Blame view
ipc/shm.c
25.5 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) |
1da177e4c
|
59 |
|
7748dbfaa
|
60 |
static int newseg(struct ipc_namespace *, struct ipc_params *); |
bc56bba8f
|
61 62 |
static void shm_open(struct vm_area_struct *vma); static void shm_close(struct vm_area_struct *vma); |
4e9823111
|
63 |
static void shm_destroy (struct ipc_namespace *ns, struct shmid_kernel *shp); |
1da177e4c
|
64 |
#ifdef CONFIG_PROC_FS |
19b4946ca
|
65 |
static int sysvipc_shm_proc_show(struct seq_file *s, void *it); |
1da177e4c
|
66 |
#endif |
ed2ddbf88
|
67 |
void shm_init_ns(struct ipc_namespace *ns) |
4e9823111
|
68 |
{ |
4e9823111
|
69 70 71 72 |
ns->shm_ctlmax = SHMMAX; ns->shm_ctlall = SHMALL; ns->shm_ctlmni = SHMMNI; ns->shm_tot = 0; |
ed2ddbf88
|
73 |
ipc_init_ids(&ns->ids[IPC_SHM_IDS]); |
4e9823111
|
74 |
} |
f4566f048
|
75 |
/* |
3e148c799
|
76 77 |
* Called with shm_ids.rw_mutex (writer) and the shp structure locked. * Only shm_ids.rw_mutex remains locked on exit. |
f4566f048
|
78 |
*/ |
01b8b07a5
|
79 |
static void do_shm_rmid(struct ipc_namespace *ns, struct kern_ipc_perm *ipcp) |
4e9823111
|
80 |
{ |
01b8b07a5
|
81 82 |
struct shmid_kernel *shp; shp = container_of(ipcp, struct shmid_kernel, shm_perm); |
4e9823111
|
83 84 85 86 87 88 89 90 |
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
|
91 |
#ifdef CONFIG_IPC_NS |
4e9823111
|
92 93 |
void shm_exit_ns(struct ipc_namespace *ns) { |
01b8b07a5
|
94 |
free_ipcs(ns, &shm_ids(ns), do_shm_rmid); |
4e9823111
|
95 |
} |
ae5e1b22f
|
96 |
#endif |
1da177e4c
|
97 98 99 |
void __init shm_init (void) { |
ed2ddbf88
|
100 |
shm_init_ns(&init_ipc_ns); |
19b4946ca
|
101 102 103 |
ipc_init_proc_interface("sysvipc/shm", " key shmid perms size cpid lpid nattch uid gid cuid cgid atime dtime ctime ", |
4e9823111
|
104 |
IPC_SHM_IDS, sysvipc_shm_proc_show); |
1da177e4c
|
105 |
} |
3e148c799
|
106 107 108 109 110 111 112 113 |
/* * 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
|
114 115 |
if (IS_ERR(ipcp)) return (struct shmid_kernel *)ipcp; |
3e148c799
|
116 117 |
return container_of(ipcp, struct shmid_kernel, shm_perm); } |
3e148c799
|
118 119 120 121 |
/* * shm_lock_(check_) routines are called in the paths where the rw_mutex * is not held. */ |
023a53557
|
122 |
static inline struct shmid_kernel *shm_lock(struct ipc_namespace *ns, int id) |
1da177e4c
|
123 |
{ |
03f02c765
|
124 |
struct kern_ipc_perm *ipcp = ipc_lock(&shm_ids(ns), id); |
b1ed88b47
|
125 126 |
if (IS_ERR(ipcp)) return (struct shmid_kernel *)ipcp; |
03f02c765
|
127 |
return container_of(ipcp, struct shmid_kernel, shm_perm); |
023a53557
|
128 129 130 131 132 |
} static inline struct shmid_kernel *shm_lock_check(struct ipc_namespace *ns, int id) { |
03f02c765
|
133 |
struct kern_ipc_perm *ipcp = ipc_lock_check(&shm_ids(ns), id); |
b1ed88b47
|
134 135 |
if (IS_ERR(ipcp)) return (struct shmid_kernel *)ipcp; |
03f02c765
|
136 |
return container_of(ipcp, struct shmid_kernel, shm_perm); |
1da177e4c
|
137 |
} |
7ca7e564e
|
138 |
static inline void shm_rmid(struct ipc_namespace *ns, struct shmid_kernel *s) |
1da177e4c
|
139 |
{ |
7ca7e564e
|
140 |
ipc_rmid(&shm_ids(ns), &s->shm_perm); |
1da177e4c
|
141 |
} |
1da177e4c
|
142 |
|
bc56bba8f
|
143 144 |
/* This is called by fork, once for every shm attach. */ static void shm_open(struct vm_area_struct *vma) |
4e9823111
|
145 |
{ |
bc56bba8f
|
146 147 |
struct file *file = vma->vm_file; struct shm_file_data *sfd = shm_file_data(file); |
1da177e4c
|
148 |
struct shmid_kernel *shp; |
bc56bba8f
|
149 |
shp = shm_lock(sfd->ns, sfd->id); |
023a53557
|
150 |
BUG_ON(IS_ERR(shp)); |
1da177e4c
|
151 |
shp->shm_atim = get_seconds(); |
b488893a3
|
152 |
shp->shm_lprid = task_tgid_vnr(current); |
1da177e4c
|
153 154 155 |
shp->shm_nattch++; shm_unlock(shp); } |
1da177e4c
|
156 157 158 |
/* * shm_destroy - free the struct shmid_kernel * |
f4566f048
|
159 |
* @ns: namespace |
1da177e4c
|
160 161 |
* @shp: struct to free * |
3e148c799
|
162 |
* It has to be called with shp and shm_ids.rw_mutex (writer) locked, |
1da177e4c
|
163 164 |
* but returns with shp unlocked and freed. */ |
4e9823111
|
165 |
static void shm_destroy(struct ipc_namespace *ns, struct shmid_kernel *shp) |
1da177e4c
|
166 |
{ |
4e9823111
|
167 |
ns->shm_tot -= (shp->shm_segsz + PAGE_SIZE - 1) >> PAGE_SHIFT; |
7ca7e564e
|
168 |
shm_rmid(ns, shp); |
1da177e4c
|
169 170 171 172 |
shm_unlock(shp); if (!is_file_hugepages(shp->shm_file)) shmem_lock(shp->shm_file, 0, shp->mlock_user); else |
6d63079ad
|
173 |
user_shm_unlock(shp->shm_file->f_path.dentry->d_inode->i_size, |
1da177e4c
|
174 175 176 177 178 179 180 |
shp->mlock_user); fput (shp->shm_file); security_shm_free(shp); ipc_rcu_putref(shp); } /* |
bc56bba8f
|
181 |
* remove the attach descriptor vma. |
1da177e4c
|
182 183 184 185 |
* 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
|
186 |
static void shm_close(struct vm_area_struct *vma) |
1da177e4c
|
187 |
{ |
bc56bba8f
|
188 189 |
struct file * file = vma->vm_file; struct shm_file_data *sfd = shm_file_data(file); |
1da177e4c
|
190 |
struct shmid_kernel *shp; |
bc56bba8f
|
191 |
struct ipc_namespace *ns = sfd->ns; |
4e9823111
|
192 |
|
3e148c799
|
193 |
down_write(&shm_ids(ns).rw_mutex); |
1da177e4c
|
194 |
/* remove from the list of attaches of the shm segment */ |
3e148c799
|
195 |
shp = shm_lock_down(ns, sfd->id); |
023a53557
|
196 |
BUG_ON(IS_ERR(shp)); |
b488893a3
|
197 |
shp->shm_lprid = task_tgid_vnr(current); |
1da177e4c
|
198 199 200 |
shp->shm_dtim = get_seconds(); shp->shm_nattch--; if(shp->shm_nattch == 0 && |
b33291c0b
|
201 |
shp->shm_perm.mode & SHM_DEST) |
4e9823111
|
202 |
shm_destroy(ns, shp); |
1da177e4c
|
203 204 |
else shm_unlock(shp); |
3e148c799
|
205 |
up_write(&shm_ids(ns).rw_mutex); |
1da177e4c
|
206 |
} |
d0217ac04
|
207 |
static int shm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) |
bc56bba8f
|
208 209 210 |
{ struct file *file = vma->vm_file; struct shm_file_data *sfd = shm_file_data(file); |
d0217ac04
|
211 |
return sfd->vm_ops->fault(vma, vmf); |
bc56bba8f
|
212 213 214 |
} #ifdef CONFIG_NUMA |
d823e3e75
|
215 |
static int shm_set_policy(struct vm_area_struct *vma, struct mempolicy *new) |
bc56bba8f
|
216 217 218 219 220 221 222 223 |
{ 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
|
224 225 |
static struct mempolicy *shm_get_policy(struct vm_area_struct *vma, unsigned long addr) |
bc56bba8f
|
226 227 228 229 230 231 232 |
{ 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); |
52cd3b074
|
233 |
else if (vma->vm_policy) |
bc56bba8f
|
234 |
pol = vma->vm_policy; |
52cd3b074
|
235 |
|
bc56bba8f
|
236 237 238 |
return pol; } #endif |
1da177e4c
|
239 240 |
static int shm_mmap(struct file * file, struct vm_area_struct * vma) { |
bc56bba8f
|
241 |
struct shm_file_data *sfd = shm_file_data(file); |
b0e15190e
|
242 |
int ret; |
bc56bba8f
|
243 244 245 246 |
ret = sfd->file->f_op->mmap(sfd->file, vma); if (ret != 0) return ret; sfd->vm_ops = vma->vm_ops; |
2e92a3bae
|
247 |
#ifdef CONFIG_MMU |
54cb8821d
|
248 |
BUG_ON(!sfd->vm_ops->fault); |
2e92a3bae
|
249 |
#endif |
bc56bba8f
|
250 251 |
vma->vm_ops = &shm_vm_ops; shm_open(vma); |
b0e15190e
|
252 253 |
return ret; |
1da177e4c
|
254 |
} |
4e9823111
|
255 256 |
static int shm_release(struct inode *ino, struct file *file) { |
bc56bba8f
|
257 |
struct shm_file_data *sfd = shm_file_data(file); |
4e9823111
|
258 |
|
bc56bba8f
|
259 260 261 |
put_ipc_ns(sfd->ns); shm_file_data(file) = NULL; kfree(sfd); |
4e9823111
|
262 263 |
return 0; } |
516dffdcd
|
264 265 266 267 268 269 270 271 272 273 274 |
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
|
275 276 277 278 279 |
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
|
280 281 282 283 284 285 286 287 288 289 290 291 292 |
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
|
293 |
} |
bc56bba8f
|
294 |
|
9a32144e9
|
295 |
static const struct file_operations shm_file_operations = { |
4e9823111
|
296 |
.mmap = shm_mmap, |
516dffdcd
|
297 |
.fsync = shm_fsync, |
4e9823111
|
298 |
.release = shm_release, |
bc56bba8f
|
299 |
.get_unmapped_area = shm_get_unmapped_area, |
1da177e4c
|
300 301 302 303 304 |
}; 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
|
305 |
.fault = shm_fault, |
bc56bba8f
|
306 307 308 |
#if defined(CONFIG_NUMA) .set_policy = shm_set_policy, .get_policy = shm_get_policy, |
1da177e4c
|
309 310 |
#endif }; |
f4566f048
|
311 312 313 314 315 |
/** * newseg - Create a new shared memory segment * @ns: namespace * @params: ptr to the structure that contains key, size and shmflg * |
3e148c799
|
316 |
* Called with shm_ids.rw_mutex held as a writer. |
f4566f048
|
317 |
*/ |
7748dbfaa
|
318 |
static int newseg(struct ipc_namespace *ns, struct ipc_params *params) |
1da177e4c
|
319 |
{ |
7748dbfaa
|
320 321 322 |
key_t key = params->key; int shmflg = params->flg; size_t size = params->u.size; |
1da177e4c
|
323 324 325 326 327 328 |
int error; struct shmid_kernel *shp; int numpages = (size + PAGE_SIZE -1) >> PAGE_SHIFT; struct file * file; char name[13]; int id; |
4e9823111
|
329 |
if (size < SHMMIN || size > ns->shm_ctlmax) |
1da177e4c
|
330 |
return -EINVAL; |
f66d45e99
|
331 |
if (ns->shm_tot + numpages > ns->shm_ctlall) |
1da177e4c
|
332 333 334 335 336 337 338 |
return -ENOSPC; shp = ipc_rcu_alloc(sizeof(*shp)); if (!shp) return -ENOMEM; shp->shm_perm.key = key; |
b33291c0b
|
339 |
shp->shm_perm.mode = (shmflg & S_IRWXUGO); |
1da177e4c
|
340 341 342 343 344 345 346 347 |
shp->mlock_user = NULL; shp->shm_perm.security = NULL; error = security_shm_alloc(shp); if (error) { ipc_rcu_putref(shp); return error; } |
9d66586f7
|
348 |
sprintf (name, "SYSV%08x", key); |
1da177e4c
|
349 |
if (shmflg & SHM_HUGETLB) { |
9d66586f7
|
350 351 |
/* hugetlb_file_setup takes care of mlock user accounting */ file = hugetlb_file_setup(name, size); |
1da177e4c
|
352 353 |
shp->mlock_user = current->user; } else { |
bf8f972d3
|
354 355 356 357 358 359 360 361 |
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
|
362 |
file = shmem_file_setup(name, size, acctflag); |
1da177e4c
|
363 364 365 366 |
} error = PTR_ERR(file); if (IS_ERR(file)) goto no_file; |
48dea404e
|
367 |
id = ipc_addid(&shm_ids(ns), &shp->shm_perm, ns->shm_ctlmni); |
283bb7fad
|
368 369 |
if (id < 0) { error = id; |
1da177e4c
|
370 |
goto no_id; |
283bb7fad
|
371 |
} |
1da177e4c
|
372 |
|
b488893a3
|
373 |
shp->shm_cprid = task_tgid_vnr(current); |
1da177e4c
|
374 375 376 377 378 |
shp->shm_lprid = 0; shp->shm_atim = shp->shm_dtim = 0; shp->shm_ctim = get_seconds(); shp->shm_segsz = size; shp->shm_nattch = 0; |
1da177e4c
|
379 |
shp->shm_file = file; |
30475cc12
|
380 381 382 383 |
/* * shmid gets reported as "inode#" in /proc/pid/maps. * proc-ps tools use this. Changing this will break them. */ |
7ca7e564e
|
384 |
file->f_dentry->d_inode->i_ino = shp->shm_perm.id; |
551110a94
|
385 |
|
4e9823111
|
386 |
ns->shm_tot += numpages; |
7ca7e564e
|
387 |
error = shp->shm_perm.id; |
1da177e4c
|
388 |
shm_unlock(shp); |
7ca7e564e
|
389 |
return error; |
1da177e4c
|
390 391 392 393 394 395 396 397 |
no_id: fput(file); no_file: security_shm_free(shp); ipc_rcu_putref(shp); return error; } |
f4566f048
|
398 |
/* |
3e148c799
|
399 |
* Called with shm_ids.rw_mutex and ipcp locked. |
f4566f048
|
400 |
*/ |
03f02c765
|
401 |
static inline int shm_security(struct kern_ipc_perm *ipcp, int shmflg) |
7748dbfaa
|
402 |
{ |
03f02c765
|
403 404 405 406 |
struct shmid_kernel *shp; shp = container_of(ipcp, struct shmid_kernel, shm_perm); return security_shm_associate(shp, shmflg); |
7748dbfaa
|
407 |
} |
f4566f048
|
408 |
/* |
3e148c799
|
409 |
* Called with shm_ids.rw_mutex and ipcp locked. |
f4566f048
|
410 |
*/ |
03f02c765
|
411 412 |
static inline int shm_more_checks(struct kern_ipc_perm *ipcp, struct ipc_params *params) |
7748dbfaa
|
413 |
{ |
03f02c765
|
414 415 416 417 |
struct shmid_kernel *shp; shp = container_of(ipcp, struct shmid_kernel, shm_perm); if (shp->shm_segsz < params->u.size) |
7748dbfaa
|
418 419 420 421 |
return -EINVAL; return 0; } |
1da177e4c
|
422 423 |
asmlinkage long sys_shmget (key_t key, size_t size, int shmflg) { |
4e9823111
|
424 |
struct ipc_namespace *ns; |
7748dbfaa
|
425 426 |
struct ipc_ops shm_ops; struct ipc_params shm_params; |
4e9823111
|
427 428 |
ns = current->nsproxy->ipc_ns; |
1da177e4c
|
429 |
|
7748dbfaa
|
430 431 432 |
shm_ops.getnew = newseg; shm_ops.associate = shm_security; shm_ops.more_checks = shm_more_checks; |
7ca7e564e
|
433 |
|
7748dbfaa
|
434 435 436 |
shm_params.key = key; shm_params.flg = shmflg; shm_params.u.size = size; |
1da177e4c
|
437 |
|
7748dbfaa
|
438 |
return ipcget(ns, &shm_ids(ns), &shm_ops, &shm_params); |
1da177e4c
|
439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 |
} 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; } } |
016d7132f
|
465 466 |
static inline unsigned long copy_shmid_from_user(struct shmid64_ds *out, void __user *buf, int version) |
1da177e4c
|
467 468 469 |
{ switch(version) { case IPC_64: |
016d7132f
|
470 |
if (copy_from_user(out, buf, sizeof(*out))) |
1da177e4c
|
471 |
return -EFAULT; |
1da177e4c
|
472 |
return 0; |
1da177e4c
|
473 474 475 476 477 478 |
case IPC_OLD: { struct shmid_ds tbuf_old; if (copy_from_user(&tbuf_old, buf, sizeof(tbuf_old))) return -EFAULT; |
016d7132f
|
479 480 481 |
out->shm_perm.uid = tbuf_old.shm_perm.uid; out->shm_perm.gid = tbuf_old.shm_perm.gid; out->shm_perm.mode = tbuf_old.shm_perm.mode; |
1da177e4c
|
482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 |
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
|
515 |
/* |
3e148c799
|
516 |
* Called with shm_ids.rw_mutex held as a reader |
f4566f048
|
517 |
*/ |
4e9823111
|
518 519 |
static void shm_get_stat(struct ipc_namespace *ns, unsigned long *rss, unsigned long *swp) |
1da177e4c
|
520 |
{ |
7ca7e564e
|
521 522 |
int next_id; int total, in_use; |
1da177e4c
|
523 524 525 |
*rss = 0; *swp = 0; |
7ca7e564e
|
526 527 528 |
in_use = shm_ids(ns).in_use; for (total = 0, next_id = 0; total < in_use; next_id++) { |
1da177e4c
|
529 530 |
struct shmid_kernel *shp; struct inode *inode; |
637c36634
|
531 |
shp = idr_find(&shm_ids(ns).ipcs_idr, next_id); |
7ca7e564e
|
532 |
if (shp == NULL) |
1da177e4c
|
533 |
continue; |
6d63079ad
|
534 |
inode = shp->shm_file->f_path.dentry->d_inode; |
1da177e4c
|
535 536 537 |
if (is_file_hugepages(shp->shm_file)) { struct address_space *mapping = inode->i_mapping; |
a55164389
|
538 539 |
struct hstate *h = hstate_file(shp->shm_file); *rss += pages_per_huge_page(h) * mapping->nrpages; |
1da177e4c
|
540 541 542 543 544 545 546 |
} 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
|
547 548 |
total++; |
1da177e4c
|
549 550 |
} } |
8d4cc8b5c
|
551 552 553 554 555 556 557 |
/* * This function handles some shmctl commands which require the rw_mutex * to be held in write mode. * NOTE: no locks must be held, the rw_mutex is taken inside this function. */ static int shmctl_down(struct ipc_namespace *ns, int shmid, int cmd, struct shmid_ds __user *buf, int version) |
1da177e4c
|
558 |
{ |
8d4cc8b5c
|
559 |
struct kern_ipc_perm *ipcp; |
016d7132f
|
560 |
struct shmid64_ds shmid64; |
1da177e4c
|
561 |
struct shmid_kernel *shp; |
8d4cc8b5c
|
562 563 564 |
int err; if (cmd == IPC_SET) { |
016d7132f
|
565 |
if (copy_shmid_from_user(&shmid64, buf, version)) |
8d4cc8b5c
|
566 567 |
return -EFAULT; } |
a5f75e7f2
|
568 569 570 |
ipcp = ipcctl_pre_down(&shm_ids(ns), shmid, cmd, &shmid64.shm_perm, 0); if (IS_ERR(ipcp)) return PTR_ERR(ipcp); |
8d4cc8b5c
|
571 |
|
a5f75e7f2
|
572 |
shp = container_of(ipcp, struct shmid_kernel, shm_perm); |
8d4cc8b5c
|
573 574 575 576 577 578 579 580 581 |
err = security_shm_shmctl(shp, cmd); if (err) goto out_unlock; switch (cmd) { case IPC_RMID: do_shm_rmid(ns, ipcp); goto out_up; case IPC_SET: |
8f4a3809c
|
582 |
ipc_update_perm(&shmid64.shm_perm, ipcp); |
8d4cc8b5c
|
583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 |
shp->shm_ctim = get_seconds(); break; default: err = -EINVAL; } out_unlock: shm_unlock(shp); out_up: up_write(&shm_ids(ns).rw_mutex); return err; } asmlinkage long sys_shmctl(int shmid, int cmd, struct shmid_ds __user *buf) { struct shmid_kernel *shp; |
1da177e4c
|
598 |
int err, version; |
4e9823111
|
599 |
struct ipc_namespace *ns; |
1da177e4c
|
600 601 602 603 604 605 606 |
if (cmd < 0 || shmid < 0) { err = -EINVAL; goto out; } version = ipc_parse_version(&cmd); |
4e9823111
|
607 |
ns = current->nsproxy->ipc_ns; |
1da177e4c
|
608 609 610 611 612 613 614 615 616 617 618 |
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
|
619 620 621 |
shminfo.shmmni = shminfo.shmseg = ns->shm_ctlmni; shminfo.shmmax = ns->shm_ctlmax; shminfo.shmall = ns->shm_ctlall; |
1da177e4c
|
622 623 624 625 |
shminfo.shmmin = SHMMIN; if(copy_shminfo_to_user (buf, &shminfo, version)) return -EFAULT; |
f4566f048
|
626 |
|
3e148c799
|
627 |
down_read(&shm_ids(ns).rw_mutex); |
7ca7e564e
|
628 |
err = ipc_get_maxid(&shm_ids(ns)); |
3e148c799
|
629 |
up_read(&shm_ids(ns).rw_mutex); |
f4566f048
|
630 |
|
1da177e4c
|
631 632 633 634 635 636 637 638 639 640 641 642 643 |
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
|
644 |
down_read(&shm_ids(ns).rw_mutex); |
4e9823111
|
645 646 647 |
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
|
648 649 |
shm_info.swap_attempts = 0; shm_info.swap_successes = 0; |
7ca7e564e
|
650 |
err = ipc_get_maxid(&shm_ids(ns)); |
3e148c799
|
651 |
up_read(&shm_ids(ns).rw_mutex); |
1da177e4c
|
652 653 654 655 656 657 658 659 660 661 662 663 664 |
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
|
665 666 667 |
if (!buf) { err = -EFAULT; |
1da177e4c
|
668 |
goto out; |
023a53557
|
669 670 671 672 673 674 675 676 |
} if (cmd == SHM_STAT) { shp = shm_lock(ns, shmid); if (IS_ERR(shp)) { err = PTR_ERR(shp); goto out; } |
7ca7e564e
|
677 |
result = shp->shm_perm.id; |
1da177e4c
|
678 |
} else { |
023a53557
|
679 680 681 682 683 |
shp = shm_lock_check(ns, shmid); if (IS_ERR(shp)) { err = PTR_ERR(shp); goto out; } |
1da177e4c
|
684 685 686 687 688 689 690 691 |
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
|
692 |
memset(&tbuf, 0, sizeof(tbuf)); |
1da177e4c
|
693 694 695 696 697 698 699 |
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
|
700 |
tbuf.shm_nattch = shp->shm_nattch; |
1da177e4c
|
701 702 703 704 705 706 707 708 709 710 |
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
|
711 712 713 |
shp = shm_lock_check(ns, shmid); if (IS_ERR(shp)) { err = PTR_ERR(shp); |
1da177e4c
|
714 715 |
goto out; } |
1da177e4c
|
716 |
|
073115d6b
|
717 718 719 |
err = audit_ipc_obj(&(shp->shm_perm)); if (err) goto out_unlock; |
1da177e4c
|
720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 |
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
|
738 |
if (!err && !(shp->shm_perm.mode & SHM_LOCKED)){ |
b33291c0b
|
739 |
shp->shm_perm.mode |= SHM_LOCKED; |
1da177e4c
|
740 741 742 743 744 |
shp->mlock_user = user; } } } else if (!is_file_hugepages(shp->shm_file)) { shmem_lock(shp->shm_file, 0, shp->mlock_user); |
b33291c0b
|
745 |
shp->shm_perm.mode &= ~SHM_LOCKED; |
1da177e4c
|
746 747 748 749 750 751 |
shp->mlock_user = NULL; } shm_unlock(shp); goto out; } case IPC_RMID: |
1da177e4c
|
752 |
case IPC_SET: |
8d4cc8b5c
|
753 754 |
err = shmctl_down(ns, shmid, cmd, buf, version); return err; |
1da177e4c
|
755 |
default: |
8d4cc8b5c
|
756 |
return -EINVAL; |
1da177e4c
|
757 |
} |
1da177e4c
|
758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 |
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
|
780 |
int acc_mode; |
bc56bba8f
|
781 |
unsigned long user_addr; |
4e9823111
|
782 |
struct ipc_namespace *ns; |
bc56bba8f
|
783 784 785 |
struct shm_file_data *sfd; struct path path; mode_t f_mode; |
1da177e4c
|
786 |
|
bc56bba8f
|
787 788 |
err = -EINVAL; if (shmid < 0) |
1da177e4c
|
789 |
goto out; |
bc56bba8f
|
790 |
else if ((addr = (ulong)shmaddr)) { |
1da177e4c
|
791 792 793 794 795 796 797 |
if (addr & (SHMLBA-1)) { if (shmflg & SHM_RND) addr &= ~(SHMLBA-1); /* round down */ else #ifndef __ARCH_FORCE_SHMLBA if (addr & ~PAGE_MASK) #endif |
bc56bba8f
|
798 |
goto out; |
1da177e4c
|
799 800 801 802 |
} flags = MAP_SHARED | MAP_FIXED; } else { if ((shmflg & SHM_REMAP)) |
bc56bba8f
|
803 |
goto out; |
1da177e4c
|
804 805 806 807 808 809 |
flags = MAP_SHARED; } if (shmflg & SHM_RDONLY) { prot = PROT_READ; |
1da177e4c
|
810 |
acc_mode = S_IRUGO; |
bc56bba8f
|
811 |
f_mode = FMODE_READ; |
1da177e4c
|
812 813 |
} else { prot = PROT_READ | PROT_WRITE; |
1da177e4c
|
814 |
acc_mode = S_IRUGO | S_IWUGO; |
bc56bba8f
|
815 |
f_mode = FMODE_READ | FMODE_WRITE; |
1da177e4c
|
816 817 818 819 820 821 822 823 824 825 |
} 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
|
826 |
ns = current->nsproxy->ipc_ns; |
023a53557
|
827 828 829 |
shp = shm_lock_check(ns, shmid); if (IS_ERR(shp)) { err = PTR_ERR(shp); |
1da177e4c
|
830 |
goto out; |
023a53557
|
831 |
} |
bc56bba8f
|
832 833 834 835 |
err = -EACCES; if (ipcperms(&shp->shm_perm, acc_mode)) goto out_unlock; |
1da177e4c
|
836 837 |
err = security_shm_shmat(shp, shmaddr, shmflg); |
bc56bba8f
|
838 839 840 841 |
if (err) goto out_unlock; path.dentry = dget(shp->shm_file->f_path.dentry); |
ce8d2cdf3
|
842 |
path.mnt = shp->shm_file->f_path.mnt; |
1da177e4c
|
843 |
shp->shm_nattch++; |
bc56bba8f
|
844 |
size = i_size_read(path.dentry->d_inode); |
1da177e4c
|
845 |
shm_unlock(shp); |
bc56bba8f
|
846 847 848 |
err = -ENOMEM; sfd = kzalloc(sizeof(*sfd), GFP_KERNEL); if (!sfd) |
ce8d2cdf3
|
849 |
goto out_put_dentry; |
bc56bba8f
|
850 |
|
ce8d2cdf3
|
851 |
file = alloc_file(path.mnt, path.dentry, f_mode, &shm_file_operations); |
bc56bba8f
|
852 853 |
if (!file) goto out_free; |
bc56bba8f
|
854 |
file->private_data = sfd; |
bc56bba8f
|
855 |
file->f_mapping = shp->shm_file->f_mapping; |
7ca7e564e
|
856 |
sfd->id = shp->shm_perm.id; |
bc56bba8f
|
857 858 859 |
sfd->ns = get_ipc_ns(ns); sfd->file = shp->shm_file; sfd->vm_ops = NULL; |
1da177e4c
|
860 861 |
down_write(¤t->mm->mmap_sem); if (addr && !(shmflg & SHM_REMAP)) { |
bc56bba8f
|
862 |
err = -EINVAL; |
1da177e4c
|
863 864 865 866 867 868 869 870 871 872 873 |
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
|
874 875 876 877 878 |
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
|
879 880 |
invalid: up_write(¤t->mm->mmap_sem); |
bc56bba8f
|
881 882 883 |
fput(file); out_nattch: |
3e148c799
|
884 885 |
down_write(&shm_ids(ns).rw_mutex); shp = shm_lock_down(ns, shmid); |
023a53557
|
886 |
BUG_ON(IS_ERR(shp)); |
1da177e4c
|
887 888 |
shp->shm_nattch--; if(shp->shm_nattch == 0 && |
b33291c0b
|
889 |
shp->shm_perm.mode & SHM_DEST) |
4e9823111
|
890 |
shm_destroy(ns, shp); |
1da177e4c
|
891 892 |
else shm_unlock(shp); |
3e148c799
|
893 |
up_write(&shm_ids(ns).rw_mutex); |
1da177e4c
|
894 |
|
1da177e4c
|
895 896 |
out: return err; |
bc56bba8f
|
897 898 899 900 901 902 903 |
out_unlock: shm_unlock(shp); goto out; out_free: kfree(sfd); |
ce8d2cdf3
|
904 |
out_put_dentry: |
bc56bba8f
|
905 |
dput(path.dentry); |
bc56bba8f
|
906 |
goto out_nattch; |
1da177e4c
|
907 |
} |
7d87e14c2
|
908 909 910 911 912 913 914 915 916 917 918 |
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
|
919 920 921 922 923 924 925 926 927 928 929 |
/* * 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
|
930 931 |
if (addr & ~PAGE_MASK) return retval; |
1da177e4c
|
932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 |
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
|
964 |
if ((vma->vm_ops == &shm_vm_ops) && |
1da177e4c
|
965 |
(vma->vm_start - addr)/PAGE_SIZE == vma->vm_pgoff) { |
6d63079ad
|
966 |
size = vma->vm_file->f_path.dentry->d_inode->i_size; |
1da177e4c
|
967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 |
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
|
986 |
size = PAGE_ALIGN(size); |
1da177e4c
|
987 988 989 990 |
while (vma && (loff_t)(vma->vm_end - addr) <= size) { next = vma->vm_next; /* finding a matching vma now does not alter retval */ |
bc56bba8f
|
991 |
if ((vma->vm_ops == &shm_vm_ops) && |
1da177e4c
|
992 993 994 995 996 997 998 999 1000 1001 1002 |
(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
|
1003 |
static int sysvipc_shm_proc_show(struct seq_file *s, void *it) |
1da177e4c
|
1004 |
{ |
19b4946ca
|
1005 |
struct shmid_kernel *shp = it; |
1da177e4c
|
1006 |
|
6c826818f
|
1007 1008 1009 1010 1011 |
#if BITS_PER_LONG <= 32 #define SIZE_SPEC "%10lu" #else #define SIZE_SPEC "%21lu" #endif |
1da177e4c
|
1012 |
|
6c826818f
|
1013 1014 1015 1016 |
return seq_printf(s, "%10d %10d %4o " SIZE_SPEC " %5u %5u " "%5lu %5u %5u %5u %5u %10lu %10lu %10lu ", |
19b4946ca
|
1017 |
shp->shm_perm.key, |
7ca7e564e
|
1018 |
shp->shm_perm.id, |
b33291c0b
|
1019 |
shp->shm_perm.mode, |
19b4946ca
|
1020 1021 1022 |
shp->shm_segsz, shp->shm_cprid, shp->shm_lprid, |
bc56bba8f
|
1023 |
shp->shm_nattch, |
19b4946ca
|
1024 1025 1026 1027 1028 1029 1030 |
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
|
1031 1032 |
} #endif |