Blame view
ipc/mqueue.c
35.3 KB
1da177e4c
|
1 2 3 4 |
/* * POSIX message queues filesystem for Linux. * * Copyright (C) 2003,2004 Krzysztof Benedyczak (golbi@mat.uni.torun.pl) |
f66e928b9
|
5 |
* Michal Wronski (michal.wronski@gmail.com) |
1da177e4c
|
6 7 8 |
* * Spinlocks: Mohamed Abbas (abbas.mohamed@intel.com) * Lockless receive & send, fd based notify: |
239521f31
|
9 |
* Manfred Spraul (manfred@colorfullife.com) |
1da177e4c
|
10 |
* |
20ca73bc7
|
11 12 |
* Audit: George Wilson (ltcgcw@us.ibm.com) * |
1da177e4c
|
13 14 |
* This file is released under the GPL. */ |
c59ede7b7
|
15 |
#include <linux/capability.h> |
1da177e4c
|
16 17 18 19 20 21 22 23 24 25 |
#include <linux/init.h> #include <linux/pagemap.h> #include <linux/file.h> #include <linux/mount.h> #include <linux/namei.h> #include <linux/sysctl.h> #include <linux/poll.h> #include <linux/mqueue.h> #include <linux/msg.h> #include <linux/skbuff.h> |
5b5c4d1a1
|
26 |
#include <linux/vmalloc.h> |
1da177e4c
|
27 28 |
#include <linux/netlink.h> #include <linux/syscalls.h> |
20ca73bc7
|
29 |
#include <linux/audit.h> |
7ed20e1ad
|
30 |
#include <linux/signal.h> |
5f921ae96
|
31 |
#include <linux/mutex.h> |
b488893a3
|
32 33 |
#include <linux/nsproxy.h> #include <linux/pid.h> |
614b84cf4
|
34 |
#include <linux/ipc_namespace.h> |
6b550f949
|
35 |
#include <linux/user_namespace.h> |
5a0e3ad6a
|
36 |
#include <linux/slab.h> |
5f921ae96
|
37 |
|
1da177e4c
|
38 39 40 41 42 43 44 45 46 47 48 49 50 |
#include <net/sock.h> #include "util.h" #define MQUEUE_MAGIC 0x19800202 #define DIRENT_SIZE 20 #define FILENT_SIZE 80 #define SEND 0 #define RECV 1 #define STATE_NONE 0 #define STATE_PENDING 1 #define STATE_READY 2 |
d6629859b
|
51 52 53 54 55 |
struct posix_msg_tree_node { struct rb_node rb_node; struct list_head msg_list; int priority; }; |
1da177e4c
|
56 57 58 59 60 61 62 63 64 65 66 |
struct ext_wait_queue { /* queue of sleeping tasks */ struct task_struct *task; struct list_head list; struct msg_msg *msg; /* ptr of loaded message */ int state; /* one of STATE_* values */ }; struct mqueue_inode_info { spinlock_t lock; struct inode vfs_inode; wait_queue_head_t wait_q; |
d6629859b
|
67 |
struct rb_root msg_tree; |
ce2d52cc1
|
68 |
struct posix_msg_tree_node *node_cache; |
1da177e4c
|
69 70 71 |
struct mq_attr attr; struct sigevent notify; |
239521f31
|
72 |
struct pid *notify_owner; |
6f9ac6d93
|
73 |
struct user_namespace *notify_user_ns; |
338cec325
|
74 |
struct user_struct *user; /* user who created, for accounting */ |
1da177e4c
|
75 76 77 78 79 80 81 82 |
struct sock *notify_sock; struct sk_buff *notify_cookie; /* for tasks waiting for free space and messages, respectively */ struct ext_wait_queue e_wait_q[2]; unsigned long qsize; /* size of queue in memory (sum of all msgs) */ }; |
92e1d5be9
|
83 |
static const struct inode_operations mqueue_dir_inode_operations; |
9a32144e9
|
84 |
static const struct file_operations mqueue_file_operations; |
b87221de6
|
85 |
static const struct super_operations mqueue_super_ops; |
1da177e4c
|
86 |
static void remove_notification(struct mqueue_inode_info *info); |
e18b890bb
|
87 |
static struct kmem_cache *mqueue_inode_cachep; |
1da177e4c
|
88 |
|
239521f31
|
89 |
static struct ctl_table_header *mq_sysctl_table; |
1da177e4c
|
90 91 92 93 94 |
static inline struct mqueue_inode_info *MQUEUE_I(struct inode *inode) { return container_of(inode, struct mqueue_inode_info, vfs_inode); } |
7eafd7c74
|
95 96 97 98 |
/* * This routine should be called with the mq_lock held. */ static inline struct ipc_namespace *__get_ns_from_inode(struct inode *inode) |
614b84cf4
|
99 |
{ |
7eafd7c74
|
100 |
return get_ipc_ns(inode->i_sb->s_fs_info); |
614b84cf4
|
101 |
} |
7eafd7c74
|
102 |
static struct ipc_namespace *get_ns_from_inode(struct inode *inode) |
614b84cf4
|
103 |
{ |
7eafd7c74
|
104 105 106 107 108 109 |
struct ipc_namespace *ns; spin_lock(&mq_lock); ns = __get_ns_from_inode(inode); spin_unlock(&mq_lock); return ns; |
614b84cf4
|
110 |
} |
d6629859b
|
111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 |
/* Auxiliary functions to manipulate messages' list */ static int msg_insert(struct msg_msg *msg, struct mqueue_inode_info *info) { struct rb_node **p, *parent = NULL; struct posix_msg_tree_node *leaf; p = &info->msg_tree.rb_node; while (*p) { parent = *p; leaf = rb_entry(parent, struct posix_msg_tree_node, rb_node); if (likely(leaf->priority == msg->m_type)) goto insert_msg; else if (msg->m_type < leaf->priority) p = &(*p)->rb_left; else p = &(*p)->rb_right; } |
ce2d52cc1
|
129 130 131 132 133 134 135 |
if (info->node_cache) { leaf = info->node_cache; info->node_cache = NULL; } else { leaf = kmalloc(sizeof(*leaf), GFP_ATOMIC); if (!leaf) return -ENOMEM; |
ce2d52cc1
|
136 137 138 |
INIT_LIST_HEAD(&leaf->msg_list); info->qsize += sizeof(*leaf); } |
d6629859b
|
139 140 141 |
leaf->priority = msg->m_type; rb_link_node(&leaf->rb_node, parent, p); rb_insert_color(&leaf->rb_node, &info->msg_tree); |
d6629859b
|
142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 |
insert_msg: info->attr.mq_curmsgs++; info->qsize += msg->m_ts; list_add_tail(&msg->m_list, &leaf->msg_list); return 0; } static inline struct msg_msg *msg_get(struct mqueue_inode_info *info) { struct rb_node **p, *parent = NULL; struct posix_msg_tree_node *leaf; struct msg_msg *msg; try_again: p = &info->msg_tree.rb_node; while (*p) { parent = *p; /* * During insert, low priorities go to the left and high to the * right. On receive, we want the highest priorities first, so * walk all the way to the right. */ p = &(*p)->rb_right; } if (!parent) { if (info->attr.mq_curmsgs) { pr_warn_once("Inconsistency in POSIX message queue, " "no tree element, but supposedly messages " "should exist! "); info->attr.mq_curmsgs = 0; } return NULL; } leaf = rb_entry(parent, struct posix_msg_tree_node, rb_node); |
ce2d52cc1
|
177 |
if (unlikely(list_empty(&leaf->msg_list))) { |
d6629859b
|
178 179 180 181 182 |
pr_warn_once("Inconsistency in POSIX message queue, " "empty leaf node but we haven't implemented " "lazy leaf delete! "); rb_erase(&leaf->rb_node, &info->msg_tree); |
ce2d52cc1
|
183 184 185 186 187 188 |
if (info->node_cache) { info->qsize -= sizeof(*leaf); kfree(leaf); } else { info->node_cache = leaf; } |
d6629859b
|
189 190 191 192 193 194 195 |
goto try_again; } else { msg = list_first_entry(&leaf->msg_list, struct msg_msg, m_list); list_del(&msg->m_list); if (list_empty(&leaf->msg_list)) { rb_erase(&leaf->rb_node, &info->msg_tree); |
ce2d52cc1
|
196 197 198 199 200 201 |
if (info->node_cache) { info->qsize -= sizeof(*leaf); kfree(leaf); } else { info->node_cache = leaf; } |
d6629859b
|
202 203 204 205 206 207 |
} } info->attr.mq_curmsgs--; info->qsize -= msg->m_ts; return msg; } |
7eafd7c74
|
208 |
static struct inode *mqueue_get_inode(struct super_block *sb, |
1b9d5ff76
|
209 |
struct ipc_namespace *ipc_ns, umode_t mode, |
7eafd7c74
|
210 |
struct mq_attr *attr) |
1da177e4c
|
211 |
{ |
86a264abe
|
212 |
struct user_struct *u = current_user(); |
1da177e4c
|
213 |
struct inode *inode; |
d40dcdb01
|
214 |
int ret = -ENOMEM; |
1da177e4c
|
215 216 |
inode = new_inode(sb); |
04715206c
|
217 218 219 220 221 222 223 224 225 226 227 |
if (!inode) goto err; inode->i_ino = get_next_ino(); inode->i_mode = mode; inode->i_uid = current_fsuid(); inode->i_gid = current_fsgid(); inode->i_mtime = inode->i_ctime = inode->i_atime = CURRENT_TIME; if (S_ISREG(mode)) { struct mqueue_inode_info *info; |
d6629859b
|
228 |
unsigned long mq_bytes, mq_treesize; |
04715206c
|
229 230 231 232 233 234 235 236 237 238 |
inode->i_fop = &mqueue_file_operations; inode->i_size = FILENT_SIZE; /* mqueue specific info */ info = MQUEUE_I(inode); spin_lock_init(&info->lock); init_waitqueue_head(&info->wait_q); INIT_LIST_HEAD(&info->e_wait_q[0].list); INIT_LIST_HEAD(&info->e_wait_q[1].list); info->notify_owner = NULL; |
6f9ac6d93
|
239 |
info->notify_user_ns = NULL; |
04715206c
|
240 241 |
info->qsize = 0; info->user = NULL; /* set when all is ok */ |
d6629859b
|
242 |
info->msg_tree = RB_ROOT; |
ce2d52cc1
|
243 |
info->node_cache = NULL; |
04715206c
|
244 |
memset(&info->attr, 0, sizeof(info->attr)); |
cef0184c1
|
245 246 247 248 |
info->attr.mq_maxmsg = min(ipc_ns->mq_msg_max, ipc_ns->mq_msg_default); info->attr.mq_msgsize = min(ipc_ns->mq_msgsize_max, ipc_ns->mq_msgsize_default); |
04715206c
|
249 250 251 252 |
if (attr) { info->attr.mq_maxmsg = attr->mq_maxmsg; info->attr.mq_msgsize = attr->mq_msgsize; } |
d6629859b
|
253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 |
/* * We used to allocate a static array of pointers and account * the size of that array as well as one msg_msg struct per * possible message into the queue size. That's no longer * accurate as the queue is now an rbtree and will grow and * shrink depending on usage patterns. We can, however, still * account one msg_msg struct per message, but the nodes are * allocated depending on priority usage, and most programs * only use one, or a handful, of priorities. However, since * this is pinned memory, we need to assume worst case, so * that means the min(mq_maxmsg, max_priorities) * struct * posix_msg_tree_node. */ mq_treesize = info->attr.mq_maxmsg * sizeof(struct msg_msg) + min_t(unsigned int, info->attr.mq_maxmsg, MQ_PRIO_MAX) * sizeof(struct posix_msg_tree_node); |
04715206c
|
269 |
|
d6629859b
|
270 271 |
mq_bytes = mq_treesize + (info->attr.mq_maxmsg * info->attr.mq_msgsize); |
1da177e4c
|
272 |
|
04715206c
|
273 274 |
spin_lock(&mq_lock); if (u->mq_bytes + mq_bytes < u->mq_bytes || |
2a4e64b8f
|
275 |
u->mq_bytes + mq_bytes > rlimit(RLIMIT_MSGQUEUE)) { |
04715206c
|
276 277 |
spin_unlock(&mq_lock); /* mqueue_evict_inode() releases info->messages */ |
d40dcdb01
|
278 |
ret = -EMFILE; |
04715206c
|
279 |
goto out_inode; |
1da177e4c
|
280 |
} |
04715206c
|
281 282 283 284 285 286 287 288 289 290 291 |
u->mq_bytes += mq_bytes; spin_unlock(&mq_lock); /* all is ok */ info->user = get_uid(u); } else if (S_ISDIR(mode)) { inc_nlink(inode); /* Some things misbehave if size == 0 on a directory */ inode->i_size = 2 * DIRENT_SIZE; inode->i_op = &mqueue_dir_inode_operations; inode->i_fop = &simple_dir_operations; |
1da177e4c
|
292 |
} |
04715206c
|
293 |
|
1da177e4c
|
294 295 |
return inode; out_inode: |
1da177e4c
|
296 |
iput(inode); |
04715206c
|
297 |
err: |
d40dcdb01
|
298 |
return ERR_PTR(ret); |
1da177e4c
|
299 300 301 302 303 |
} static int mqueue_fill_super(struct super_block *sb, void *data, int silent) { struct inode *inode; |
7eafd7c74
|
304 |
struct ipc_namespace *ns = data; |
1da177e4c
|
305 306 307 308 309 |
sb->s_blocksize = PAGE_CACHE_SIZE; sb->s_blocksize_bits = PAGE_CACHE_SHIFT; sb->s_magic = MQUEUE_MAGIC; sb->s_op = &mqueue_super_ops; |
48fde701a
|
310 311 312 |
inode = mqueue_get_inode(sb, ns, S_IFDIR | S_ISVTX | S_IRWXUGO, NULL); if (IS_ERR(inode)) return PTR_ERR(inode); |
1da177e4c
|
313 |
|
48fde701a
|
314 315 316 317 |
sb->s_root = d_make_root(inode); if (!sb->s_root) return -ENOMEM; return 0; |
1da177e4c
|
318 |
} |
ceefda693
|
319 |
static struct dentry *mqueue_mount(struct file_system_type *fs_type, |
454e2398b
|
320 |
int flags, const char *dev_name, |
ceefda693
|
321 |
void *data) |
1da177e4c
|
322 |
{ |
a636b702e
|
323 324 325 326 327 328 329 330 331 332 |
if (!(flags & MS_KERNMOUNT)) { struct ipc_namespace *ns = current->nsproxy->ipc_ns; /* Don't allow mounting unless the caller has CAP_SYS_ADMIN * over the ipc namespace. */ if (!ns_capable(ns->user_ns, CAP_SYS_ADMIN)) return ERR_PTR(-EPERM); data = ns; } |
ceefda693
|
333 |
return mount_ns(fs_type, flags, data, mqueue_fill_super); |
1da177e4c
|
334 |
} |
51cc50685
|
335 |
static void init_once(void *foo) |
1da177e4c
|
336 337 |
{ struct mqueue_inode_info *p = (struct mqueue_inode_info *) foo; |
a35afb830
|
338 |
inode_init_once(&p->vfs_inode); |
1da177e4c
|
339 340 341 342 343 |
} static struct inode *mqueue_alloc_inode(struct super_block *sb) { struct mqueue_inode_info *ei; |
e94b17660
|
344 |
ei = kmem_cache_alloc(mqueue_inode_cachep, GFP_KERNEL); |
1da177e4c
|
345 346 347 348 |
if (!ei) return NULL; return &ei->vfs_inode; } |
fa0d7e3de
|
349 |
static void mqueue_i_callback(struct rcu_head *head) |
1da177e4c
|
350 |
{ |
fa0d7e3de
|
351 |
struct inode *inode = container_of(head, struct inode, i_rcu); |
1da177e4c
|
352 353 |
kmem_cache_free(mqueue_inode_cachep, MQUEUE_I(inode)); } |
fa0d7e3de
|
354 355 356 357 |
static void mqueue_destroy_inode(struct inode *inode) { call_rcu(&inode->i_rcu, mqueue_i_callback); } |
6d8af64c1
|
358 |
static void mqueue_evict_inode(struct inode *inode) |
1da177e4c
|
359 360 361 |
{ struct mqueue_inode_info *info; struct user_struct *user; |
d6629859b
|
362 |
unsigned long mq_bytes, mq_treesize; |
7eafd7c74
|
363 |
struct ipc_namespace *ipc_ns; |
d6629859b
|
364 |
struct msg_msg *msg; |
1da177e4c
|
365 |
|
dbd5768f8
|
366 |
clear_inode(inode); |
6d8af64c1
|
367 368 |
if (S_ISDIR(inode->i_mode)) |
1da177e4c
|
369 |
return; |
6d8af64c1
|
370 |
|
7eafd7c74
|
371 |
ipc_ns = get_ns_from_inode(inode); |
1da177e4c
|
372 373 |
info = MQUEUE_I(inode); spin_lock(&info->lock); |
d6629859b
|
374 375 |
while ((msg = msg_get(info)) != NULL) free_msg(msg); |
ce2d52cc1
|
376 |
kfree(info->node_cache); |
1da177e4c
|
377 |
spin_unlock(&info->lock); |
8834cf796
|
378 |
/* Total amount of bytes accounted for the mqueue */ |
d6629859b
|
379 380 381 382 383 384 |
mq_treesize = info->attr.mq_maxmsg * sizeof(struct msg_msg) + min_t(unsigned int, info->attr.mq_maxmsg, MQ_PRIO_MAX) * sizeof(struct posix_msg_tree_node); mq_bytes = mq_treesize + (info->attr.mq_maxmsg * info->attr.mq_msgsize); |
1da177e4c
|
385 386 387 388 |
user = info->user; if (user) { spin_lock(&mq_lock); user->mq_bytes -= mq_bytes; |
7eafd7c74
|
389 390 391 392 393 394 395 396 |
/* * get_ns_from_inode() ensures that the * (ipc_ns = sb->s_fs_info) is either a valid ipc_ns * to which we now hold a reference, or it is NULL. * We can't put it here under mq_lock, though. */ if (ipc_ns) ipc_ns->mq_queues_count--; |
1da177e4c
|
397 398 399 |
spin_unlock(&mq_lock); free_uid(user); } |
7eafd7c74
|
400 401 |
if (ipc_ns) put_ipc_ns(ipc_ns); |
1da177e4c
|
402 403 404 |
} static int mqueue_create(struct inode *dir, struct dentry *dentry, |
ebfc3b49a
|
405 |
umode_t mode, bool excl) |
1da177e4c
|
406 407 408 409 |
{ struct inode *inode; struct mq_attr *attr = dentry->d_fsdata; int error; |
7eafd7c74
|
410 |
struct ipc_namespace *ipc_ns; |
1da177e4c
|
411 412 |
spin_lock(&mq_lock); |
7eafd7c74
|
413 414 415 416 417 |
ipc_ns = __get_ns_from_inode(dir); if (!ipc_ns) { error = -EACCES; goto out_unlock; } |
f3713fd9c
|
418 419 420 |
if (ipc_ns->mq_queues_count >= ipc_ns->mq_queues_max && !capable(CAP_SYS_RESOURCE)) { |
1da177e4c
|
421 |
error = -ENOSPC; |
614b84cf4
|
422 |
goto out_unlock; |
1da177e4c
|
423 |
} |
614b84cf4
|
424 |
ipc_ns->mq_queues_count++; |
1da177e4c
|
425 |
spin_unlock(&mq_lock); |
7eafd7c74
|
426 |
inode = mqueue_get_inode(dir->i_sb, ipc_ns, mode, attr); |
d40dcdb01
|
427 428 |
if (IS_ERR(inode)) { error = PTR_ERR(inode); |
1da177e4c
|
429 |
spin_lock(&mq_lock); |
614b84cf4
|
430 431 |
ipc_ns->mq_queues_count--; goto out_unlock; |
1da177e4c
|
432 |
} |
7eafd7c74
|
433 |
put_ipc_ns(ipc_ns); |
1da177e4c
|
434 435 436 437 438 439 |
dir->i_size += DIRENT_SIZE; dir->i_ctime = dir->i_mtime = dir->i_atime = CURRENT_TIME; d_instantiate(dentry, inode); dget(dentry); return 0; |
614b84cf4
|
440 |
out_unlock: |
1da177e4c
|
441 |
spin_unlock(&mq_lock); |
7eafd7c74
|
442 443 |
if (ipc_ns) put_ipc_ns(ipc_ns); |
1da177e4c
|
444 445 446 447 448 |
return error; } static int mqueue_unlink(struct inode *dir, struct dentry *dentry) { |
239521f31
|
449 |
struct inode *inode = dentry->d_inode; |
1da177e4c
|
450 451 452 |
dir->i_ctime = dir->i_mtime = dir->i_atime = CURRENT_TIME; dir->i_size -= DIRENT_SIZE; |
239521f31
|
453 454 455 |
drop_nlink(inode); dput(dentry); return 0; |
1da177e4c
|
456 457 458 459 460 461 462 463 464 465 |
} /* * This is routine for system read from queue file. * To avoid mess with doing here some sort of mq_receive we allow * to read only queue size & notification info (the only values * that are interesting from user point of view and aren't accessible * through std routines) */ static ssize_t mqueue_read_file(struct file *filp, char __user *u_data, |
f1a43f93f
|
466 |
size_t count, loff_t *off) |
1da177e4c
|
467 |
{ |
496ad9aa8
|
468 |
struct mqueue_inode_info *info = MQUEUE_I(file_inode(filp)); |
1da177e4c
|
469 |
char buffer[FILENT_SIZE]; |
f1a43f93f
|
470 |
ssize_t ret; |
1da177e4c
|
471 472 473 474 475 476 477 478 479 480 |
spin_lock(&info->lock); snprintf(buffer, sizeof(buffer), "QSIZE:%-10lu NOTIFY:%-5d SIGNO:%-5d NOTIFY_PID:%-6d ", info->qsize, info->notify_owner ? info->notify.sigev_notify : 0, (info->notify_owner && info->notify.sigev_notify == SIGEV_SIGNAL) ? info->notify.sigev_signo : 0, |
6c5f3e7b4
|
481 |
pid_vnr(info->notify_owner)); |
1da177e4c
|
482 483 |
spin_unlock(&info->lock); buffer[sizeof(buffer)-1] = '\0'; |
1da177e4c
|
484 |
|
f1a43f93f
|
485 486 487 488 |
ret = simple_read_from_buffer(u_data, count, off, buffer, strlen(buffer)); if (ret <= 0) return ret; |
1da177e4c
|
489 |
|
496ad9aa8
|
490 |
file_inode(filp)->i_atime = file_inode(filp)->i_ctime = CURRENT_TIME; |
f1a43f93f
|
491 |
return ret; |
1da177e4c
|
492 |
} |
75e1fcc0b
|
493 |
static int mqueue_flush_file(struct file *filp, fl_owner_t id) |
1da177e4c
|
494 |
{ |
496ad9aa8
|
495 |
struct mqueue_inode_info *info = MQUEUE_I(file_inode(filp)); |
1da177e4c
|
496 497 |
spin_lock(&info->lock); |
a03fcb730
|
498 |
if (task_tgid(current) == info->notify_owner) |
1da177e4c
|
499 500 501 502 503 504 505 506 |
remove_notification(info); spin_unlock(&info->lock); return 0; } static unsigned int mqueue_poll_file(struct file *filp, struct poll_table_struct *poll_tab) { |
496ad9aa8
|
507 |
struct mqueue_inode_info *info = MQUEUE_I(file_inode(filp)); |
1da177e4c
|
508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 |
int retval = 0; poll_wait(filp, &info->wait_q, poll_tab); spin_lock(&info->lock); if (info->attr.mq_curmsgs) retval = POLLIN | POLLRDNORM; if (info->attr.mq_curmsgs < info->attr.mq_maxmsg) retval |= POLLOUT | POLLWRNORM; spin_unlock(&info->lock); return retval; } /* Adds current to info->e_wait_q[sr] before element with smaller prio */ static void wq_add(struct mqueue_inode_info *info, int sr, struct ext_wait_queue *ewp) { struct ext_wait_queue *walk; ewp->task = current; list_for_each_entry(walk, &info->e_wait_q[sr].list, list) { if (walk->task->static_prio <= current->static_prio) { list_add_tail(&ewp->list, &walk->list); return; } } list_add_tail(&ewp->list, &info->e_wait_q[sr].list); } /* * Puts current task to sleep. Caller must hold queue lock. After return * lock isn't held. * sr: SEND or RECV */ static int wq_sleep(struct mqueue_inode_info *info, int sr, |
9ca7d8e68
|
546 |
ktime_t *timeout, struct ext_wait_queue *ewp) |
1da177e4c
|
547 548 549 550 551 552 553 554 555 556 |
{ int retval; signed long time; wq_add(info, sr, ewp); for (;;) { set_current_state(TASK_INTERRUPTIBLE); spin_unlock(&info->lock); |
32ea845d5
|
557 558 |
time = schedule_hrtimeout_range_clock(timeout, 0, HRTIMER_MODE_ABS, CLOCK_REALTIME); |
1da177e4c
|
559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 |
while (ewp->state == STATE_PENDING) cpu_relax(); if (ewp->state == STATE_READY) { retval = 0; goto out; } spin_lock(&info->lock); if (ewp->state == STATE_READY) { retval = 0; goto out_unlock; } if (signal_pending(current)) { retval = -ERESTARTSYS; break; } if (time == 0) { retval = -ETIMEDOUT; break; } } list_del(&ewp->list); out_unlock: spin_unlock(&info->lock); out: return retval; } /* * Returns waiting task that should be serviced first or NULL if none exists */ static struct ext_wait_queue *wq_get_first_waiter( struct mqueue_inode_info *info, int sr) { struct list_head *ptr; ptr = info->e_wait_q[sr].list.prev; if (ptr == &info->e_wait_q[sr].list) return NULL; return list_entry(ptr, struct ext_wait_queue, list); } |
1da177e4c
|
601 602 603 |
static inline void set_cookie(struct sk_buff *skb, char code) { |
239521f31
|
604 |
((char *)skb->data)[NOTIFY_COOKIE_LEN-1] = code; |
1da177e4c
|
605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 |
} /* * The next function is only to split too long sys_mq_timedsend */ static void __do_notify(struct mqueue_inode_info *info) { /* notification * invoked when there is registered process and there isn't process * waiting synchronously for message AND state of queue changed from * empty to not empty. Here we are sure that no one is waiting * synchronously. */ if (info->notify_owner && info->attr.mq_curmsgs == 1) { struct siginfo sig_i; switch (info->notify.sigev_notify) { case SIGEV_NONE: break; case SIGEV_SIGNAL: /* sends signal */ sig_i.si_signo = info->notify.sigev_signo; sig_i.si_errno = 0; sig_i.si_code = SI_MESGQ; sig_i.si_value = info->notify.sigev_value; |
6b550f949
|
630 631 |
/* map current pid/uid into info->owner's namespaces */ rcu_read_lock(); |
a6684999f
|
632 633 |
sig_i.si_pid = task_tgid_nr_ns(current, ns_of_pid(info->notify_owner)); |
76b6db010
|
634 |
sig_i.si_uid = from_kuid_munged(info->notify_user_ns, current_uid()); |
6b550f949
|
635 |
rcu_read_unlock(); |
1da177e4c
|
636 |
|
a03fcb730
|
637 638 |
kill_pid_info(info->notify.sigev_signo, &sig_i, info->notify_owner); |
1da177e4c
|
639 640 641 |
break; case SIGEV_THREAD: set_cookie(info->notify_cookie, NOTIFY_WOKENUP); |
7ee015e0f
|
642 |
netlink_sendskb(info->notify_sock, info->notify_cookie); |
1da177e4c
|
643 644 645 |
break; } /* after notification unregisters process */ |
a03fcb730
|
646 |
put_pid(info->notify_owner); |
6f9ac6d93
|
647 |
put_user_ns(info->notify_user_ns); |
a03fcb730
|
648 |
info->notify_owner = NULL; |
6f9ac6d93
|
649 |
info->notify_user_ns = NULL; |
1da177e4c
|
650 651 652 |
} wake_up(&info->wait_q); } |
9ca7d8e68
|
653 654 |
static int prepare_timeout(const struct timespec __user *u_abs_timeout, ktime_t *expires, struct timespec *ts) |
1da177e4c
|
655 |
{ |
9ca7d8e68
|
656 657 658 659 |
if (copy_from_user(ts, u_abs_timeout, sizeof(struct timespec))) return -EFAULT; if (!timespec_valid(ts)) return -EINVAL; |
1da177e4c
|
660 |
|
9ca7d8e68
|
661 662 |
*expires = timespec_to_ktime(*ts); return 0; |
1da177e4c
|
663 664 665 666 |
} static void remove_notification(struct mqueue_inode_info *info) { |
a03fcb730
|
667 |
if (info->notify_owner != NULL && |
1da177e4c
|
668 669 |
info->notify.sigev_notify == SIGEV_THREAD) { set_cookie(info->notify_cookie, NOTIFY_REMOVED); |
7ee015e0f
|
670 |
netlink_sendskb(info->notify_sock, info->notify_cookie); |
1da177e4c
|
671 |
} |
a03fcb730
|
672 |
put_pid(info->notify_owner); |
6f9ac6d93
|
673 |
put_user_ns(info->notify_user_ns); |
a03fcb730
|
674 |
info->notify_owner = NULL; |
6f9ac6d93
|
675 |
info->notify_user_ns = NULL; |
1da177e4c
|
676 |
} |
614b84cf4
|
677 |
static int mq_attr_ok(struct ipc_namespace *ipc_ns, struct mq_attr *attr) |
1da177e4c
|
678 |
{ |
2c12ea498
|
679 680 |
int mq_treesize; unsigned long total_size; |
1da177e4c
|
681 |
if (attr->mq_maxmsg <= 0 || attr->mq_msgsize <= 0) |
113289cc0
|
682 |
return -EINVAL; |
1da177e4c
|
683 |
if (capable(CAP_SYS_RESOURCE)) { |
02967ea08
|
684 685 |
if (attr->mq_maxmsg > HARD_MSGMAX || attr->mq_msgsize > HARD_MSGSIZEMAX) |
113289cc0
|
686 |
return -EINVAL; |
1da177e4c
|
687 |
} else { |
614b84cf4
|
688 689 |
if (attr->mq_maxmsg > ipc_ns->mq_msg_max || attr->mq_msgsize > ipc_ns->mq_msgsize_max) |
113289cc0
|
690 |
return -EINVAL; |
1da177e4c
|
691 692 693 |
} /* check for overflow */ if (attr->mq_msgsize > ULONG_MAX/attr->mq_maxmsg) |
113289cc0
|
694 |
return -EOVERFLOW; |
2c12ea498
|
695 696 697 698 699 |
mq_treesize = attr->mq_maxmsg * sizeof(struct msg_msg) + min_t(unsigned int, attr->mq_maxmsg, MQ_PRIO_MAX) * sizeof(struct posix_msg_tree_node); total_size = attr->mq_maxmsg * attr->mq_msgsize; if (total_size + mq_treesize < total_size) |
113289cc0
|
700 701 |
return -EOVERFLOW; return 0; |
1da177e4c
|
702 703 704 705 706 |
} /* * Invoked when creating a new queue via sys_mq_open */ |
765927b2d
|
707 708 |
static struct file *do_create(struct ipc_namespace *ipc_ns, struct inode *dir, struct path *path, int oflag, umode_t mode, |
614b84cf4
|
709 |
struct mq_attr *attr) |
1da177e4c
|
710 |
{ |
745ca2475
|
711 |
const struct cred *cred = current_cred(); |
1da177e4c
|
712 |
int ret; |
564f6993f
|
713 |
if (attr) { |
113289cc0
|
714 715 |
ret = mq_attr_ok(ipc_ns, attr); if (ret) |
765927b2d
|
716 |
return ERR_PTR(ret); |
1da177e4c
|
717 |
/* store for use during create */ |
765927b2d
|
718 |
path->dentry->d_fsdata = attr; |
113289cc0
|
719 720 721 722 723 724 725 726 727 |
} else { struct mq_attr def_attr; def_attr.mq_maxmsg = min(ipc_ns->mq_msg_max, ipc_ns->mq_msg_default); def_attr.mq_msgsize = min(ipc_ns->mq_msgsize_max, ipc_ns->mq_msgsize_default); ret = mq_attr_ok(ipc_ns, &def_attr); if (ret) |
765927b2d
|
728 |
return ERR_PTR(ret); |
1da177e4c
|
729 |
} |
ce3b0f8d5
|
730 |
mode &= ~current_umask(); |
765927b2d
|
731 732 |
ret = vfs_create(dir, path->dentry, mode, true); path->dentry->d_fsdata = NULL; |
312b90fbe
|
733 734 735 |
if (ret) return ERR_PTR(ret); return dentry_open(path, oflag, cred); |
1da177e4c
|
736 737 738 |
} /* Opens existing queue */ |
765927b2d
|
739 |
static struct file *do_open(struct path *path, int oflag) |
1da177e4c
|
740 |
{ |
745ca2475
|
741 742 |
static const int oflag2acc[O_ACCMODE] = { MAY_READ, MAY_WRITE, MAY_READ | MAY_WRITE }; |
765927b2d
|
743 744 745 746 747 748 749 |
int acc; if ((oflag & O_ACCMODE) == (O_RDWR | O_WRONLY)) return ERR_PTR(-EINVAL); acc = oflag2acc[oflag & O_ACCMODE]; if (inode_permission(path->dentry->d_inode, acc)) return ERR_PTR(-EACCES); return dentry_open(path, oflag, current_cred()); |
1da177e4c
|
750 |
} |
df0a42837
|
751 |
SYSCALL_DEFINE4(mq_open, const char __user *, u_name, int, oflag, umode_t, mode, |
d5460c997
|
752 |
struct mq_attr __user *, u_attr) |
1da177e4c
|
753 |
{ |
765927b2d
|
754 |
struct path path; |
1da177e4c
|
755 |
struct file *filp; |
91a27b2a7
|
756 |
struct filename *name; |
564f6993f
|
757 |
struct mq_attr attr; |
1da177e4c
|
758 |
int fd, error; |
7eafd7c74
|
759 |
struct ipc_namespace *ipc_ns = current->nsproxy->ipc_ns; |
312b90fbe
|
760 761 762 |
struct vfsmount *mnt = ipc_ns->mq_mnt; struct dentry *root = mnt->mnt_root; int ro; |
1da177e4c
|
763 |
|
564f6993f
|
764 765 766 767 |
if (u_attr && copy_from_user(&attr, u_attr, sizeof(struct mq_attr))) return -EFAULT; audit_mq_open(oflag, mode, u_attr ? &attr : NULL); |
20ca73bc7
|
768 |
|
1da177e4c
|
769 770 |
if (IS_ERR(name = getname(u_name))) return PTR_ERR(name); |
269f21344
|
771 |
fd = get_unused_fd_flags(O_CLOEXEC); |
1da177e4c
|
772 773 |
if (fd < 0) goto out_putname; |
312b90fbe
|
774 |
ro = mnt_want_write(mnt); /* we'll drop it in any case */ |
765927b2d
|
775 776 |
error = 0; mutex_lock(&root->d_inode->i_mutex); |
91a27b2a7
|
777 |
path.dentry = lookup_one_len(name->name, root, strlen(name->name)); |
765927b2d
|
778 779 |
if (IS_ERR(path.dentry)) { error = PTR_ERR(path.dentry); |
4294a8eed
|
780 |
goto out_putfd; |
1da177e4c
|
781 |
} |
312b90fbe
|
782 |
path.mnt = mntget(mnt); |
1da177e4c
|
783 784 |
if (oflag & O_CREAT) { |
765927b2d
|
785 |
if (path.dentry->d_inode) { /* entry already exists */ |
adb5c2473
|
786 |
audit_inode(name, path.dentry, 0); |
8d8ffefaa
|
787 788 |
if (oflag & O_EXCL) { error = -EEXIST; |
7c7dce920
|
789 |
goto out; |
8d8ffefaa
|
790 |
} |
765927b2d
|
791 |
filp = do_open(&path, oflag); |
1da177e4c
|
792 |
} else { |
312b90fbe
|
793 794 795 796 |
if (ro) { error = ro; goto out; } |
79f6530cb
|
797 |
audit_inode_parent_hidden(name, root); |
765927b2d
|
798 799 |
filp = do_create(ipc_ns, root->d_inode, &path, oflag, mode, |
564f6993f
|
800 |
u_attr ? &attr : NULL); |
1da177e4c
|
801 |
} |
7c7dce920
|
802 |
} else { |
765927b2d
|
803 |
if (!path.dentry->d_inode) { |
8d8ffefaa
|
804 |
error = -ENOENT; |
7c7dce920
|
805 |
goto out; |
8d8ffefaa
|
806 |
} |
adb5c2473
|
807 |
audit_inode(name, path.dentry, 0); |
765927b2d
|
808 |
filp = do_open(&path, oflag); |
7c7dce920
|
809 |
} |
1da177e4c
|
810 |
|
765927b2d
|
811 812 813 |
if (!IS_ERR(filp)) fd_install(fd, filp); else |
1da177e4c
|
814 |
error = PTR_ERR(filp); |
7c7dce920
|
815 |
out: |
765927b2d
|
816 |
path_put(&path); |
7c7dce920
|
817 |
out_putfd: |
765927b2d
|
818 819 820 821 822 |
if (error) { put_unused_fd(fd); fd = error; } mutex_unlock(&root->d_inode->i_mutex); |
38d78e587
|
823 824 |
if (!ro) mnt_drop_write(mnt); |
1da177e4c
|
825 826 827 828 |
out_putname: putname(name); return fd; } |
d5460c997
|
829 |
SYSCALL_DEFINE1(mq_unlink, const char __user *, u_name) |
1da177e4c
|
830 831 |
{ int err; |
91a27b2a7
|
832 |
struct filename *name; |
1da177e4c
|
833 834 |
struct dentry *dentry; struct inode *inode = NULL; |
7eafd7c74
|
835 |
struct ipc_namespace *ipc_ns = current->nsproxy->ipc_ns; |
312b90fbe
|
836 |
struct vfsmount *mnt = ipc_ns->mq_mnt; |
1da177e4c
|
837 838 839 840 |
name = getname(u_name); if (IS_ERR(name)) return PTR_ERR(name); |
79f6530cb
|
841 |
audit_inode_parent_hidden(name, mnt->mnt_root); |
312b90fbe
|
842 843 844 845 |
err = mnt_want_write(mnt); if (err) goto out_name; mutex_lock_nested(&mnt->mnt_root->d_inode->i_mutex, I_MUTEX_PARENT); |
91a27b2a7
|
846 847 |
dentry = lookup_one_len(name->name, mnt->mnt_root, strlen(name->name)); |
1da177e4c
|
848 849 850 851 |
if (IS_ERR(dentry)) { err = PTR_ERR(dentry); goto out_unlock; } |
1da177e4c
|
852 |
inode = dentry->d_inode; |
312b90fbe
|
853 854 855 |
if (!inode) { err = -ENOENT; } else { |
7de9c6ee3
|
856 |
ihold(inode); |
b21996e36
|
857 |
err = vfs_unlink(dentry->d_parent->d_inode, dentry, NULL); |
312b90fbe
|
858 |
} |
1da177e4c
|
859 860 861 |
dput(dentry); out_unlock: |
312b90fbe
|
862 |
mutex_unlock(&mnt->mnt_root->d_inode->i_mutex); |
1da177e4c
|
863 864 |
if (inode) iput(inode); |
312b90fbe
|
865 866 867 |
mnt_drop_write(mnt); out_name: putname(name); |
1da177e4c
|
868 869 870 871 872 873 874 875 876 877 878 879 880 881 |
return err; } /* Pipelined send and receive functions. * * If a receiver finds no waiting message, then it registers itself in the * list of waiting receivers. A sender checks that list before adding the new * message into the message array. If there is a waiting receiver, then it * bypasses the message array and directly hands the message over to the * receiver. * The receiver accepts the message and returns without grabbing the queue * spinlock. Therefore an intermediate STATE_PENDING state and memory barriers * are necessary. The same algorithm is used for sysv semaphores, see |
7b7a317cf
|
882 |
* ipc/sem.c for more details. |
1da177e4c
|
883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 |
* * The same algorithm is used for senders. */ /* pipelined_send() - send a message directly to the task waiting in * sys_mq_timedreceive() (without inserting message into a queue). */ static inline void pipelined_send(struct mqueue_inode_info *info, struct msg_msg *message, struct ext_wait_queue *receiver) { receiver->msg = message; list_del(&receiver->list); receiver->state = STATE_PENDING; wake_up_process(receiver->task); |
d59dd4620
|
898 |
smp_wmb(); |
1da177e4c
|
899 900 901 902 903 904 905 906 907 908 909 910 911 912 |
receiver->state = STATE_READY; } /* pipelined_receive() - if there is task waiting in sys_mq_timedsend() * gets its message and put to the queue (we have one free place for sure). */ static inline void pipelined_receive(struct mqueue_inode_info *info) { struct ext_wait_queue *sender = wq_get_first_waiter(info, SEND); if (!sender) { /* for poll */ wake_up_interruptible(&info->wait_q); return; } |
d6629859b
|
913 914 |
if (msg_insert(sender->msg, info)) return; |
1da177e4c
|
915 916 917 |
list_del(&sender->list); sender->state = STATE_PENDING; wake_up_process(sender->task); |
d59dd4620
|
918 |
smp_wmb(); |
1da177e4c
|
919 920 |
sender->state = STATE_READY; } |
c4ea37c26
|
921 922 923 |
SYSCALL_DEFINE5(mq_timedsend, mqd_t, mqdes, const char __user *, u_msg_ptr, size_t, msg_len, unsigned int, msg_prio, const struct timespec __user *, u_abs_timeout) |
1da177e4c
|
924 |
{ |
2903ff019
|
925 |
struct fd f; |
1da177e4c
|
926 927 928 929 930 |
struct inode *inode; struct ext_wait_queue wait; struct ext_wait_queue *receiver; struct msg_msg *msg_ptr; struct mqueue_inode_info *info; |
9ca7d8e68
|
931 932 |
ktime_t expires, *timeout = NULL; struct timespec ts; |
ce2d52cc1
|
933 |
struct posix_msg_tree_node *new_leaf = NULL; |
2903ff019
|
934 |
int ret = 0; |
1da177e4c
|
935 |
|
c32c8af43
|
936 |
if (u_abs_timeout) { |
9ca7d8e68
|
937 938 939 940 |
int res = prepare_timeout(u_abs_timeout, &expires, &ts); if (res) return res; timeout = &expires; |
c32c8af43
|
941 |
} |
20ca73bc7
|
942 |
|
1da177e4c
|
943 944 |
if (unlikely(msg_prio >= (unsigned long) MQ_PRIO_MAX)) return -EINVAL; |
9ca7d8e68
|
945 |
audit_mq_sendrecv(mqdes, msg_len, msg_prio, timeout ? &ts : NULL); |
1da177e4c
|
946 |
|
2903ff019
|
947 948 |
f = fdget(mqdes); if (unlikely(!f.file)) { |
8d8ffefaa
|
949 |
ret = -EBADF; |
1da177e4c
|
950 |
goto out; |
8d8ffefaa
|
951 |
} |
1da177e4c
|
952 |
|
496ad9aa8
|
953 |
inode = file_inode(f.file); |
2903ff019
|
954 |
if (unlikely(f.file->f_op != &mqueue_file_operations)) { |
8d8ffefaa
|
955 |
ret = -EBADF; |
1da177e4c
|
956 |
goto out_fput; |
8d8ffefaa
|
957 |
} |
1da177e4c
|
958 |
info = MQUEUE_I(inode); |
bfcec7087
|
959 |
audit_inode(NULL, f.file->f_path.dentry, 0); |
1da177e4c
|
960 |
|
2903ff019
|
961 |
if (unlikely(!(f.file->f_mode & FMODE_WRITE))) { |
8d8ffefaa
|
962 |
ret = -EBADF; |
1da177e4c
|
963 |
goto out_fput; |
8d8ffefaa
|
964 |
} |
1da177e4c
|
965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 |
if (unlikely(msg_len > info->attr.mq_msgsize)) { ret = -EMSGSIZE; goto out_fput; } /* First try to allocate memory, before doing anything with * existing queues. */ msg_ptr = load_msg(u_msg_ptr, msg_len); if (IS_ERR(msg_ptr)) { ret = PTR_ERR(msg_ptr); goto out_fput; } msg_ptr->m_ts = msg_len; msg_ptr->m_type = msg_prio; |
ce2d52cc1
|
980 981 982 983 984 985 986 |
/* * msg_insert really wants us to have a valid, spare node struct so * it doesn't have to kmalloc a GFP_ATOMIC allocation, but it will * fall back to that if necessary. */ if (!info->node_cache) new_leaf = kmalloc(sizeof(*new_leaf), GFP_KERNEL); |
1da177e4c
|
987 |
spin_lock(&info->lock); |
ce2d52cc1
|
988 989 |
if (!info->node_cache && new_leaf) { /* Save our speculative allocation into the cache */ |
ce2d52cc1
|
990 991 992 993 994 995 996 |
INIT_LIST_HEAD(&new_leaf->msg_list); info->node_cache = new_leaf; info->qsize += sizeof(*new_leaf); new_leaf = NULL; } else { kfree(new_leaf); } |
1da177e4c
|
997 |
if (info->attr.mq_curmsgs == info->attr.mq_maxmsg) { |
2903ff019
|
998 |
if (f.file->f_flags & O_NONBLOCK) { |
1da177e4c
|
999 |
ret = -EAGAIN; |
1da177e4c
|
1000 1001 1002 1003 1004 |
} else { wait.task = current; wait.msg = (void *) msg_ptr; wait.state = STATE_NONE; ret = wq_sleep(info, SEND, timeout, &wait); |
ce2d52cc1
|
1005 1006 1007 1008 1009 |
/* * wq_sleep must be called with info->lock held, and * returns with the lock released */ goto out_free; |
1da177e4c
|
1010 |
} |
1da177e4c
|
1011 1012 1013 1014 1015 1016 |
} else { receiver = wq_get_first_waiter(info, RECV); if (receiver) { pipelined_send(info, msg_ptr, receiver); } else { /* adds message to the queue */ |
ce2d52cc1
|
1017 1018 1019 |
ret = msg_insert(msg_ptr, info); if (ret) goto out_unlock; |
1da177e4c
|
1020 1021 1022 1023 |
__do_notify(info); } inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; |
1da177e4c
|
1024 |
} |
ce2d52cc1
|
1025 1026 1027 1028 1029 |
out_unlock: spin_unlock(&info->lock); out_free: if (ret) free_msg(msg_ptr); |
1da177e4c
|
1030 |
out_fput: |
2903ff019
|
1031 |
fdput(f); |
1da177e4c
|
1032 1033 1034 |
out: return ret; } |
c4ea37c26
|
1035 1036 1037 |
SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr, size_t, msg_len, unsigned int __user *, u_msg_prio, const struct timespec __user *, u_abs_timeout) |
1da177e4c
|
1038 |
{ |
1da177e4c
|
1039 1040 |
ssize_t ret; struct msg_msg *msg_ptr; |
2903ff019
|
1041 |
struct fd f; |
1da177e4c
|
1042 1043 1044 |
struct inode *inode; struct mqueue_inode_info *info; struct ext_wait_queue wait; |
9ca7d8e68
|
1045 1046 |
ktime_t expires, *timeout = NULL; struct timespec ts; |
ce2d52cc1
|
1047 |
struct posix_msg_tree_node *new_leaf = NULL; |
1da177e4c
|
1048 |
|
c32c8af43
|
1049 |
if (u_abs_timeout) { |
9ca7d8e68
|
1050 1051 1052 1053 |
int res = prepare_timeout(u_abs_timeout, &expires, &ts); if (res) return res; timeout = &expires; |
c32c8af43
|
1054 |
} |
20ca73bc7
|
1055 |
|
9ca7d8e68
|
1056 |
audit_mq_sendrecv(mqdes, msg_len, 0, timeout ? &ts : NULL); |
1da177e4c
|
1057 |
|
2903ff019
|
1058 1059 |
f = fdget(mqdes); if (unlikely(!f.file)) { |
8d8ffefaa
|
1060 |
ret = -EBADF; |
1da177e4c
|
1061 |
goto out; |
8d8ffefaa
|
1062 |
} |
1da177e4c
|
1063 |
|
496ad9aa8
|
1064 |
inode = file_inode(f.file); |
2903ff019
|
1065 |
if (unlikely(f.file->f_op != &mqueue_file_operations)) { |
8d8ffefaa
|
1066 |
ret = -EBADF; |
1da177e4c
|
1067 |
goto out_fput; |
8d8ffefaa
|
1068 |
} |
1da177e4c
|
1069 |
info = MQUEUE_I(inode); |
bfcec7087
|
1070 |
audit_inode(NULL, f.file->f_path.dentry, 0); |
1da177e4c
|
1071 |
|
2903ff019
|
1072 |
if (unlikely(!(f.file->f_mode & FMODE_READ))) { |
8d8ffefaa
|
1073 |
ret = -EBADF; |
1da177e4c
|
1074 |
goto out_fput; |
8d8ffefaa
|
1075 |
} |
1da177e4c
|
1076 1077 1078 1079 1080 1081 |
/* checks if buffer is big enough */ if (unlikely(msg_len < info->attr.mq_msgsize)) { ret = -EMSGSIZE; goto out_fput; } |
ce2d52cc1
|
1082 1083 1084 1085 1086 1087 1088 |
/* * msg_insert really wants us to have a valid, spare node struct so * it doesn't have to kmalloc a GFP_ATOMIC allocation, but it will * fall back to that if necessary. */ if (!info->node_cache) new_leaf = kmalloc(sizeof(*new_leaf), GFP_KERNEL); |
1da177e4c
|
1089 |
spin_lock(&info->lock); |
ce2d52cc1
|
1090 1091 1092 |
if (!info->node_cache && new_leaf) { /* Save our speculative allocation into the cache */ |
ce2d52cc1
|
1093 1094 1095 1096 1097 1098 |
INIT_LIST_HEAD(&new_leaf->msg_list); info->node_cache = new_leaf; info->qsize += sizeof(*new_leaf); } else { kfree(new_leaf); } |
1da177e4c
|
1099 |
if (info->attr.mq_curmsgs == 0) { |
2903ff019
|
1100 |
if (f.file->f_flags & O_NONBLOCK) { |
1da177e4c
|
1101 1102 |
spin_unlock(&info->lock); ret = -EAGAIN; |
1da177e4c
|
1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 |
} else { wait.task = current; wait.state = STATE_NONE; ret = wq_sleep(info, RECV, timeout, &wait); msg_ptr = wait.msg; } } else { msg_ptr = msg_get(info); inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME; /* There is now free space in queue. */ pipelined_receive(info); spin_unlock(&info->lock); ret = 0; } if (ret == 0) { ret = msg_ptr->m_ts; if ((u_msg_prio && put_user(msg_ptr->m_type, u_msg_prio)) || store_msg(u_msg_ptr, msg_ptr, msg_ptr->m_ts)) { ret = -EFAULT; } free_msg(msg_ptr); } out_fput: |
2903ff019
|
1130 |
fdput(f); |
1da177e4c
|
1131 1132 1133 1134 1135 1136 1137 1138 1139 |
out: return ret; } /* * Notes: the case when user wants us to deregister (with NULL as pointer) * and he isn't currently owner of notification, will be silently discarded. * It isn't explicitly defined in the POSIX. */ |
c4ea37c26
|
1140 1141 |
SYSCALL_DEFINE2(mq_notify, mqd_t, mqdes, const struct sigevent __user *, u_notification) |
1da177e4c
|
1142 |
{ |
2903ff019
|
1143 1144 |
int ret; struct fd f; |
1da177e4c
|
1145 1146 1147 1148 1149 |
struct sock *sock; struct inode *inode; struct sigevent notification; struct mqueue_inode_info *info; struct sk_buff *nc; |
20114f71b
|
1150 |
if (u_notification) { |
1da177e4c
|
1151 1152 1153 |
if (copy_from_user(¬ification, u_notification, sizeof(struct sigevent))) return -EFAULT; |
20114f71b
|
1154 1155 1156 |
} audit_mq_notify(mqdes, u_notification ? ¬ification : NULL); |
1da177e4c
|
1157 |
|
20114f71b
|
1158 1159 1160 |
nc = NULL; sock = NULL; if (u_notification != NULL) { |
1da177e4c
|
1161 1162 1163 1164 1165 |
if (unlikely(notification.sigev_notify != SIGEV_NONE && notification.sigev_notify != SIGEV_SIGNAL && notification.sigev_notify != SIGEV_THREAD)) return -EINVAL; if (notification.sigev_notify == SIGEV_SIGNAL && |
7ed20e1ad
|
1166 |
!valid_signal(notification.sigev_signo)) { |
1da177e4c
|
1167 1168 1169 |
return -EINVAL; } if (notification.sigev_notify == SIGEV_THREAD) { |
c3d8d1e30
|
1170 |
long timeo; |
1da177e4c
|
1171 1172 |
/* create the notify skb */ nc = alloc_skb(NOTIFY_COOKIE_LEN, GFP_KERNEL); |
8d8ffefaa
|
1173 1174 |
if (!nc) { ret = -ENOMEM; |
1da177e4c
|
1175 |
goto out; |
8d8ffefaa
|
1176 |
} |
1da177e4c
|
1177 1178 1179 |
if (copy_from_user(nc->data, notification.sigev_value.sival_ptr, NOTIFY_COOKIE_LEN)) { |
8d8ffefaa
|
1180 |
ret = -EFAULT; |
1da177e4c
|
1181 1182 1183 1184 1185 1186 1187 |
goto out; } /* TODO: add a header? */ skb_put(nc, NOTIFY_COOKIE_LEN); /* and attach it to the socket */ retry: |
2903ff019
|
1188 1189 |
f = fdget(notification.sigev_signo); if (!f.file) { |
8d8ffefaa
|
1190 |
ret = -EBADF; |
1da177e4c
|
1191 |
goto out; |
8d8ffefaa
|
1192 |
} |
2903ff019
|
1193 1194 |
sock = netlink_getsockbyfilp(f.file); fdput(f); |
1da177e4c
|
1195 1196 1197 1198 1199 |
if (IS_ERR(sock)) { ret = PTR_ERR(sock); sock = NULL; goto out; } |
c3d8d1e30
|
1200 |
timeo = MAX_SCHEDULE_TIMEOUT; |
9457afee8
|
1201 |
ret = netlink_attachskb(sock, nc, &timeo, NULL); |
1da177e4c
|
1202 |
if (ret == 1) |
8d8ffefaa
|
1203 |
goto retry; |
1da177e4c
|
1204 1205 1206 1207 1208 1209 1210 |
if (ret) { sock = NULL; nc = NULL; goto out; } } } |
2903ff019
|
1211 1212 |
f = fdget(mqdes); if (!f.file) { |
8d8ffefaa
|
1213 |
ret = -EBADF; |
1da177e4c
|
1214 |
goto out; |
8d8ffefaa
|
1215 |
} |
1da177e4c
|
1216 |
|
496ad9aa8
|
1217 |
inode = file_inode(f.file); |
2903ff019
|
1218 |
if (unlikely(f.file->f_op != &mqueue_file_operations)) { |
8d8ffefaa
|
1219 |
ret = -EBADF; |
1da177e4c
|
1220 |
goto out_fput; |
8d8ffefaa
|
1221 |
} |
1da177e4c
|
1222 1223 1224 1225 1226 |
info = MQUEUE_I(inode); ret = 0; spin_lock(&info->lock); if (u_notification == NULL) { |
a03fcb730
|
1227 |
if (info->notify_owner == task_tgid(current)) { |
1da177e4c
|
1228 1229 1230 |
remove_notification(info); inode->i_atime = inode->i_ctime = CURRENT_TIME; } |
a03fcb730
|
1231 |
} else if (info->notify_owner != NULL) { |
1da177e4c
|
1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 |
ret = -EBUSY; } else { switch (notification.sigev_notify) { case SIGEV_NONE: info->notify.sigev_notify = SIGEV_NONE; break; case SIGEV_THREAD: info->notify_sock = sock; info->notify_cookie = nc; sock = NULL; nc = NULL; info->notify.sigev_notify = SIGEV_THREAD; break; case SIGEV_SIGNAL: info->notify.sigev_signo = notification.sigev_signo; info->notify.sigev_value = notification.sigev_value; info->notify.sigev_notify = SIGEV_SIGNAL; break; } |
a03fcb730
|
1251 1252 |
info->notify_owner = get_pid(task_tgid(current)); |
6f9ac6d93
|
1253 |
info->notify_user_ns = get_user_ns(current_user_ns()); |
1da177e4c
|
1254 1255 1256 1257 |
inode->i_atime = inode->i_ctime = CURRENT_TIME; } spin_unlock(&info->lock); out_fput: |
2903ff019
|
1258 |
fdput(f); |
1da177e4c
|
1259 |
out: |
3ab08fe20
|
1260 |
if (sock) |
1da177e4c
|
1261 |
netlink_detachskb(sock, nc); |
3ab08fe20
|
1262 |
else if (nc) |
1da177e4c
|
1263 |
dev_kfree_skb(nc); |
3ab08fe20
|
1264 |
|
1da177e4c
|
1265 1266 |
return ret; } |
c4ea37c26
|
1267 1268 1269 |
SYSCALL_DEFINE3(mq_getsetattr, mqd_t, mqdes, const struct mq_attr __user *, u_mqstat, struct mq_attr __user *, u_omqstat) |
1da177e4c
|
1270 1271 1272 |
{ int ret; struct mq_attr mqstat, omqstat; |
2903ff019
|
1273 |
struct fd f; |
1da177e4c
|
1274 1275 1276 1277 1278 1279 1280 1281 1282 |
struct inode *inode; struct mqueue_inode_info *info; if (u_mqstat != NULL) { if (copy_from_user(&mqstat, u_mqstat, sizeof(struct mq_attr))) return -EFAULT; if (mqstat.mq_flags & (~O_NONBLOCK)) return -EINVAL; } |
2903ff019
|
1283 1284 |
f = fdget(mqdes); if (!f.file) { |
8d8ffefaa
|
1285 |
ret = -EBADF; |
1da177e4c
|
1286 |
goto out; |
8d8ffefaa
|
1287 |
} |
1da177e4c
|
1288 |
|
496ad9aa8
|
1289 |
inode = file_inode(f.file); |
2903ff019
|
1290 |
if (unlikely(f.file->f_op != &mqueue_file_operations)) { |
8d8ffefaa
|
1291 |
ret = -EBADF; |
1da177e4c
|
1292 |
goto out_fput; |
8d8ffefaa
|
1293 |
} |
1da177e4c
|
1294 1295 1296 1297 1298 |
info = MQUEUE_I(inode); spin_lock(&info->lock); omqstat = info->attr; |
2903ff019
|
1299 |
omqstat.mq_flags = f.file->f_flags & O_NONBLOCK; |
1da177e4c
|
1300 |
if (u_mqstat) { |
7392906ea
|
1301 |
audit_mq_getsetattr(mqdes, &mqstat); |
2903ff019
|
1302 |
spin_lock(&f.file->f_lock); |
1da177e4c
|
1303 |
if (mqstat.mq_flags & O_NONBLOCK) |
2903ff019
|
1304 |
f.file->f_flags |= O_NONBLOCK; |
1da177e4c
|
1305 |
else |
2903ff019
|
1306 1307 |
f.file->f_flags &= ~O_NONBLOCK; spin_unlock(&f.file->f_lock); |
1da177e4c
|
1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 |
inode->i_atime = inode->i_ctime = CURRENT_TIME; } spin_unlock(&info->lock); ret = 0; if (u_omqstat != NULL && copy_to_user(u_omqstat, &omqstat, sizeof(struct mq_attr))) ret = -EFAULT; out_fput: |
2903ff019
|
1320 |
fdput(f); |
1da177e4c
|
1321 1322 1323 |
out: return ret; } |
92e1d5be9
|
1324 |
static const struct inode_operations mqueue_dir_inode_operations = { |
1da177e4c
|
1325 1326 1327 1328 |
.lookup = simple_lookup, .create = mqueue_create, .unlink = mqueue_unlink, }; |
9a32144e9
|
1329 |
static const struct file_operations mqueue_file_operations = { |
1da177e4c
|
1330 1331 1332 |
.flush = mqueue_flush_file, .poll = mqueue_poll_file, .read = mqueue_read_file, |
6038f373a
|
1333 |
.llseek = default_llseek, |
1da177e4c
|
1334 |
}; |
b87221de6
|
1335 |
static const struct super_operations mqueue_super_ops = { |
1da177e4c
|
1336 1337 |
.alloc_inode = mqueue_alloc_inode, .destroy_inode = mqueue_destroy_inode, |
6d8af64c1
|
1338 |
.evict_inode = mqueue_evict_inode, |
1da177e4c
|
1339 |
.statfs = simple_statfs, |
1da177e4c
|
1340 1341 1342 1343 |
}; static struct file_system_type mqueue_fs_type = { .name = "mqueue", |
ceefda693
|
1344 |
.mount = mqueue_mount, |
1da177e4c
|
1345 |
.kill_sb = kill_litter_super, |
bc1b69ed2
|
1346 |
.fs_flags = FS_USERNS_MOUNT, |
1da177e4c
|
1347 |
}; |
7eafd7c74
|
1348 1349 1350 1351 1352 1353 |
int mq_init_ns(struct ipc_namespace *ns) { ns->mq_queues_count = 0; ns->mq_queues_max = DFLT_QUEUESMAX; ns->mq_msg_max = DFLT_MSGMAX; ns->mq_msgsize_max = DFLT_MSGSIZEMAX; |
cef0184c1
|
1354 1355 |
ns->mq_msg_default = DFLT_MSG; ns->mq_msgsize_default = DFLT_MSGSIZE; |
7eafd7c74
|
1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 |
ns->mq_mnt = kern_mount_data(&mqueue_fs_type, ns); if (IS_ERR(ns->mq_mnt)) { int err = PTR_ERR(ns->mq_mnt); ns->mq_mnt = NULL; return err; } return 0; } void mq_clear_sbinfo(struct ipc_namespace *ns) { ns->mq_mnt->mnt_sb->s_fs_info = NULL; } void mq_put_mnt(struct ipc_namespace *ns) { |
6f686574c
|
1373 |
kern_unmount(ns->mq_mnt); |
7eafd7c74
|
1374 |
} |
1da177e4c
|
1375 1376 1377 1378 1379 1380 |
static int __init init_mqueue_fs(void) { int error; mqueue_inode_cachep = kmem_cache_create("mqueue_inode_cache", sizeof(struct mqueue_inode_info), 0, |
20c2df83d
|
1381 |
SLAB_HWCACHE_ALIGN, init_once); |
1da177e4c
|
1382 1383 |
if (mqueue_inode_cachep == NULL) return -ENOMEM; |
2329e392a
|
1384 |
/* ignore failures - they are not fatal */ |
bdc8e5f85
|
1385 |
mq_sysctl_table = mq_register_sysctl_table(); |
1da177e4c
|
1386 1387 1388 1389 |
error = register_filesystem(&mqueue_fs_type); if (error) goto out_sysctl; |
7eafd7c74
|
1390 |
spin_lock_init(&mq_lock); |
6f686574c
|
1391 1392 |
error = mq_init_ns(&init_ipc_ns); if (error) |
1da177e4c
|
1393 |
goto out_filesystem; |
1da177e4c
|
1394 |
|
1da177e4c
|
1395 1396 1397 1398 1399 1400 1401 |
return 0; out_filesystem: unregister_filesystem(&mqueue_fs_type); out_sysctl: if (mq_sysctl_table) unregister_sysctl_table(mq_sysctl_table); |
1a1d92c10
|
1402 |
kmem_cache_destroy(mqueue_inode_cachep); |
1da177e4c
|
1403 1404 |
return error; } |
6d08a2567
|
1405 |
device_initcall(init_mqueue_fs); |