inode *ino, struct file *file) { |
bc56bba8f
|
273 |
struct shm_file_data *sfd = shm_file_data(file); |
4e9823111
|
274 |
|
bc56bba8f
|
275 276 277 |
put_ipc_ns(sfd->ns); shm_file_data(file) = NULL; kfree(sfd); |
4e9823111
|
278 279 |
return 0; } |
516dffdcd
|
280 281 282 283 284 285 286 287 288 289 290 |
static int shm_fsync(struct file *file, struct dentry *dentry, int datasync) { int (*fsync) (struct file *, struct dentry *, int datasync); struct shm_file_data *sfd = shm_file_data(file); int ret = -EINVAL; fsync = sfd->file->f_op->fsync; if (fsync) ret = fsync(sfd->file, sfd->file->f_path.dentry, datasync); return ret; } |
bc56bba8f
|
291 292 293 294 295 |
static unsigned long shm_get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags) { struct shm_file_data *sfd = shm_file_data(file); |
516dffdcd
|
296 297 298 299 300 301 302 303 304 305 306 307 308 |
return get_unmapped_area(sfd->file, addr, len, pgoff, flags); } int is_file_shm_hugepages(struct file *file) { int ret = 0; if (file->f_op == &shm_file_operations) { struct shm_file_data *sfd; sfd = shm_file_data(file); ret = is_file_hugepages(sfd->file); } return ret; |
bc56bba8f
|
309 |
} |
bc56bba8f
|
310 |
|
9a32144e9
|
311 |
static const struct file_operations shm_file_operations = { |
4e9823111
|
312 |
.mmap = shm_mmap, |
516dffdcd
|
313 |
.fsync = shm_fsync, |
4e9823111
|
314 |
.release = shm_release, |
bc56bba8f
|
315 |
.get_unmapped_area = shm_get_unmapped_area, |
1da177e4c
|
316 317 318 319 320 |
}; static struct vm_operations_struct shm_vm_ops = { .open = shm_open, /* callback for a new vm-area open */ .close = shm_close, /* callback for when the vm-area is released */ |
54cb8821d
|
321 |
.fault = shm_fault, |
bc56bba8f
|
322 323 324 |
#if defined(CONFIG_NUMA) .set_policy = shm_set_policy, .get_policy = shm_get_policy, |
1da177e4c
|
325 326 |
#endif }; |
f4566f048
|
327 328 329 330 331 |
/** * newseg - Create a new shared memory segment * @ns: namespace * @params: ptr to the structure that contains key, size and shmflg * |
3e148c799
|
332 |
* Called with shm_ids.rw_mutex held as a writer. |
f4566f048
|
333 |
*/ |
7748dbfaa
|
334 |
static int newseg(struct ipc_namespace *ns, struct ipc_params *params) |
1da177e4c
|
335 |
{ |
7748dbfaa
|
336 337 338 |
key_t key = params->key; int shmflg = params->flg; size_t size = params->u.size; |
1da177e4c
|
339 340 341 342 343 344 |
int error; struct shmid_kernel *shp; int numpages = (size + PAGE_SIZE -1) >> PAGE_SHIFT; struct file * file; char name[13]; int id; |
4e9823111
|
345 |
if (size < SHMMIN || size > ns->shm_ctlmax) |
1da177e4c
|
346 |
return -EINVAL; |
f66d45e99
|
347 |
if (ns->shm_tot + numpages > ns->shm_ctlall) |
1da177e4c
|
348 349 350 351 352 353 354 |
return -ENOSPC; shp = ipc_rcu_alloc(sizeof(*shp)); if (!shp) return -ENOMEM; shp->shm_perm.key = key; |
b33291c0b
|
355 |
shp->shm_perm.mode = (shmflg & S_IRWXUGO); |
1da177e4c
|
356 357 358 359 360 361 362 363 |
shp->mlock_user = NULL; shp->shm_perm.security = NULL; error = security_shm_alloc(shp); if (error) { ipc_rcu_putref(shp); return error; } |
9d66586f7
|
364 |
sprintf (name, "SYSV%08x", key); |
1da177e4c
|
365 |
if (shmflg & SHM_HUGETLB) { |
9d66586f7
|
366 367 |
/* hugetlb_file_setup takes care of mlock user accounting */ file = hugetlb_file_setup(name, size); |
1da177e4c
|
368 369 |
shp->mlock_user = current->user; } else { |
bf8f972d3
|
370 371 372 373 374 375 376 377 |
int acctflag = VM_ACCOUNT; /* * Do not allow no accounting for OVERCOMMIT_NEVER, even * if it's asked for. */ if ((shmflg & SHM_NORESERVE) && sysctl_overcommit_memory != OVERCOMMIT_NEVER) acctflag = 0; |
bf8f972d3
|
378 |
file = shmem_file_setup(name, size, acctflag); |
1da177e4c
|
379 380 381 382 |
} error = PTR_ERR(file); if (IS_ERR(file)) goto no_file; |
4e9823111
|
383 |
id = shm_addid(ns, shp); |
283bb7fad
|
384 385 |
if (id < 0) { error = id; |
1da177e4c
|
386 |
goto no_id; |
283bb7fad
|
387 |
} |
1da177e4c
|
388 |
|
b488893a3
|
389 |
shp->shm_cprid = task_tgid_vnr(current); |
1da177e4c
|
390 391 392 393 394 |
shp->shm_lprid = 0; shp->shm_atim = shp->shm_dtim = 0; shp->shm_ctim = get_seconds(); shp->shm_segsz = size; shp->shm_nattch = 0; |
1b531f213
|
395 |
shp->shm_perm.id = shm_buildid(id, shp->shm_perm.seq); |
1da177e4c
|
396 |
shp->shm_file = file; |
30475cc12
|
397 398 399 400 |
/* * shmid gets reported as "inode#" in /proc/pid/maps. * proc-ps tools use this. Changing this will break them. */ |
7ca7e564e
|
401 |
file->f_dentry->d_inode->i_ino = shp->shm_perm.id; |
551110a94
|
402 |
|
4e9823111
|
403 |
ns->shm_tot += numpages; |
7ca7e564e
|
404 |
error = shp->shm_perm.id; |
1da177e4c
|
405 |
shm_unlock(shp); |
7ca7e564e
|
406 |
return error; |
1da177e4c
|
407 408 409 410 411 412 413 414 |
no_id: fput(file); no_file: security_shm_free(shp); ipc_rcu_putref(shp); return error; } |
f4566f048
|
415 |
/* |
3e148c799
|
416 |
* Called with shm_ids.rw_mutex and ipcp locked. |
f4566f048
|
417 |
*/ |
03f02c765
|
418 |
static inline int shm_security(struct kern_ipc_perm *ipcp, int shmflg) |
7748dbfaa
|
419 |
{ |
03f02c765
|
420 421 422 423 |
struct shmid_kernel *shp; shp = container_of(ipcp, struct shmid_kernel, shm_perm); return security_shm_associate(shp, shmflg); |
7748dbfaa
|
424 |
} |
f4566f048
|
425 |
/* |
3e148c799
|
426 |
* Called with shm_ids.rw_mutex and ipcp locked. |
f4566f048
|
427 |
*/ |
03f02c765
|
428 429 |
static inline int shm_more_checks(struct kern_ipc_perm *ipcp, struct ipc_params *params) |
7748dbfaa
|
430 |
{ |
03f02c765
|
431 432 433 434 |
struct shmid_kernel *shp; shp = container_of(ipcp, struct shmid_kernel, shm_perm); if (shp->shm_segsz < params->u.size) |
7748dbfaa
|
435 436 437 438 |
return -EINVAL; return 0; } |
1da177e4c
|
439 440 |
asmlinkage long sys_shmget (key_t key, size_t size, int shmflg) { |
4e9823111
|
441 |
struct ipc_namespace *ns; |
7748dbfaa
|
442 443 |
struct ipc_ops shm_ops; struct ipc_params shm_params; |
4e9823111
|
444 445 |
ns = current->nsproxy->ipc_ns; |
1da177e4c
|
446 |
|
7748dbfaa
|
447 448 449 |
shm_ops.getnew = newseg; shm_ops.associate = shm_security; shm_ops.more_checks = shm_more_checks; |
7ca7e564e
|
450 |
|
7748dbfaa
|
451 452 453 |
shm_params.key = key; shm_params.flg = shmflg; shm_params.u.size = size; |
1da177e4c
|
454 |
|
7748dbfaa
|
455 |
return ipcget(ns, &shm_ids(ns), &shm_ops, &shm_params); |
1da177e4c
|
456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 |
} static inline unsigned long copy_shmid_to_user(void __user *buf, struct shmid64_ds *in, int version) { switch(version) { case IPC_64: return copy_to_user(buf, in, sizeof(*in)); case IPC_OLD: { struct shmid_ds out; ipc64_perm_to_ipc_perm(&in->shm_perm, &out.shm_perm); out.shm_segsz = in->shm_segsz; out.shm_atime = in->shm_atime; out.shm_dtime = in->shm_dtime; out.shm_ctime = in->shm_ctime; out.shm_cpid = in->shm_cpid; out.shm_lpid = in->shm_lpid; out.shm_nattch = in->shm_nattch; return copy_to_user(buf, &out, sizeof(out)); } default: return -EINVAL; } } struct shm_setbuf { uid_t uid; gid_t gid; mode_t mode; }; static inline unsigned long copy_shmid_from_user(struct shm_setbuf *out, void __user *buf, int version) { switch(version) { case IPC_64: { struct shmid64_ds tbuf; if (copy_from_user(&tbuf, buf, sizeof(tbuf))) return -EFAULT; out->uid = tbuf.shm_perm.uid; out->gid = tbuf.shm_perm.gid; |
b33291c0b
|
501 |
out->mode = tbuf.shm_perm.mode; |
1da177e4c
|
502 503 504 505 506 507 508 509 510 511 512 513 |
return 0; } case IPC_OLD: { struct shmid_ds tbuf_old; if (copy_from_user(&tbuf_old, buf, sizeof(tbuf_old))) return -EFAULT; out->uid = tbuf_old.shm_perm.uid; out->gid = tbuf_old.shm_perm.gid; |
b33291c0b
|
514 |
out->mode = tbuf_old.shm_perm.mode; |
1da177e4c
|
515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 |
return 0; } default: return -EINVAL; } } static inline unsigned long copy_shminfo_to_user(void __user *buf, struct shminfo64 *in, int version) { switch(version) { case IPC_64: return copy_to_user(buf, in, sizeof(*in)); case IPC_OLD: { struct shminfo out; if(in->shmmax > INT_MAX) out.shmmax = INT_MAX; else out.shmmax = (int)in->shmmax; out.shmmin = in->shmmin; out.shmmni = in->shmmni; out.shmseg = in->shmseg; out.shmall = in->shmall; return copy_to_user(buf, &out, sizeof(out)); } default: return -EINVAL; } } |
f4566f048
|
548 |
/* |
3e148c799
|
549 |
* Called with shm_ids.rw_mutex held as a reader |
f4566f048
|
550 |
*/ |
4e9823111
|
551 552 |
static void shm_get_stat(struct ipc_namespace *ns, unsigned long *rss, unsigned long *swp) |
1da177e4c
|
553 |
{ |
7ca7e564e
|
554 555 |
int next_id; int total, in_use; |
1da177e4c
|
556 557 558 |
*rss = 0; *swp = 0; |
7ca7e564e
|
559 560 561 |
in_use = shm_ids(ns).in_use; for (total = 0, next_id = 0; total < in_use; next_id++) { |
1da177e4c
|
562 563 |
struct shmid_kernel *shp; struct inode *inode; |
637c36634
|
564 |
shp = idr_find(&shm_ids(ns).ipcs_idr, next_id); |
7ca7e564e
|
565 |
if (shp == NULL) |
1da177e4c
|
566 |
continue; |
6d63079ad
|
567 |
inode = shp->shm_file->f_path.dentry->d_inode; |
1da177e4c
|
568 569 570 571 572 573 574 575 576 577 578 |
if (is_file_hugepages(shp->shm_file)) { struct address_space *mapping = inode->i_mapping; *rss += (HPAGE_SIZE/PAGE_SIZE)*mapping->nrpages; } else { struct shmem_inode_info *info = SHMEM_I(inode); spin_lock(&info->lock); *rss += inode->i_mapping->nrpages; *swp += info->swapped; spin_unlock(&info->lock); } |
7ca7e564e
|
579 580 |
total++; |
1da177e4c
|
581 582 583 584 585 586 587 588 |
} } asmlinkage long sys_shmctl (int shmid, int cmd, struct shmid_ds __user *buf) { struct shm_setbuf setbuf; struct shmid_kernel *shp; int err, version; |
4e9823111
|
589 |
struct ipc_namespace *ns; |
1da177e4c
|
590 591 592 593 594 595 596 |
if (cmd < 0 || shmid < 0) { err = -EINVAL; goto out; } version = ipc_parse_version(&cmd); |
4e9823111
|
597 |
ns = current->nsproxy->ipc_ns; |
1da177e4c
|
598 599 600 601 602 603 604 605 606 607 608 |
switch (cmd) { /* replace with proc interface ? */ case IPC_INFO: { struct shminfo64 shminfo; err = security_shm_shmctl(NULL, cmd); if (err) return err; memset(&shminfo,0,sizeof(shminfo)); |
4e9823111
|
609 610 611 |
shminfo.shmmni = shminfo.shmseg = ns->shm_ctlmni; shminfo.shmmax = ns->shm_ctlmax; shminfo.shmall = ns->shm_ctlall; |
1da177e4c
|
612 613 614 615 |
shminfo.shmmin = SHMMIN; if(copy_shminfo_to_user (buf, &shminfo, version)) return -EFAULT; |
f4566f048
|
616 |
|
3e148c799
|
617 |
down_read(&shm_ids(ns).rw_mutex); |
7ca7e564e
|
618 |
err = ipc_get_maxid(&shm_ids(ns)); |
3e148c799
|
619 |
up_read(&shm_ids(ns).rw_mutex); |
f4566f048
|
620 |
|
1da177e4c
|
621 622 623 624 625 626 627 628 629 630 631 632 633 |
if(err<0) err = 0; goto out; } case SHM_INFO: { struct shm_info shm_info; err = security_shm_shmctl(NULL, cmd); if (err) return err; memset(&shm_info,0,sizeof(shm_info)); |
3e148c799
|
634 |
down_read(&shm_ids(ns).rw_mutex); |
4e9823111
|
635 636 637 |
shm_info.used_ids = shm_ids(ns).in_use; shm_get_stat (ns, &shm_info.shm_rss, &shm_info.shm_swp); shm_info.shm_tot = ns->shm_tot; |
1da177e4c
|
638 639 |
shm_info.swap_attempts = 0; shm_info.swap_successes = 0; |
7ca7e564e
|
640 |
err = ipc_get_maxid(&shm_ids(ns)); |
3e148c799
|
641 |
up_read(&shm_ids(ns).rw_mutex); |
1da177e4c
|
642 643 644 645 646 647 648 649 650 651 652 653 654 |
if(copy_to_user (buf, &shm_info, sizeof(shm_info))) { err = -EFAULT; goto out; } err = err < 0 ? 0 : err; goto out; } case SHM_STAT: case IPC_STAT: { struct shmid64_ds tbuf; int result; |
023a53557
|
655 656 657 |
if (!buf) { err = -EFAULT; |
1da177e4c
|
658 |
goto out; |
023a53557
|
659 660 661 662 663 664 665 666 |
} if (cmd == SHM_STAT) { shp = shm_lock(ns, shmid); if (IS_ERR(shp)) { err = PTR_ERR(shp); goto out; } |
7ca7e564e
|
667 |
result = shp->shm_perm.id; |
1da177e4c
|
668 |
} else { |
023a53557
|
669 670 671 672 673 |
shp = shm_lock_check(ns, shmid); if (IS_ERR(shp)) { err = PTR_ERR(shp); goto out; } |
1da177e4c
|
674 675 676 677 678 679 680 681 |
result = 0; } err=-EACCES; if (ipcperms (&shp->shm_perm, S_IRUGO)) goto out_unlock; err = security_shm_shmctl(shp, cmd); if (err) goto out_unlock; |
023a53557
|
682 |
memset(&tbuf, 0, sizeof(tbuf)); |
1da177e4c
|
683 684 685 686 687 688 689 |
kernel_to_ipc64_perm(&shp->shm_perm, &tbuf.shm_perm); tbuf.shm_segsz = shp->shm_segsz; tbuf.shm_atime = shp->shm_atim; tbuf.shm_dtime = shp->shm_dtim; tbuf.shm_ctime = shp->shm_ctim; tbuf.shm_cpid = shp->shm_cprid; tbuf.shm_lpid = shp->shm_lprid; |
bc56bba8f
|
690 |
tbuf.shm_nattch = shp->shm_nattch; |
1da177e4c
|
691 692 693 694 695 696 697 698 699 700 |
shm_unlock(shp); if(copy_shmid_to_user (buf, &tbuf, version)) err = -EFAULT; else err = result; goto out; } case SHM_LOCK: case SHM_UNLOCK: { |
023a53557
|
701 702 703 |
shp = shm_lock_check(ns, shmid); if (IS_ERR(shp)) { err = PTR_ERR(shp); |
1da177e4c
|
704 705 |
goto out; } |
1da177e4c
|
706 |
|
073115d6b
|
707 708 709 |
err = audit_ipc_obj(&(shp->shm_perm)); if (err) goto out_unlock; |
1da177e4c
|
710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 |
if (!capable(CAP_IPC_LOCK)) { err = -EPERM; if (current->euid != shp->shm_perm.uid && current->euid != shp->shm_perm.cuid) goto out_unlock; if (cmd == SHM_LOCK && !current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur) goto out_unlock; } err = security_shm_shmctl(shp, cmd); if (err) goto out_unlock; if(cmd==SHM_LOCK) { struct user_struct * user = current->user; if (!is_file_hugepages(shp->shm_file)) { err = shmem_lock(shp->shm_file, 1, user); |
7be77e20d
|
728 |
if (!err && !(shp->shm_perm.mode & SHM_LOCKED)){ |
b33291c0b
|
729 |
shp->shm_perm.mode |= SHM_LOCKED; |
1da177e4c
|
730 731 732 733 734 |
shp->mlock_user = user; } } } else if (!is_file_hugepages(shp->shm_file)) { shmem_lock(shp->shm_file, 0, shp->mlock_user); |
b33291c0b
|
735 |
shp->shm_perm.mode &= ~SHM_LOCKED; |
1da177e4c
|
736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 |
shp->mlock_user = NULL; } shm_unlock(shp); goto out; } case IPC_RMID: { /* * We cannot simply remove the file. The SVID states * that the block remains until the last person * detaches from it, then is deleted. A shmat() on * an RMID segment is legal in older Linux and if * we change it apps break... * * Instead we set a destroyed flag, and then blow * the name away when the usage hits zero. */ |
3e148c799
|
753 754 |
down_write(&shm_ids(ns).rw_mutex); shp = shm_lock_check_down(ns, shmid); |
023a53557
|
755 756 |
if (IS_ERR(shp)) { err = PTR_ERR(shp); |
1da177e4c
|
757 |
goto out_up; |
023a53557
|
758 |
} |
1da177e4c
|
759 |
|
073115d6b
|
760 761 762 |
err = audit_ipc_obj(&(shp->shm_perm)); if (err) goto out_unlock_up; |
1da177e4c
|
763 764 765 766 767 768 769 770 771 772 |
if (current->euid != shp->shm_perm.uid && current->euid != shp->shm_perm.cuid && !capable(CAP_SYS_ADMIN)) { err=-EPERM; goto out_unlock_up; } err = security_shm_shmctl(shp, cmd); if (err) goto out_unlock_up; |
01b8b07a5
|
773 |
do_shm_rmid(ns, &shp->shm_perm); |
3e148c799
|
774 |
up_write(&shm_ids(ns).rw_mutex); |
1da177e4c
|
775 776 777 778 779 |
goto out; } case IPC_SET: { |
023a53557
|
780 781 782 783 |
if (!buf) { err = -EFAULT; goto out; } |
1da177e4c
|
784 785 786 787 |
if (copy_shmid_from_user (&setbuf, buf, version)) { err = -EFAULT; goto out; } |
3e148c799
|
788 789 |
down_write(&shm_ids(ns).rw_mutex); shp = shm_lock_check_down(ns, shmid); |
023a53557
|
790 791 |
if (IS_ERR(shp)) { err = PTR_ERR(shp); |
1da177e4c
|
792 |
goto out_up; |
023a53557
|
793 |
} |
073115d6b
|
794 795 796 |
err = audit_ipc_obj(&(shp->shm_perm)); if (err) goto out_unlock_up; |
ac03221a4
|
797 |
err = audit_ipc_set_perm(0, setbuf.uid, setbuf.gid, setbuf.mode); |
073115d6b
|
798 799 |
if (err) goto out_unlock_up; |
1da177e4c
|
800 801 802 803 804 805 806 807 808 809 810 811 812 |
err=-EPERM; if (current->euid != shp->shm_perm.uid && current->euid != shp->shm_perm.cuid && !capable(CAP_SYS_ADMIN)) { goto out_unlock_up; } err = security_shm_shmctl(shp, cmd); if (err) goto out_unlock_up; shp->shm_perm.uid = setbuf.uid; shp->shm_perm.gid = setbuf.gid; |
b33291c0b
|
813 |
shp->shm_perm.mode = (shp->shm_perm.mode & ~S_IRWXUGO) |
1da177e4c
|
814 815 816 817 818 819 820 821 822 823 824 825 826 827 |
| (setbuf.mode & S_IRWXUGO); shp->shm_ctim = get_seconds(); break; } default: err = -EINVAL; goto out; } err = 0; out_unlock_up: shm_unlock(shp); out_up: |
3e148c799
|
828 |
up_write(&shm_ids(ns).rw_mutex); |
1da177e4c
|
829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 |
goto out; out_unlock: shm_unlock(shp); out: return err; } /* * Fix shmaddr, allocate descriptor, map shm, add attach descriptor to lists. * * NOTE! Despite the name, this is NOT a direct system call entrypoint. The * "raddr" thing points to kernel space, and there has to be a wrapper around * this. */ long do_shmat(int shmid, char __user *shmaddr, int shmflg, ulong *raddr) { struct shmid_kernel *shp; unsigned long addr; unsigned long size; struct file * file; int err; unsigned long flags; unsigned long prot; |
1da177e4c
|
852 |
int acc_mode; |
bc56bba8f
|
853 |
unsigned long user_addr; |
4e9823111
|
854 |
struct ipc_namespace *ns; |
bc56bba8f
|
855 856 857 |
struct shm_file_data *sfd; struct path path; mode_t f_mode; |
1da177e4c
|
858 |
|
bc56bba8f
|
859 860 |
err = -EINVAL; if (shmid < 0) |
1da177e4c
|
861 |
goto out; |
bc56bba8f
|
862 |
else if ((addr = (ulong)shmaddr)) { |
1da177e4c
|
863 864 865 866 867 868 869 |
if (addr & (SHMLBA-1)) { if (shmflg & SHM_RND) addr &= ~(SHMLBA-1); /* round down */ else #ifndef __ARCH_FORCE_SHMLBA if (addr & ~PAGE_MASK) #endif |
bc56bba8f
|
870 |
goto out; |
1da177e4c
|
871 872 873 874 |
} flags = MAP_SHARED | MAP_FIXED; } else { if ((shmflg & SHM_REMAP)) |
bc56bba8f
|
875 |
goto out; |
1da177e4c
|
876 877 878 879 880 881 |
flags = MAP_SHARED; } if (shmflg & SHM_RDONLY) { prot = PROT_READ; |
1da177e4c
|
882 |
acc_mode = S_IRUGO; |
bc56bba8f
|
883 |
f_mode = FMODE_READ; |
1da177e4c
|
884 885 |
} else { prot = PROT_READ | PROT_WRITE; |
1da177e4c
|
886 |
acc_mode = S_IRUGO | S_IWUGO; |
bc56bba8f
|
887 |
f_mode = FMODE_READ | FMODE_WRITE; |
1da177e4c
|
888 889 890 891 892 893 894 895 896 897 |
} if (shmflg & SHM_EXEC) { prot |= PROT_EXEC; acc_mode |= S_IXUGO; } /* * We cannot rely on the fs check since SYSV IPC does have an * additional creator id... */ |
4e9823111
|
898 |
ns = current->nsproxy->ipc_ns; |
023a53557
|
899 900 901 |
shp = shm_lock_check(ns, shmid); if (IS_ERR(shp)) { err = PTR_ERR(shp); |
1da177e4c
|
902 |
goto out; |
023a53557
|
903 |
} |
bc56bba8f
|
904 905 906 907 |
err = -EACCES; if (ipcperms(&shp->shm_perm, acc_mode)) goto out_unlock; |
1da177e4c
|
908 909 |
err = security_shm_shmat(shp, shmaddr, shmflg); |
bc56bba8f
|
910 911 912 913 |
if (err) goto out_unlock; path.dentry = dget(shp->shm_file->f_path.dentry); |
ce8d2cdf3
|
914 |
path.mnt = shp->shm_file->f_path.mnt; |
1da177e4c
|
915 |
shp->shm_nattch++; |
bc56bba8f
|
916 |
size = i_size_read(path.dentry->d_inode); |
1da177e4c
|
917 |
shm_unlock(shp); |
bc56bba8f
|
918 919 920 |
err = -ENOMEM; sfd = kzalloc(sizeof(*sfd), GFP_KERNEL); if (!sfd) |
ce8d2cdf3
|
921 |
goto out_put_dentry; |
bc56bba8f
|
922 923 |
err = -ENOMEM; |
ce8d2cdf3
|
924 925 |
file = alloc_file(path.mnt, path.dentry, f_mode, &shm_file_operations); |
bc56bba8f
|
926 927 |
if (!file) goto out_free; |
bc56bba8f
|
928 |
file->private_data = sfd; |
bc56bba8f
|
929 |
file->f_mapping = shp->shm_file->f_mapping; |
7ca7e564e
|
930 |
sfd->id = shp->shm_perm.id; |
bc56bba8f
|
931 932 933 |
sfd->ns = get_ipc_ns(ns); sfd->file = shp->shm_file; sfd->vm_ops = NULL; |
1da177e4c
|
934 935 |
down_write(¤t->mm->mmap_sem); if (addr && !(shmflg & SHM_REMAP)) { |
bc56bba8f
|
936 |
err = -EINVAL; |
1da177e4c
|
937 938 939 940 941 942 943 944 945 946 947 |
if (find_vma_intersection(current->mm, addr, addr + size)) goto invalid; /* * If shm segment goes below stack, make sure there is some * space left for the stack to grow (at least 4 pages). */ if (addr < current->mm->start_stack && addr > current->mm->start_stack - size - PAGE_SIZE * 5) goto invalid; } |
bc56bba8f
|
948 949 950 951 952 |
user_addr = do_mmap (file, addr, size, prot, flags, 0); *raddr = user_addr; err = 0; if (IS_ERR_VALUE(user_addr)) err = (long)user_addr; |
1da177e4c
|
953 954 |
invalid: up_write(¤t->mm->mmap_sem); |
bc56bba8f
|
955 956 957 |
fput(file); out_nattch: |
3e148c799
|
958 959 |
down_write(&shm_ids(ns).rw_mutex); shp = shm_lock_down(ns, shmid); |
023a53557
|
960 |
BUG_ON(IS_ERR(shp)); |
1da177e4c
|
961 962 |
shp->shm_nattch--; if(shp->shm_nattch == 0 && |
b33291c0b
|
963 |
shp->shm_perm.mode & SHM_DEST) |
4e9823111
|
964 |
shm_destroy(ns, shp); |
1da177e4c
|
965 966 |
else shm_unlock(shp); |
3e148c799
|
967 |
up_write(&shm_ids(ns).rw_mutex); |
1da177e4c
|
968 |
|
1da177e4c
|
969 970 |
out: return err; |
bc56bba8f
|
971 972 973 974 975 976 977 |
out_unlock: shm_unlock(shp); goto out; out_free: kfree(sfd); |
ce8d2cdf3
|
978 |
out_put_dentry: |
bc56bba8f
|
979 |
dput(path.dentry); |
bc56bba8f
|
980 |
goto out_nattch; |
1da177e4c
|
981 |
} |
7d87e14c2
|
982 983 984 985 986 987 988 989 990 991 992 |
asmlinkage long sys_shmat(int shmid, char __user *shmaddr, int shmflg) { unsigned long ret; long err; err = do_shmat(shmid, shmaddr, shmflg, &ret); if (err) return err; force_successful_syscall_return(); return (long)ret; } |
1da177e4c
|
993 994 995 996 997 998 999 1000 1001 1002 1003 |
/* * detach and kill segment if marked destroyed. * The work is done in shm_close. */ asmlinkage long sys_shmdt(char __user *shmaddr) { struct mm_struct *mm = current->mm; struct vm_area_struct *vma, *next; unsigned long addr = (unsigned long)shmaddr; loff_t size = 0; int retval = -EINVAL; |
df1e2fb54
|
1004 1005 |
if (addr & ~PAGE_MASK) return retval; |
1da177e4c
|
1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 |
down_write(&mm->mmap_sem); /* * This function tries to be smart and unmap shm segments that * were modified by partial mlock or munmap calls: * - It first determines the size of the shm segment that should be * unmapped: It searches for a vma that is backed by shm and that * started at address shmaddr. It records it's size and then unmaps * it. * - Then it unmaps all shm vmas that started at shmaddr and that * are within the initially determined size. * Errors from do_munmap are ignored: the function only fails if * it's called with invalid parameters or if it's called to unmap * a part of a vma. Both calls in this function are for full vmas, * the parameters are directly copied from the vma itself and always * valid - therefore do_munmap cannot fail. (famous last words?) */ /* * If it had been mremap()'d, the starting address would not * match the usual checks anyway. So assume all vma's are * above the starting address given. */ vma = find_vma(mm, addr); while (vma) { next = vma->vm_next; /* * Check if the starting address would match, i.e. it's * a fragment created by mprotect() and/or munmap(), or it * otherwise it starts at this address with no hassles. */ |
bc56bba8f
|
1038 |
if ((vma->vm_ops == &shm_vm_ops) && |
1da177e4c
|
1039 |
(vma->vm_start - addr)/PAGE_SIZE == vma->vm_pgoff) { |
6d63079ad
|
1040 |
size = vma->vm_file->f_path.dentry->d_inode->i_size; |
1da177e4c
|
1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 |
do_munmap(mm, vma->vm_start, vma->vm_end - vma->vm_start); /* * We discovered the size of the shm segment, so * break out of here and fall through to the next * loop that uses the size information to stop * searching for matching vma's. */ retval = 0; vma = next; break; } vma = next; } /* * We need look no further than the maximum address a fragment * could possibly have landed at. Also cast things to loff_t to * prevent overflows and make comparisions vs. equal-width types. */ |
8e36709d8
|
1060 |
size = PAGE_ALIGN(size); |
1da177e4c
|
1061 1062 1063 1064 |
while (vma && (loff_t)(vma->vm_end - addr) <= size) { next = vma->vm_next; /* finding a matching vma now does not alter retval */ |
bc56bba8f
|
1065 |
if ((vma->vm_ops == &shm_vm_ops) && |
1da177e4c
|
1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 |
(vma->vm_start - addr)/PAGE_SIZE == vma->vm_pgoff) do_munmap(mm, vma->vm_start, vma->vm_end - vma->vm_start); vma = next; } up_write(&mm->mmap_sem); return retval; } #ifdef CONFIG_PROC_FS |
19b4946ca
|
1077 |
static int sysvipc_shm_proc_show(struct seq_file *s, void *it) |
1da177e4c
|
1078 |
{ |
19b4946ca
|
1079 1080 |
struct shmid_kernel *shp = it; char *format; |
1da177e4c
|
1081 |
|
1da177e4c
|
1082 1083 1084 1085 |
#define SMALL_STRING "%10d %10d %4o %10u %5u %5u %5d %5u %5u %5u %5u %10lu %10lu %10lu " #define BIG_STRING "%10d %10d %4o %21u %5u %5u %5d %5u %5u %5u %5u %10lu %10lu %10lu " |
1da177e4c
|
1086 |
|
19b4946ca
|
1087 1088 1089 1090 1091 1092 |
if (sizeof(size_t) <= sizeof(int)) format = SMALL_STRING; else format = BIG_STRING; return seq_printf(s, format, shp->shm_perm.key, |
7ca7e564e
|
1093 |
shp->shm_perm.id, |
b33291c0b
|
1094 |
shp->shm_perm.mode, |
19b4946ca
|
1095 1096 1097 |
shp->shm_segsz, shp->shm_cprid, shp->shm_lprid, |
bc56bba8f
|
1098 |
shp->shm_nattch, |
19b4946ca
|
1099 1100 1101 1102 1103 1104 1105 |
shp->shm_perm.uid, shp->shm_perm.gid, shp->shm_perm.cuid, shp->shm_perm.cgid, shp->shm_atim, shp->shm_dtim, shp->shm_ctim); |
1da177e4c
|
1106 1107 |
} #endif |