Blame view
fs/select.c
34.8 KB
b24413180 License cleanup: ... |
1 |
// SPDX-License-Identifier: GPL-2.0 |
1da177e4c Linux-2.6.12-rc2 |
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 |
/* * This file contains the procedures for the handling of select and poll * * Created for Linux based loosely upon Mathius Lattner's minix * patches by Peter MacDonald. Heavily edited by Linus. * * 4 February 1994 * COFF/ELF binary emulation. If the process has the STICKY_TIMEOUTS * flag set in its personality we do *not* modify the given timeout * parameter to reflect time remaining. * * 24 January 2000 * Changed sys_poll()/do_poll() to use PAGE_SIZE chunk-based allocation * of fds to overcome nfds < 16390 descriptors limit (Tigran Aivazian). */ |
022a16924 ROUND_UP macro cl... |
17 |
#include <linux/kernel.h> |
3f07c0144 sched/headers: Pr... |
18 19 |
#include <linux/sched/signal.h> #include <linux/sched/rt.h> |
1da177e4c Linux-2.6.12-rc2 |
20 |
#include <linux/syscalls.h> |
630d9c472 fs: reduce the us... |
21 |
#include <linux/export.h> |
1da177e4c Linux-2.6.12-rc2 |
22 |
#include <linux/slab.h> |
1da177e4c Linux-2.6.12-rc2 |
23 24 25 |
#include <linux/poll.h> #include <linux/personality.h> /* for STICKY_TIMEOUTS */ #include <linux/file.h> |
9f3acc314 [PATCH] split lin... |
26 |
#include <linux/fdtable.h> |
1da177e4c Linux-2.6.12-rc2 |
27 |
#include <linux/fs.h> |
b835996f6 [PATCH] files: lo... |
28 |
#include <linux/rcupdate.h> |
8ff3e8e85 select: switch se... |
29 |
#include <linux/hrtimer.h> |
9745cdb36 select: use freez... |
30 |
#include <linux/freezer.h> |
076bb0c82 net: rename inclu... |
31 |
#include <net/busy_poll.h> |
2d19309cf fs/select: add vm... |
32 |
#include <linux/vmalloc.h> |
1da177e4c Linux-2.6.12-rc2 |
33 |
|
7c0f6ba68 Replace <asm/uacc... |
34 |
#include <linux/uaccess.h> |
1da177e4c Linux-2.6.12-rc2 |
35 |
|
90d6e24a3 hrtimer: make sel... |
36 37 38 39 40 41 42 43 44 45 46 47 |
/* * Estimate expected accuracy in ns from a timeval. * * After quite a bit of churning around, we've settled on * a simple thing of taking 0.1% of the timeout as the * slack, with a cap of 100 msec. * "nice" tasks get a 0.5% slack instead. * * Consider this comment an open invitation to come up with even * better solutions.. */ |
5ae87e79e poll/select: avoi... |
48 |
#define MAX_SLACK (100 * NSEC_PER_MSEC) |
766b9f928 fs: poll/select/r... |
49 |
static long __estimate_accuracy(struct timespec64 *tv) |
90d6e24a3 hrtimer: make sel... |
50 |
{ |
96d2ab484 hrtimer: fix sign... |
51 |
long slack; |
90d6e24a3 hrtimer: make sel... |
52 |
int divfactor = 1000; |
5ae87e79e poll/select: avoi... |
53 54 |
if (tv->tv_sec < 0) return 0; |
4ce105d30 hrtimer: incorpor... |
55 |
if (task_nice(current) > 0) |
90d6e24a3 hrtimer: make sel... |
56 |
divfactor = divfactor / 5; |
5ae87e79e poll/select: avoi... |
57 58 |
if (tv->tv_sec > MAX_SLACK / (NSEC_PER_SEC/divfactor)) return MAX_SLACK; |
90d6e24a3 hrtimer: make sel... |
59 60 |
slack = tv->tv_nsec / divfactor; slack += tv->tv_sec * (NSEC_PER_SEC/divfactor); |
5ae87e79e poll/select: avoi... |
61 62 |
if (slack > MAX_SLACK) return MAX_SLACK; |
96d2ab484 hrtimer: fix sign... |
63 |
|
90d6e24a3 hrtimer: make sel... |
64 65 |
return slack; } |
766b9f928 fs: poll/select/r... |
66 |
u64 select_estimate_accuracy(struct timespec64 *tv) |
90d6e24a3 hrtimer: make sel... |
67 |
{ |
da8b44d5a timer: convert ti... |
68 |
u64 ret; |
766b9f928 fs: poll/select/r... |
69 |
struct timespec64 now; |
90d6e24a3 hrtimer: make sel... |
70 71 72 73 |
/* * Realtime tasks get a slack of 0 for obvious reasons. */ |
4ce105d30 hrtimer: incorpor... |
74 |
if (rt_task(current)) |
90d6e24a3 hrtimer: make sel... |
75 |
return 0; |
766b9f928 fs: poll/select/r... |
76 77 |
ktime_get_ts64(&now); now = timespec64_sub(*tv, now); |
90d6e24a3 hrtimer: make sel... |
78 79 80 81 82 |
ret = __estimate_accuracy(&now); if (ret < current->timer_slack_ns) return current->timer_slack_ns; return ret; } |
1da177e4c Linux-2.6.12-rc2 |
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 |
struct poll_table_page { struct poll_table_page * next; struct poll_table_entry * entry; struct poll_table_entry entries[0]; }; #define POLL_TABLE_FULL(table) \ ((unsigned long)((table)->entry+1) > PAGE_SIZE + (unsigned long)(table)) /* * Ok, Peter made a complicated, but straightforward multiple_wait() function. * I have rewritten this, taking some shortcuts: This code may not be easy to * follow, but it should be free of race-conditions, and it's practical. If you * understand what I'm doing here, then you understand how the linux * sleep/wakeup mechanism works. * * Two very simple procedures, poll_wait() and poll_freewait() make all the * work. poll_wait() is an inline-function defined in <linux/poll.h>, * as all select/poll functions have to call it to add an entry to the * poll table. */ |
75c96f858 [PATCH] make some... |
104 105 |
static void __pollwait(struct file *filp, wait_queue_head_t *wait_address, poll_table *p); |
1da177e4c Linux-2.6.12-rc2 |
106 107 108 109 |
void poll_initwait(struct poll_wqueues *pwq) { init_poll_funcptr(&pwq->pt, __pollwait); |
5f820f648 poll: allow f_op-... |
110 |
pwq->polling_task = current; |
b2add73db poll/select: init... |
111 |
pwq->triggered = 0; |
1da177e4c Linux-2.6.12-rc2 |
112 113 |
pwq->error = 0; pwq->table = NULL; |
70674f95c [PATCH] Optimize ... |
114 |
pwq->inline_index = 0; |
1da177e4c Linux-2.6.12-rc2 |
115 |
} |
1da177e4c Linux-2.6.12-rc2 |
116 |
EXPORT_SYMBOL(poll_initwait); |
70674f95c [PATCH] Optimize ... |
117 118 |
static void free_poll_entry(struct poll_table_entry *entry) { |
ccf6780dc Style fix in fs/s... |
119 |
remove_wait_queue(entry->wait_address, &entry->wait); |
70674f95c [PATCH] Optimize ... |
120 121 |
fput(entry->filp); } |
1da177e4c Linux-2.6.12-rc2 |
122 123 124 |
void poll_freewait(struct poll_wqueues *pwq) { struct poll_table_page * p = pwq->table; |
70674f95c [PATCH] Optimize ... |
125 126 127 |
int i; for (i = 0; i < pwq->inline_index; i++) free_poll_entry(pwq->inline_entries + i); |
1da177e4c Linux-2.6.12-rc2 |
128 129 130 131 132 133 134 |
while (p) { struct poll_table_entry * entry; struct poll_table_page *old; entry = p->entry; do { entry--; |
70674f95c [PATCH] Optimize ... |
135 |
free_poll_entry(entry); |
1da177e4c Linux-2.6.12-rc2 |
136 137 138 139 140 141 |
} while (entry > p->entries); old = p; p = p->next; free_page((unsigned long) old); } } |
1da177e4c Linux-2.6.12-rc2 |
142 |
EXPORT_SYMBOL(poll_freewait); |
5f820f648 poll: allow f_op-... |
143 |
static struct poll_table_entry *poll_get_entry(struct poll_wqueues *p) |
1da177e4c Linux-2.6.12-rc2 |
144 |
{ |
1da177e4c Linux-2.6.12-rc2 |
145 |
struct poll_table_page *table = p->table; |
70674f95c [PATCH] Optimize ... |
146 147 |
if (p->inline_index < N_INLINE_POLL_ENTRIES) return p->inline_entries + p->inline_index++; |
1da177e4c Linux-2.6.12-rc2 |
148 149 150 151 152 153 |
if (!table || POLL_TABLE_FULL(table)) { struct poll_table_page *new_table; new_table = (struct poll_table_page *) __get_free_page(GFP_KERNEL); if (!new_table) { p->error = -ENOMEM; |
70674f95c [PATCH] Optimize ... |
154 |
return NULL; |
1da177e4c Linux-2.6.12-rc2 |
155 156 157 158 159 160 |
} new_table->entry = new_table->entries; new_table->next = table; p->table = new_table; table = new_table; } |
70674f95c [PATCH] Optimize ... |
161 162 |
return table->entry++; } |
ac6424b98 sched/wait: Renam... |
163 |
static int __pollwake(wait_queue_entry_t *wait, unsigned mode, int sync, void *key) |
5f820f648 poll: allow f_op-... |
164 165 166 167 168 169 170 171 172 |
{ struct poll_wqueues *pwq = wait->private; DECLARE_WAITQUEUE(dummy_wait, pwq->polling_task); /* * Although this function is called under waitqueue lock, LOCK * doesn't imply write barrier and the users expect write * barrier semantics on wakeup functions. The following * smp_wmb() is equivalent to smp_wmb() in try_to_wake_up() |
b92b8b35a locking/arch: Ren... |
173 |
* and is paired with smp_store_mb() in poll_schedule_timeout. |
5f820f648 poll: allow f_op-... |
174 175 176 177 178 179 180 181 182 183 184 185 186 187 |
*/ smp_wmb(); pwq->triggered = 1; /* * Perform the default wake up operation using a dummy * waitqueue. * * TODO: This is hacky but there currently is no interface to * pass in @sync. @sync is scheduled to be removed and once * that happens, wake_up_process() can be used directly. */ return default_wake_function(&dummy_wait, mode, sync, key); } |
ac6424b98 sched/wait: Renam... |
188 |
static int pollwake(wait_queue_entry_t *wait, unsigned mode, int sync, void *key) |
4938d7e02 poll: avoid extra... |
189 190 191 192 193 194 195 196 |
{ struct poll_table_entry *entry; entry = container_of(wait, struct poll_table_entry, wait); if (key && !((unsigned long)key & entry->key)) return 0; return __pollwake(wait, mode, sync, key); } |
70674f95c [PATCH] Optimize ... |
197 198 199 200 |
/* Add a new entry */ static void __pollwait(struct file *filp, wait_queue_head_t *wait_address, poll_table *p) { |
5f820f648 poll: allow f_op-... |
201 202 |
struct poll_wqueues *pwq = container_of(p, struct poll_wqueues, pt); struct poll_table_entry *entry = poll_get_entry(pwq); |
70674f95c [PATCH] Optimize ... |
203 204 |
if (!entry) return; |
cb0942b81 make get_file() r... |
205 |
entry->filp = get_file(filp); |
70674f95c [PATCH] Optimize ... |
206 |
entry->wait_address = wait_address; |
626cf2366 poll: add poll_re... |
207 |
entry->key = p->_key; |
5f820f648 poll: allow f_op-... |
208 209 |
init_waitqueue_func_entry(&entry->wait, pollwake); entry->wait.private = pwq; |
ccf6780dc Style fix in fs/s... |
210 |
add_wait_queue(wait_address, &entry->wait); |
1da177e4c Linux-2.6.12-rc2 |
211 |
} |
5f820f648 poll: allow f_op-... |
212 213 214 215 216 217 218 |
int poll_schedule_timeout(struct poll_wqueues *pwq, int state, ktime_t *expires, unsigned long slack) { int rc = -EINTR; set_current_state(state); if (!pwq->triggered) |
59612d187 Revert "select: u... |
219 |
rc = schedule_hrtimeout_range(expires, slack, HRTIMER_MODE_ABS); |
5f820f648 poll: allow f_op-... |
220 221 222 223 224 |
__set_current_state(TASK_RUNNING); /* * Prepare for the next iteration. * |
b92b8b35a locking/arch: Ren... |
225 |
* The following smp_store_mb() serves two purposes. First, it's |
5f820f648 poll: allow f_op-... |
226 227 228 229 230 231 232 |
* the counterpart rmb of the wmb in pollwake() such that data * written before wake up is always visible after wake up. * Second, the full barrier guarantees that triggered clearing * doesn't pass event check of the next iteration. Note that * this problem doesn't exist for the first iteration as * add_wait_queue() has full barrier semantics. */ |
b92b8b35a locking/arch: Ren... |
233 |
smp_store_mb(pwq->triggered, 0); |
5f820f648 poll: allow f_op-... |
234 235 236 237 |
return rc; } EXPORT_SYMBOL(poll_schedule_timeout); |
b773ad40a select: add poll_... |
238 239 |
/** * poll_select_set_timeout - helper function to setup the timeout value |
766b9f928 fs: poll/select/r... |
240 |
* @to: pointer to timespec64 variable for the final timeout |
b773ad40a select: add poll_... |
241 242 243 244 245 246 247 248 |
* @sec: seconds (from user space) * @nsec: nanoseconds (from user space) * * Note, we do not use a timespec for the user space value here, That * way we can use the function for timeval and compat interfaces as well. * * Returns -EINVAL if sec/nsec are not normalized. Otherwise 0. */ |
766b9f928 fs: poll/select/r... |
249 |
int poll_select_set_timeout(struct timespec64 *to, time64_t sec, long nsec) |
b773ad40a select: add poll_... |
250 |
{ |
766b9f928 fs: poll/select/r... |
251 |
struct timespec64 ts = {.tv_sec = sec, .tv_nsec = nsec}; |
b773ad40a select: add poll_... |
252 |
|
766b9f928 fs: poll/select/r... |
253 |
if (!timespec64_valid(&ts)) |
b773ad40a select: add poll_... |
254 255 256 257 258 259 |
return -EINVAL; /* Optimize for the zero timeout value here */ if (!sec && !nsec) { to->tv_sec = to->tv_nsec = 0; } else { |
766b9f928 fs: poll/select/r... |
260 261 |
ktime_get_ts64(to); *to = timespec64_add_safe(*to, ts); |
b773ad40a select: add poll_... |
262 263 264 |
} return 0; } |
766b9f928 fs: poll/select/r... |
265 266 |
static int poll_select_copy_remaining(struct timespec64 *end_time, void __user *p, |
b773ad40a select: add poll_... |
267 268 |
int timeval, int ret) { |
766b9f928 fs: poll/select/r... |
269 |
struct timespec64 rts64; |
b773ad40a select: add poll_... |
270 271 272 273 274 275 276 277 278 279 280 281 |
struct timespec rts; struct timeval rtv; if (!p) return ret; if (current->personality & STICKY_TIMEOUTS) goto sticky; /* No update for zero timeout */ if (!end_time->tv_sec && !end_time->tv_nsec) return ret; |
766b9f928 fs: poll/select/r... |
282 283 284 285 286 287 |
ktime_get_ts64(&rts64); rts64 = timespec64_sub(*end_time, rts64); if (rts64.tv_sec < 0) rts64.tv_sec = rts64.tv_nsec = 0; rts = timespec64_to_timespec(rts64); |
b773ad40a select: add poll_... |
288 289 |
if (timeval) { |
65329bf46 fs/select.c: fix ... |
290 291 |
if (sizeof(rtv) > sizeof(rtv.tv_sec) + sizeof(rtv.tv_usec)) memset(&rtv, 0, sizeof(rtv)); |
766b9f928 fs: poll/select/r... |
292 293 |
rtv.tv_sec = rts64.tv_sec; rtv.tv_usec = rts64.tv_nsec / NSEC_PER_USEC; |
b773ad40a select: add poll_... |
294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 |
if (!copy_to_user(p, &rtv, sizeof(rtv))) return ret; } else if (!copy_to_user(p, &rts, sizeof(rts))) return ret; /* * If an application puts its timeval in read-only memory, we * don't want the Linux-specific update to the timeval to * cause a fault after the select has completed * successfully. However, because we're not updating the * timeval, we can't restart the system call. */ sticky: if (ret == -ERESTARTNOHAND) ret = -EINTR; return ret; } |
e99ca56ce move compat selec... |
314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 |
/* * Scalable version of the fd_set. */ typedef struct { unsigned long *in, *out, *ex; unsigned long *res_in, *res_out, *res_ex; } fd_set_bits; /* * How many longwords for "nr" bits? */ #define FDS_BITPERLONG (8*sizeof(long)) #define FDS_LONGS(nr) (((nr)+FDS_BITPERLONG-1)/FDS_BITPERLONG) #define FDS_BYTES(nr) (FDS_LONGS(nr)*sizeof(long)) /* * We do a VERIFY_WRITE here even though we are only reading this time: * we'll write to it eventually.. * * Use "unsigned long" accesses to let user-mode fd_set's be long-aligned. */ static inline int get_fd_set(unsigned long nr, void __user *ufdset, unsigned long *fdset) { nr = FDS_BYTES(nr); if (ufdset) return copy_from_user(fdset, ufdset, nr) ? -EFAULT : 0; memset(fdset, 0, nr); return 0; } static inline unsigned long __must_check set_fd_set(unsigned long nr, void __user *ufdset, unsigned long *fdset) { if (ufdset) return __copy_to_user(ufdset, fdset, FDS_BYTES(nr)); return 0; } static inline void zero_fd_set(unsigned long nr, unsigned long *fdset) { memset(fdset, 0, FDS_BYTES(nr)); } |
1da177e4c Linux-2.6.12-rc2 |
360 361 362 363 364 365 366 367 368 369 370 |
#define FDS_IN(fds, n) (fds->in + n) #define FDS_OUT(fds, n) (fds->out + n) #define FDS_EX(fds, n) (fds->ex + n) #define BITS(fds, n) (*FDS_IN(fds, n)|*FDS_OUT(fds, n)|*FDS_EX(fds, n)) static int max_select_fd(unsigned long n, fd_set_bits *fds) { unsigned long *open_fds; unsigned long set; int max; |
badf16621 [PATCH] files: br... |
371 |
struct fdtable *fdt; |
1da177e4c Linux-2.6.12-rc2 |
372 373 |
/* handle last in-complete long-word first */ |
8ded2bbc1 posix_types.h: Cl... |
374 375 |
set = ~(~0UL << (n & (BITS_PER_LONG-1))); n /= BITS_PER_LONG; |
badf16621 [PATCH] files: br... |
376 |
fdt = files_fdtable(current->files); |
1fd36adcd Replace the fd_se... |
377 |
open_fds = fdt->open_fds + n; |
1da177e4c Linux-2.6.12-rc2 |
378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 |
max = 0; if (set) { set &= BITS(fds, n); if (set) { if (!(set & ~*open_fds)) goto get_max; return -EBADF; } } while (n) { open_fds--; n--; set = BITS(fds, n); if (!set) continue; if (set & ~*open_fds) return -EBADF; if (max) continue; get_max: do { max++; set >>= 1; } while (set); |
8ded2bbc1 posix_types.h: Cl... |
402 |
max += n * BITS_PER_LONG; |
1da177e4c Linux-2.6.12-rc2 |
403 404 405 406 |
} return max; } |
1da177e4c Linux-2.6.12-rc2 |
407 408 409 |
#define POLLIN_SET (POLLRDNORM | POLLRDBAND | POLLIN | POLLHUP | POLLERR) #define POLLOUT_SET (POLLWRBAND | POLLWRNORM | POLLOUT | POLLERR) #define POLLEX_SET (POLLPRI) |
4938d7e02 poll: avoid extra... |
410 |
static inline void wait_key_set(poll_table *wait, unsigned long in, |
2d48d67fa net: poll/select ... |
411 412 |
unsigned long out, unsigned long bit, unsigned int ll_flag) |
4938d7e02 poll: avoid extra... |
413 |
{ |
2d48d67fa net: poll/select ... |
414 |
wait->_key = POLLEX_SET | ll_flag; |
626cf2366 poll: add poll_re... |
415 416 417 418 |
if (in & bit) wait->_key |= POLLIN_SET; if (out & bit) wait->_key |= POLLOUT_SET; |
4938d7e02 poll: avoid extra... |
419 |
} |
e99ca56ce move compat selec... |
420 |
static int do_select(int n, fd_set_bits *fds, struct timespec64 *end_time) |
1da177e4c Linux-2.6.12-rc2 |
421 |
{ |
8ff3e8e85 select: switch se... |
422 |
ktime_t expire, *to = NULL; |
1da177e4c Linux-2.6.12-rc2 |
423 424 |
struct poll_wqueues table; poll_table *wait; |
8ff3e8e85 select: switch se... |
425 |
int retval, i, timed_out = 0; |
da8b44d5a timer: convert ti... |
426 |
u64 slack = 0; |
cbf55001b net: rename low l... |
427 |
unsigned int busy_flag = net_busy_loop_on() ? POLL_BUSY_LOOP : 0; |
37056719b net: Track start ... |
428 |
unsigned long busy_start = 0; |
1da177e4c Linux-2.6.12-rc2 |
429 |
|
b835996f6 [PATCH] files: lo... |
430 |
rcu_read_lock(); |
1da177e4c Linux-2.6.12-rc2 |
431 |
retval = max_select_fd(n, fds); |
b835996f6 [PATCH] files: lo... |
432 |
rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
433 434 435 436 437 438 439 |
if (retval < 0) return retval; n = retval; poll_initwait(&table); wait = &table.pt; |
8ff3e8e85 select: switch se... |
440 |
if (end_time && !end_time->tv_sec && !end_time->tv_nsec) { |
626cf2366 poll: add poll_re... |
441 |
wait->_qproc = NULL; |
8ff3e8e85 select: switch se... |
442 443 |
timed_out = 1; } |
96d2ab484 hrtimer: fix sign... |
444 |
if (end_time && !timed_out) |
231f3d393 select: rename es... |
445 |
slack = select_estimate_accuracy(end_time); |
90d6e24a3 hrtimer: make sel... |
446 |
|
1da177e4c Linux-2.6.12-rc2 |
447 448 449 |
retval = 0; for (;;) { unsigned long *rinp, *routp, *rexp, *inp, *outp, *exp; |
cbf55001b net: rename low l... |
450 |
bool can_busy_loop = false; |
1da177e4c Linux-2.6.12-rc2 |
451 |
|
1da177e4c Linux-2.6.12-rc2 |
452 453 454 455 456 457 |
inp = fds->in; outp = fds->out; exp = fds->ex; rinp = fds->res_in; routp = fds->res_out; rexp = fds->res_ex; for (i = 0; i < n; ++rinp, ++routp, ++rexp) { unsigned long in, out, ex, all_bits, bit = 1, mask, j; unsigned long res_in = 0, res_out = 0, res_ex = 0; |
1da177e4c Linux-2.6.12-rc2 |
458 459 460 461 |
in = *inp++; out = *outp++; ex = *exp++; all_bits = in | out | ex; if (all_bits == 0) { |
8ded2bbc1 posix_types.h: Cl... |
462 |
i += BITS_PER_LONG; |
1da177e4c Linux-2.6.12-rc2 |
463 464 |
continue; } |
8ded2bbc1 posix_types.h: Cl... |
465 |
for (j = 0; j < BITS_PER_LONG; ++j, ++i, bit <<= 1) { |
2903ff019 switch simple cas... |
466 |
struct fd f; |
1da177e4c Linux-2.6.12-rc2 |
467 468 469 470 |
if (i >= n) break; if (!(bit & all_bits)) continue; |
2903ff019 switch simple cas... |
471 472 473 474 |
f = fdget(i); if (f.file) { const struct file_operations *f_op; f_op = f.file->f_op; |
1da177e4c Linux-2.6.12-rc2 |
475 |
mask = DEFAULT_POLLMASK; |
72c2d5319 file->f_op is nev... |
476 |
if (f_op->poll) { |
2d48d67fa net: poll/select ... |
477 |
wait_key_set(wait, in, out, |
cbf55001b net: rename low l... |
478 |
bit, busy_flag); |
2903ff019 switch simple cas... |
479 |
mask = (*f_op->poll)(f.file, wait); |
4938d7e02 poll: avoid extra... |
480 |
} |
2903ff019 switch simple cas... |
481 |
fdput(f); |
1da177e4c Linux-2.6.12-rc2 |
482 483 484 |
if ((mask & POLLIN_SET) && (in & bit)) { res_in |= bit; retval++; |
626cf2366 poll: add poll_re... |
485 |
wait->_qproc = NULL; |
1da177e4c Linux-2.6.12-rc2 |
486 487 488 489 |
} if ((mask & POLLOUT_SET) && (out & bit)) { res_out |= bit; retval++; |
626cf2366 poll: add poll_re... |
490 |
wait->_qproc = NULL; |
1da177e4c Linux-2.6.12-rc2 |
491 492 493 494 |
} if ((mask & POLLEX_SET) && (ex & bit)) { res_ex |= bit; retval++; |
626cf2366 poll: add poll_re... |
495 |
wait->_qproc = NULL; |
1da177e4c Linux-2.6.12-rc2 |
496 |
} |
2d48d67fa net: poll/select ... |
497 |
/* got something, stop busy polling */ |
cbf55001b net: rename low l... |
498 499 500 501 502 503 504 505 506 507 |
if (retval) { can_busy_loop = false; busy_flag = 0; /* * only remember a returned * POLL_BUSY_LOOP if we asked for it */ } else if (busy_flag & mask) can_busy_loop = true; |
1da177e4c Linux-2.6.12-rc2 |
508 |
} |
1da177e4c Linux-2.6.12-rc2 |
509 510 511 512 513 514 515 |
} if (res_in) *rinp = res_in; if (res_out) *routp = res_out; if (res_ex) *rexp = res_ex; |
55d853849 Fix performance r... |
516 |
cond_resched(); |
1da177e4c Linux-2.6.12-rc2 |
517 |
} |
626cf2366 poll: add poll_re... |
518 |
wait->_qproc = NULL; |
8ff3e8e85 select: switch se... |
519 |
if (retval || timed_out || signal_pending(current)) |
1da177e4c Linux-2.6.12-rc2 |
520 |
break; |
f5264481c trivial: small cl... |
521 |
if (table.error) { |
1da177e4c Linux-2.6.12-rc2 |
522 523 524 |
retval = table.error; break; } |
9f72949f6 [PATCH] Add psele... |
525 |
|
cbf55001b net: rename low l... |
526 |
/* only if found POLL_BUSY_LOOP sockets && not out of time */ |
76b1e9b98 net/fs: change bu... |
527 |
if (can_busy_loop && !need_resched()) { |
37056719b net: Track start ... |
528 529 |
if (!busy_start) { busy_start = busy_loop_current_time(); |
76b1e9b98 net/fs: change bu... |
530 531 |
continue; } |
37056719b net: Track start ... |
532 |
if (!busy_loop_timeout(busy_start)) |
76b1e9b98 net/fs: change bu... |
533 534 535 |
continue; } busy_flag = 0; |
2d48d67fa net: poll/select ... |
536 |
|
8ff3e8e85 select: switch se... |
537 538 539 540 541 542 |
/* * If this is the first loop and we have a timeout * given, then we convert to ktime_t and set the to * pointer to the expiry value. */ if (end_time && !to) { |
766b9f928 fs: poll/select/r... |
543 |
expire = timespec64_to_ktime(*end_time); |
8ff3e8e85 select: switch se... |
544 |
to = &expire; |
9f72949f6 [PATCH] Add psele... |
545 |
} |
8ff3e8e85 select: switch se... |
546 |
|
5f820f648 poll: allow f_op-... |
547 548 |
if (!poll_schedule_timeout(&table, TASK_INTERRUPTIBLE, to, slack)) |
8ff3e8e85 select: switch se... |
549 |
timed_out = 1; |
1da177e4c Linux-2.6.12-rc2 |
550 |
} |
1da177e4c Linux-2.6.12-rc2 |
551 552 |
poll_freewait(&table); |
1da177e4c Linux-2.6.12-rc2 |
553 554 |
return retval; } |
1da177e4c Linux-2.6.12-rc2 |
555 556 557 558 559 560 561 562 |
/* * We can actually return ERESTARTSYS instead of EINTR, but I'd * like to be certain this leads to no problems. So I return * EINTR just for safety. * * Update: ERESTARTSYS breaks at least the xview clock binary, so * I'm trying ERESTARTNOHAND which restart only when you want to. */ |
a2dcb44c3 [PATCH] make osf_... |
563 |
int core_sys_select(int n, fd_set __user *inp, fd_set __user *outp, |
766b9f928 fs: poll/select/r... |
564 |
fd_set __user *exp, struct timespec64 *end_time) |
1da177e4c Linux-2.6.12-rc2 |
565 566 |
{ fd_set_bits fds; |
29ff2db55 [PATCH] select() ... |
567 |
void *bits; |
bbea9f696 [PATCH] fdtable: ... |
568 |
int ret, max_fds; |
2d19309cf fs/select: add vm... |
569 |
size_t size, alloc_size; |
badf16621 [PATCH] files: br... |
570 |
struct fdtable *fdt; |
70674f95c [PATCH] Optimize ... |
571 |
/* Allocate small arguments on the stack to save memory and be faster */ |
30c14e40e [PATCH] avoid una... |
572 |
long stack_fds[SELECT_STACK_ALLOC/sizeof(long)]; |
1da177e4c Linux-2.6.12-rc2 |
573 |
|
1da177e4c Linux-2.6.12-rc2 |
574 575 576 |
ret = -EINVAL; if (n < 0) goto out_nofds; |
bbea9f696 [PATCH] fdtable: ... |
577 |
/* max_fds can increase, so grab it once to avoid race */ |
b835996f6 [PATCH] files: lo... |
578 |
rcu_read_lock(); |
badf16621 [PATCH] files: br... |
579 |
fdt = files_fdtable(current->files); |
bbea9f696 [PATCH] fdtable: ... |
580 |
max_fds = fdt->max_fds; |
b835996f6 [PATCH] files: lo... |
581 |
rcu_read_unlock(); |
bbea9f696 [PATCH] fdtable: ... |
582 583 |
if (n > max_fds) n = max_fds; |
1da177e4c Linux-2.6.12-rc2 |
584 585 586 587 588 589 |
/* * We need 6 bitmaps (in/out/ex for both incoming and outgoing), * since we used fdset we need to allocate memory in units of * long-words. */ |
1da177e4c Linux-2.6.12-rc2 |
590 |
size = FDS_BYTES(n); |
b04eb6aa0 [PATCH] select: d... |
591 592 593 594 |
bits = stack_fds; if (size > sizeof(stack_fds) / 6) { /* Not enough space in on-stack array; must use kmalloc */ ret = -ENOMEM; |
2d19309cf fs/select: add vm... |
595 596 597 598 |
if (size > (SIZE_MAX / 6)) goto out_nofds; alloc_size = 6 * size; |
752ade68c treewide: use kv[... |
599 |
bits = kvmalloc(alloc_size, GFP_KERNEL); |
b04eb6aa0 [PATCH] select: d... |
600 601 602 |
if (!bits) goto out_nofds; } |
29ff2db55 [PATCH] select() ... |
603 604 605 606 607 608 |
fds.in = bits; fds.out = bits + size; fds.ex = bits + 2*size; fds.res_in = bits + 3*size; fds.res_out = bits + 4*size; fds.res_ex = bits + 5*size; |
1da177e4c Linux-2.6.12-rc2 |
609 610 611 612 613 614 615 616 |
if ((ret = get_fd_set(n, inp, fds.in)) || (ret = get_fd_set(n, outp, fds.out)) || (ret = get_fd_set(n, exp, fds.ex))) goto out; zero_fd_set(n, fds.res_in); zero_fd_set(n, fds.res_out); zero_fd_set(n, fds.res_ex); |
8ff3e8e85 select: switch se... |
617 |
ret = do_select(n, &fds, end_time); |
1da177e4c Linux-2.6.12-rc2 |
618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 |
if (ret < 0) goto out; if (!ret) { ret = -ERESTARTNOHAND; if (signal_pending(current)) goto out; ret = 0; } if (set_fd_set(n, inp, fds.res_in) || set_fd_set(n, outp, fds.res_out) || set_fd_set(n, exp, fds.res_ex)) ret = -EFAULT; out: |
70674f95c [PATCH] Optimize ... |
634 |
if (bits != stack_fds) |
2d19309cf fs/select: add vm... |
635 |
kvfree(bits); |
1da177e4c Linux-2.6.12-rc2 |
636 637 638 |
out_nofds: return ret; } |
5a8a82b1d [CVE-2009-0029] S... |
639 640 |
SYSCALL_DEFINE5(select, int, n, fd_set __user *, inp, fd_set __user *, outp, fd_set __user *, exp, struct timeval __user *, tvp) |
9f72949f6 [PATCH] Add psele... |
641 |
{ |
766b9f928 fs: poll/select/r... |
642 |
struct timespec64 end_time, *to = NULL; |
9f72949f6 [PATCH] Add psele... |
643 644 645 646 647 648 |
struct timeval tv; int ret; if (tvp) { if (copy_from_user(&tv, tvp, sizeof(tv))) return -EFAULT; |
8ff3e8e85 select: switch se... |
649 |
to = &end_time; |
4d36a9e65 select: deal with... |
650 651 652 |
if (poll_select_set_timeout(to, tv.tv_sec + (tv.tv_usec / USEC_PER_SEC), (tv.tv_usec % USEC_PER_SEC) * NSEC_PER_USEC)) |
9f72949f6 [PATCH] Add psele... |
653 |
return -EINVAL; |
9f72949f6 [PATCH] Add psele... |
654 |
} |
8ff3e8e85 select: switch se... |
655 656 |
ret = core_sys_select(n, inp, outp, exp, to); ret = poll_select_copy_remaining(&end_time, tvp, 1, ret); |
9f72949f6 [PATCH] Add psele... |
657 658 659 |
return ret; } |
c9da9f212 [CVE-2009-0029] M... |
660 661 662 |
static long do_pselect(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, struct timespec __user *tsp, const sigset_t __user *sigmask, size_t sigsetsize) |
9f72949f6 [PATCH] Add psele... |
663 |
{ |
9f72949f6 [PATCH] Add psele... |
664 |
sigset_t ksigmask, sigsaved; |
766b9f928 fs: poll/select/r... |
665 666 |
struct timespec ts; struct timespec64 ts64, end_time, *to = NULL; |
9f72949f6 [PATCH] Add psele... |
667 668 669 670 671 |
int ret; if (tsp) { if (copy_from_user(&ts, tsp, sizeof(ts))) return -EFAULT; |
766b9f928 fs: poll/select/r... |
672 |
ts64 = timespec_to_timespec64(ts); |
9f72949f6 [PATCH] Add psele... |
673 |
|
8ff3e8e85 select: switch se... |
674 |
to = &end_time; |
766b9f928 fs: poll/select/r... |
675 |
if (poll_select_set_timeout(to, ts64.tv_sec, ts64.tv_nsec)) |
9f72949f6 [PATCH] Add psele... |
676 |
return -EINVAL; |
9f72949f6 [PATCH] Add psele... |
677 678 679 680 681 682 683 684 685 686 687 688 |
} if (sigmask) { /* XXX: Don't preclude handling different sized sigset_t's. */ if (sigsetsize != sizeof(sigset_t)) return -EINVAL; if (copy_from_user(&ksigmask, sigmask, sizeof(ksigmask))) return -EFAULT; sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP)); sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); } |
62568510b Fix timeouts in s... |
689 |
ret = core_sys_select(n, inp, outp, exp, to); |
8ff3e8e85 select: switch se... |
690 |
ret = poll_select_copy_remaining(&end_time, tsp, 0, ret); |
9f72949f6 [PATCH] Add psele... |
691 692 693 694 695 696 697 698 699 700 |
if (ret == -ERESTARTNOHAND) { /* * Don't restore the signal mask yet. Let do_signal() deliver * the signal on the way back to userspace, before the signal * mask is restored. */ if (sigmask) { memcpy(¤t->saved_sigmask, &sigsaved, sizeof(sigsaved)); |
4e4c22c71 signals: add set_... |
701 |
set_restore_sigmask(); |
9f72949f6 [PATCH] Add psele... |
702 703 704 705 706 707 708 709 710 711 712 713 714 |
} } else if (sigmask) sigprocmask(SIG_SETMASK, &sigsaved, NULL); return ret; } /* * Most architectures can't handle 7-argument syscalls. So we provide a * 6-argument version where the sixth argument is a pointer to a structure * which has a pointer to the sigset_t itself followed by a size_t containing * the sigset size. */ |
d4e82042c [CVE-2009-0029] S... |
715 716 717 |
SYSCALL_DEFINE6(pselect6, int, n, fd_set __user *, inp, fd_set __user *, outp, fd_set __user *, exp, struct timespec __user *, tsp, void __user *, sig) |
9f72949f6 [PATCH] Add psele... |
718 719 720 721 722 723 |
{ size_t sigsetsize = 0; sigset_t __user *up = NULL; if (sig) { if (!access_ok(VERIFY_READ, sig, sizeof(void *)+sizeof(size_t)) |
e110ab94e [PATCH] fix __use... |
724 |
|| __get_user(up, (sigset_t __user * __user *)sig) |
9f72949f6 [PATCH] Add psele... |
725 |
|| __get_user(sigsetsize, |
e110ab94e [PATCH] fix __use... |
726 |
(size_t __user *)(sig+sizeof(void *)))) |
9f72949f6 [PATCH] Add psele... |
727 728 |
return -EFAULT; } |
c9da9f212 [CVE-2009-0029] M... |
729 |
return do_pselect(n, inp, outp, exp, tsp, up, sigsetsize); |
9f72949f6 [PATCH] Add psele... |
730 |
} |
9f72949f6 [PATCH] Add psele... |
731 |
|
5d0e52830 Add generic sys_o... |
732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 |
#ifdef __ARCH_WANT_SYS_OLD_SELECT struct sel_arg_struct { unsigned long n; fd_set __user *inp, *outp, *exp; struct timeval __user *tvp; }; SYSCALL_DEFINE1(old_select, struct sel_arg_struct __user *, arg) { struct sel_arg_struct a; if (copy_from_user(&a, arg, sizeof(a))) return -EFAULT; return sys_select(a.n, a.inp, a.outp, a.exp, a.tvp); } #endif |
1da177e4c Linux-2.6.12-rc2 |
748 749 750 751 752 753 754 |
struct poll_list { struct poll_list *next; int len; struct pollfd entries[0]; }; #define POLLFD_PER_PAGE ((PAGE_SIZE-sizeof(struct poll_list)) / sizeof(struct pollfd)) |
4a4b69f79 [PATCH] Poll clea... |
755 756 757 758 759 |
/* * Fish for pollable events on the pollfd->fd file descriptor. We're only * interested in events matching the pollfd->events mask, and the result * matching that mask is both recorded in pollfd->revents and returned. The * pwait poll_table will be used by the fd-provided poll handler for waiting, |
626cf2366 poll: add poll_re... |
760 |
* if pwait->_qproc is non-NULL. |
4a4b69f79 [PATCH] Poll clea... |
761 |
*/ |
2d48d67fa net: poll/select ... |
762 |
static inline unsigned int do_pollfd(struct pollfd *pollfd, poll_table *pwait, |
cbf55001b net: rename low l... |
763 764 |
bool *can_busy_poll, unsigned int busy_flag) |
1da177e4c Linux-2.6.12-rc2 |
765 |
{ |
4a4b69f79 [PATCH] Poll clea... |
766 767 768 769 770 771 |
unsigned int mask; int fd; mask = 0; fd = pollfd->fd; if (fd >= 0) { |
2903ff019 switch simple cas... |
772 |
struct fd f = fdget(fd); |
4a4b69f79 [PATCH] Poll clea... |
773 |
mask = POLLNVAL; |
2903ff019 switch simple cas... |
774 |
if (f.file) { |
4a4b69f79 [PATCH] Poll clea... |
775 |
mask = DEFAULT_POLLMASK; |
72c2d5319 file->f_op is nev... |
776 |
if (f.file->f_op->poll) { |
626cf2366 poll: add poll_re... |
777 |
pwait->_key = pollfd->events|POLLERR|POLLHUP; |
cbf55001b net: rename low l... |
778 |
pwait->_key |= busy_flag; |
2903ff019 switch simple cas... |
779 |
mask = f.file->f_op->poll(f.file, pwait); |
cbf55001b net: rename low l... |
780 781 |
if (mask & busy_flag) *can_busy_poll = true; |
4938d7e02 poll: avoid extra... |
782 |
} |
4a4b69f79 [PATCH] Poll clea... |
783 784 |
/* Mask out unneeded events. */ mask &= pollfd->events | POLLERR | POLLHUP; |
2903ff019 switch simple cas... |
785 |
fdput(f); |
1da177e4c Linux-2.6.12-rc2 |
786 |
} |
1da177e4c Linux-2.6.12-rc2 |
787 |
} |
4a4b69f79 [PATCH] Poll clea... |
788 789 790 |
pollfd->revents = mask; return mask; |
1da177e4c Linux-2.6.12-rc2 |
791 |
} |
ccec5ee30 poll: plug an unu... |
792 |
static int do_poll(struct poll_list *list, struct poll_wqueues *wait, |
766b9f928 fs: poll/select/r... |
793 |
struct timespec64 *end_time) |
1da177e4c Linux-2.6.12-rc2 |
794 |
{ |
1da177e4c Linux-2.6.12-rc2 |
795 |
poll_table* pt = &wait->pt; |
8ff3e8e85 select: switch se... |
796 797 |
ktime_t expire, *to = NULL; int timed_out = 0, count = 0; |
da8b44d5a timer: convert ti... |
798 |
u64 slack = 0; |
cbf55001b net: rename low l... |
799 |
unsigned int busy_flag = net_busy_loop_on() ? POLL_BUSY_LOOP : 0; |
37056719b net: Track start ... |
800 |
unsigned long busy_start = 0; |
1da177e4c Linux-2.6.12-rc2 |
801 |
|
9f72949f6 [PATCH] Add psele... |
802 |
/* Optimise the no-wait case */ |
8ff3e8e85 select: switch se... |
803 |
if (end_time && !end_time->tv_sec && !end_time->tv_nsec) { |
626cf2366 poll: add poll_re... |
804 |
pt->_qproc = NULL; |
8ff3e8e85 select: switch se... |
805 806 |
timed_out = 1; } |
9bf084f70 do_poll: return -... |
807 |
|
96d2ab484 hrtimer: fix sign... |
808 |
if (end_time && !timed_out) |
231f3d393 select: rename es... |
809 |
slack = select_estimate_accuracy(end_time); |
90d6e24a3 hrtimer: make sel... |
810 |
|
1da177e4c Linux-2.6.12-rc2 |
811 812 |
for (;;) { struct poll_list *walk; |
cbf55001b net: rename low l... |
813 |
bool can_busy_loop = false; |
9f72949f6 [PATCH] Add psele... |
814 |
|
4a4b69f79 [PATCH] Poll clea... |
815 816 817 818 819 820 821 822 |
for (walk = list; walk != NULL; walk = walk->next) { struct pollfd * pfd, * pfd_end; pfd = walk->entries; pfd_end = pfd + walk->len; for (; pfd != pfd_end; pfd++) { /* * Fish for events. If we found one, record it |
626cf2366 poll: add poll_re... |
823 |
* and kill poll_table->_qproc, so we don't |
4a4b69f79 [PATCH] Poll clea... |
824 825 826 827 |
* needlessly register any other waiters after * this. They'll get immediately deregistered * when we break out and return. */ |
cbf55001b net: rename low l... |
828 829 |
if (do_pollfd(pfd, pt, &can_busy_loop, busy_flag)) { |
4a4b69f79 [PATCH] Poll clea... |
830 |
count++; |
626cf2366 poll: add poll_re... |
831 |
pt->_qproc = NULL; |
cbf55001b net: rename low l... |
832 833 834 |
/* found something, stop busy polling */ busy_flag = 0; can_busy_loop = false; |
4a4b69f79 [PATCH] Poll clea... |
835 836 |
} } |
1da177e4c Linux-2.6.12-rc2 |
837 |
} |
4a4b69f79 [PATCH] Poll clea... |
838 839 |
/* * All waiters have already been registered, so don't provide |
626cf2366 poll: add poll_re... |
840 |
* a poll_table->_qproc to them on the next loop iteration. |
4a4b69f79 [PATCH] Poll clea... |
841 |
*/ |
626cf2366 poll: add poll_re... |
842 |
pt->_qproc = NULL; |
9bf084f70 do_poll: return -... |
843 844 845 846 847 |
if (!count) { count = wait->error; if (signal_pending(current)) count = -EINTR; } |
8ff3e8e85 select: switch se... |
848 |
if (count || timed_out) |
1da177e4c Linux-2.6.12-rc2 |
849 |
break; |
9f72949f6 [PATCH] Add psele... |
850 |
|
cbf55001b net: rename low l... |
851 |
/* only if found POLL_BUSY_LOOP sockets && not out of time */ |
76b1e9b98 net/fs: change bu... |
852 |
if (can_busy_loop && !need_resched()) { |
37056719b net: Track start ... |
853 854 |
if (!busy_start) { busy_start = busy_loop_current_time(); |
76b1e9b98 net/fs: change bu... |
855 856 |
continue; } |
37056719b net: Track start ... |
857 |
if (!busy_loop_timeout(busy_start)) |
76b1e9b98 net/fs: change bu... |
858 859 860 |
continue; } busy_flag = 0; |
91e2fd337 net: avoid callin... |
861 |
|
8ff3e8e85 select: switch se... |
862 863 864 865 866 867 |
/* * If this is the first loop and we have a timeout * given, then we convert to ktime_t and set the to * pointer to the expiry value. */ if (end_time && !to) { |
766b9f928 fs: poll/select/r... |
868 |
expire = timespec64_to_ktime(*end_time); |
8ff3e8e85 select: switch se... |
869 |
to = &expire; |
9f72949f6 [PATCH] Add psele... |
870 |
} |
5f820f648 poll: allow f_op-... |
871 |
if (!poll_schedule_timeout(wait, TASK_INTERRUPTIBLE, to, slack)) |
8ff3e8e85 select: switch se... |
872 |
timed_out = 1; |
1da177e4c Linux-2.6.12-rc2 |
873 |
} |
1da177e4c Linux-2.6.12-rc2 |
874 875 |
return count; } |
70674f95c [PATCH] Optimize ... |
876 877 |
#define N_STACK_PPS ((sizeof(stack_pps) - sizeof(struct poll_list)) / \ sizeof(struct pollfd)) |
e99ca56ce move compat selec... |
878 |
static int do_sys_poll(struct pollfd __user *ufds, unsigned int nfds, |
766b9f928 fs: poll/select/r... |
879 |
struct timespec64 *end_time) |
1da177e4c Linux-2.6.12-rc2 |
880 881 |
{ struct poll_wqueues table; |
252e5725c do_sys_poll: simp... |
882 |
int err = -EFAULT, fdcount, len, size; |
30c14e40e [PATCH] avoid una... |
883 884 885 886 |
/* Allocate small arguments on the stack to save memory and be faster - use long to make sure the buffer is aligned properly on 64 bit archs to avoid unaligned access */ long stack_pps[POLL_STACK_ALLOC/sizeof(long)]; |
252e5725c do_sys_poll: simp... |
887 888 889 |
struct poll_list *const head = (struct poll_list *)stack_pps; struct poll_list *walk = head; unsigned long todo = nfds; |
1da177e4c Linux-2.6.12-rc2 |
890 |
|
d554ed895 fs: use rlimit he... |
891 |
if (nfds > rlimit(RLIMIT_NOFILE)) |
1da177e4c Linux-2.6.12-rc2 |
892 |
return -EINVAL; |
252e5725c do_sys_poll: simp... |
893 894 895 896 897 898 |
len = min_t(unsigned int, nfds, N_STACK_PPS); for (;;) { walk->next = NULL; walk->len = len; if (!len) break; |
1da177e4c Linux-2.6.12-rc2 |
899 |
|
252e5725c do_sys_poll: simp... |
900 901 902 903 904 905 906 |
if (copy_from_user(walk->entries, ufds + nfds-todo, sizeof(struct pollfd) * walk->len)) goto out_fds; todo -= walk->len; if (!todo) break; |
1da177e4c Linux-2.6.12-rc2 |
907 |
|
252e5725c do_sys_poll: simp... |
908 909 910 911 912 |
len = min(todo, POLLFD_PER_PAGE); size = sizeof(struct poll_list) + sizeof(struct pollfd) * len; walk = walk->next = kmalloc(size, GFP_KERNEL); if (!walk) { err = -ENOMEM; |
1da177e4c Linux-2.6.12-rc2 |
913 914 |
goto out_fds; } |
1da177e4c Linux-2.6.12-rc2 |
915 |
} |
9f72949f6 [PATCH] Add psele... |
916 |
|
252e5725c do_sys_poll: simp... |
917 |
poll_initwait(&table); |
ccec5ee30 poll: plug an unu... |
918 |
fdcount = do_poll(head, &table, end_time); |
252e5725c do_sys_poll: simp... |
919 |
poll_freewait(&table); |
1da177e4c Linux-2.6.12-rc2 |
920 |
|
252e5725c do_sys_poll: simp... |
921 |
for (walk = head; walk; walk = walk->next) { |
1da177e4c Linux-2.6.12-rc2 |
922 923 |
struct pollfd *fds = walk->entries; int j; |
252e5725c do_sys_poll: simp... |
924 925 |
for (j = 0; j < walk->len; j++, ufds++) if (__put_user(fds[j].revents, &ufds->revents)) |
1da177e4c Linux-2.6.12-rc2 |
926 |
goto out_fds; |
1da177e4c Linux-2.6.12-rc2 |
927 |
} |
252e5725c do_sys_poll: simp... |
928 |
|
1da177e4c Linux-2.6.12-rc2 |
929 |
err = fdcount; |
1da177e4c Linux-2.6.12-rc2 |
930 |
out_fds: |
252e5725c do_sys_poll: simp... |
931 932 933 934 935 |
walk = head->next; while (walk) { struct poll_list *pos = walk; walk = walk->next; kfree(pos); |
1da177e4c Linux-2.6.12-rc2 |
936 |
} |
252e5725c do_sys_poll: simp... |
937 |
|
1da177e4c Linux-2.6.12-rc2 |
938 939 |
return err; } |
9f72949f6 [PATCH] Add psele... |
940 |
|
3075d9da0 Use ERESTART_REST... |
941 942 |
static long do_restart_poll(struct restart_block *restart_block) { |
8ff3e8e85 select: switch se... |
943 944 |
struct pollfd __user *ufds = restart_block->poll.ufds; int nfds = restart_block->poll.nfds; |
766b9f928 fs: poll/select/r... |
945 |
struct timespec64 *to = NULL, end_time; |
3075d9da0 Use ERESTART_REST... |
946 |
int ret; |
8ff3e8e85 select: switch se... |
947 948 949 950 951 952 953 |
if (restart_block->poll.has_timeout) { end_time.tv_sec = restart_block->poll.tv_sec; end_time.tv_nsec = restart_block->poll.tv_nsec; to = &end_time; } ret = do_sys_poll(ufds, nfds, to); |
3075d9da0 Use ERESTART_REST... |
954 955 |
if (ret == -EINTR) { restart_block->fn = do_restart_poll; |
3075d9da0 Use ERESTART_REST... |
956 957 958 959 |
ret = -ERESTART_RESTARTBLOCK; } return ret; } |
5a8a82b1d [CVE-2009-0029] S... |
960 |
SYSCALL_DEFINE3(poll, struct pollfd __user *, ufds, unsigned int, nfds, |
faf309009 sys_poll: fix inc... |
961 |
int, timeout_msecs) |
9f72949f6 [PATCH] Add psele... |
962 |
{ |
766b9f928 fs: poll/select/r... |
963 |
struct timespec64 end_time, *to = NULL; |
3075d9da0 Use ERESTART_REST... |
964 |
int ret; |
9f72949f6 [PATCH] Add psele... |
965 |
|
8ff3e8e85 select: switch se... |
966 967 968 969 |
if (timeout_msecs >= 0) { to = &end_time; poll_select_set_timeout(to, timeout_msecs / MSEC_PER_SEC, NSEC_PER_MSEC * (timeout_msecs % MSEC_PER_SEC)); |
9f72949f6 [PATCH] Add psele... |
970 |
} |
8ff3e8e85 select: switch se... |
971 |
ret = do_sys_poll(ufds, nfds, to); |
3075d9da0 Use ERESTART_REST... |
972 973 |
if (ret == -EINTR) { struct restart_block *restart_block; |
8ff3e8e85 select: switch se... |
974 |
|
f56141e3e all arches, signa... |
975 |
restart_block = ¤t->restart_block; |
3075d9da0 Use ERESTART_REST... |
976 |
restart_block->fn = do_restart_poll; |
8ff3e8e85 select: switch se... |
977 978 979 980 981 982 983 984 985 |
restart_block->poll.ufds = ufds; restart_block->poll.nfds = nfds; if (timeout_msecs >= 0) { restart_block->poll.tv_sec = end_time.tv_sec; restart_block->poll.tv_nsec = end_time.tv_nsec; restart_block->poll.has_timeout = 1; } else restart_block->poll.has_timeout = 0; |
3075d9da0 Use ERESTART_REST... |
986 987 988 |
ret = -ERESTART_RESTARTBLOCK; } return ret; |
9f72949f6 [PATCH] Add psele... |
989 |
} |
d4e82042c [CVE-2009-0029] S... |
990 991 992 |
SYSCALL_DEFINE5(ppoll, struct pollfd __user *, ufds, unsigned int, nfds, struct timespec __user *, tsp, const sigset_t __user *, sigmask, size_t, sigsetsize) |
9f72949f6 [PATCH] Add psele... |
993 994 |
{ sigset_t ksigmask, sigsaved; |
766b9f928 fs: poll/select/r... |
995 996 |
struct timespec ts; struct timespec64 end_time, *to = NULL; |
9f72949f6 [PATCH] Add psele... |
997 998 999 1000 1001 |
int ret; if (tsp) { if (copy_from_user(&ts, tsp, sizeof(ts))) return -EFAULT; |
8ff3e8e85 select: switch se... |
1002 1003 1004 |
to = &end_time; if (poll_select_set_timeout(to, ts.tv_sec, ts.tv_nsec)) return -EINVAL; |
9f72949f6 [PATCH] Add psele... |
1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 |
} if (sigmask) { /* XXX: Don't preclude handling different sized sigset_t's. */ if (sigsetsize != sizeof(sigset_t)) return -EINVAL; if (copy_from_user(&ksigmask, sigmask, sizeof(ksigmask))) return -EFAULT; sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP)); sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); } |
8ff3e8e85 select: switch se... |
1017 |
ret = do_sys_poll(ufds, nfds, to); |
9f72949f6 [PATCH] Add psele... |
1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 |
/* We can restart this syscall, usually */ if (ret == -EINTR) { /* * Don't restore the signal mask yet. Let do_signal() deliver * the signal on the way back to userspace, before the signal * mask is restored. */ if (sigmask) { memcpy(¤t->saved_sigmask, &sigsaved, sizeof(sigsaved)); |
4e4c22c71 signals: add set_... |
1029 |
set_restore_sigmask(); |
9f72949f6 [PATCH] Add psele... |
1030 1031 1032 1033 |
} ret = -ERESTARTNOHAND; } else if (sigmask) sigprocmask(SIG_SETMASK, &sigsaved, NULL); |
8ff3e8e85 select: switch se... |
1034 |
ret = poll_select_copy_remaining(&end_time, tsp, 0, ret); |
9f72949f6 [PATCH] Add psele... |
1035 1036 1037 |
return ret; } |
e99ca56ce move compat selec... |
1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 |
#ifdef CONFIG_COMPAT #define __COMPAT_NFDBITS (8 * sizeof(compat_ulong_t)) static int compat_poll_select_copy_remaining(struct timespec *end_time, void __user *p, int timeval, int ret) { struct timespec ts; if (!p) return ret; if (current->personality & STICKY_TIMEOUTS) goto sticky; /* No update for zero timeout */ if (!end_time->tv_sec && !end_time->tv_nsec) return ret; ktime_get_ts(&ts); ts = timespec_sub(*end_time, ts); if (ts.tv_sec < 0) ts.tv_sec = ts.tv_nsec = 0; if (timeval) { struct compat_timeval rtv; rtv.tv_sec = ts.tv_sec; rtv.tv_usec = ts.tv_nsec / NSEC_PER_USEC; if (!copy_to_user(p, &rtv, sizeof(rtv))) return ret; } else { struct compat_timespec rts; rts.tv_sec = ts.tv_sec; rts.tv_nsec = ts.tv_nsec; if (!copy_to_user(p, &rts, sizeof(rts))) return ret; } /* * If an application puts its timeval in read-only memory, we * don't want the Linux-specific update to the timeval to * cause a fault after the select has completed * successfully. However, because we're not updating the * timeval, we can't restart the system call. */ sticky: if (ret == -ERESTARTNOHAND) ret = -EINTR; return ret; } /* * Ooo, nasty. We need here to frob 32-bit unsigned longs to * 64-bit unsigned longs. */ static int compat_get_fd_set(unsigned long nr, compat_ulong_t __user *ufdset, unsigned long *fdset) { |
e99ca56ce move compat selec... |
1102 |
if (ufdset) { |
464d62421 select: switch co... |
1103 |
return compat_get_bitmap(fdset, ufdset, nr); |
e99ca56ce move compat selec... |
1104 |
} else { |
79de3cbe9 fs/select: Fix me... |
1105 |
zero_fd_set(nr, fdset); |
464d62421 select: switch co... |
1106 |
return 0; |
e99ca56ce move compat selec... |
1107 |
} |
e99ca56ce move compat selec... |
1108 1109 1110 1111 1112 1113 |
} static int compat_set_fd_set(unsigned long nr, compat_ulong_t __user *ufdset, unsigned long *fdset) { |
e99ca56ce move compat selec... |
1114 1115 |
if (!ufdset) return 0; |
464d62421 select: switch co... |
1116 |
return compat_put_bitmap(ufdset, fdset, nr); |
e99ca56ce move compat selec... |
1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 |
} /* * This is a virtual copy of sys_select from fs/select.c and probably * should be compared to it from time to time */ /* * We can actually return ERESTARTSYS instead of EINTR, but I'd * like to be certain this leads to no problems. So I return * EINTR just for safety. * * Update: ERESTARTSYS breaks at least the xview clock binary, so * I'm trying ERESTARTNOHAND which restart only when you want to. */ static int compat_core_sys_select(int n, compat_ulong_t __user *inp, compat_ulong_t __user *outp, compat_ulong_t __user *exp, struct timespec *end_time) { fd_set_bits fds; void *bits; int size, max_fds, ret = -EINVAL; struct fdtable *fdt; long stack_fds[SELECT_STACK_ALLOC/sizeof(long)]; if (n < 0) goto out_nofds; /* max_fds can increase, so grab it once to avoid race */ rcu_read_lock(); fdt = files_fdtable(current->files); max_fds = fdt->max_fds; rcu_read_unlock(); if (n > max_fds) n = max_fds; /* * We need 6 bitmaps (in/out/ex for both incoming and outgoing), * since we used fdset we need to allocate memory in units of * long-words. */ size = FDS_BYTES(n); bits = stack_fds; if (size > sizeof(stack_fds) / 6) { bits = kmalloc(6 * size, GFP_KERNEL); ret = -ENOMEM; if (!bits) goto out_nofds; } fds.in = (unsigned long *) bits; fds.out = (unsigned long *) (bits + size); fds.ex = (unsigned long *) (bits + 2*size); fds.res_in = (unsigned long *) (bits + 3*size); fds.res_out = (unsigned long *) (bits + 4*size); fds.res_ex = (unsigned long *) (bits + 5*size); if ((ret = compat_get_fd_set(n, inp, fds.in)) || (ret = compat_get_fd_set(n, outp, fds.out)) || (ret = compat_get_fd_set(n, exp, fds.ex))) goto out; zero_fd_set(n, fds.res_in); zero_fd_set(n, fds.res_out); zero_fd_set(n, fds.res_ex); ret = do_select(n, &fds, end_time); if (ret < 0) goto out; if (!ret) { ret = -ERESTARTNOHAND; if (signal_pending(current)) goto out; ret = 0; } if (compat_set_fd_set(n, inp, fds.res_in) || compat_set_fd_set(n, outp, fds.res_out) || compat_set_fd_set(n, exp, fds.res_ex)) ret = -EFAULT; out: if (bits != stack_fds) kfree(bits); out_nofds: return ret; } COMPAT_SYSCALL_DEFINE5(select, int, n, compat_ulong_t __user *, inp, compat_ulong_t __user *, outp, compat_ulong_t __user *, exp, struct compat_timeval __user *, tvp) { struct timespec end_time, *to = NULL; struct compat_timeval tv; int ret; if (tvp) { if (copy_from_user(&tv, tvp, sizeof(tv))) return -EFAULT; to = &end_time; if (poll_select_set_timeout(to, tv.tv_sec + (tv.tv_usec / USEC_PER_SEC), (tv.tv_usec % USEC_PER_SEC) * NSEC_PER_USEC)) return -EINVAL; } ret = compat_core_sys_select(n, inp, outp, exp, to); ret = compat_poll_select_copy_remaining(&end_time, tvp, 1, ret); return ret; } struct compat_sel_arg_struct { compat_ulong_t n; compat_uptr_t inp; compat_uptr_t outp; compat_uptr_t exp; compat_uptr_t tvp; }; COMPAT_SYSCALL_DEFINE1(old_select, struct compat_sel_arg_struct __user *, arg) { struct compat_sel_arg_struct a; if (copy_from_user(&a, arg, sizeof(a))) return -EFAULT; return compat_sys_select(a.n, compat_ptr(a.inp), compat_ptr(a.outp), compat_ptr(a.exp), compat_ptr(a.tvp)); } static long do_compat_pselect(int n, compat_ulong_t __user *inp, compat_ulong_t __user *outp, compat_ulong_t __user *exp, struct compat_timespec __user *tsp, compat_sigset_t __user *sigmask, compat_size_t sigsetsize) { compat_sigset_t ss32; sigset_t ksigmask, sigsaved; struct compat_timespec ts; struct timespec end_time, *to = NULL; int ret; if (tsp) { if (copy_from_user(&ts, tsp, sizeof(ts))) return -EFAULT; to = &end_time; if (poll_select_set_timeout(to, ts.tv_sec, ts.tv_nsec)) return -EINVAL; } if (sigmask) { if (sigsetsize != sizeof(compat_sigset_t)) return -EINVAL; if (copy_from_user(&ss32, sigmask, sizeof(ss32))) return -EFAULT; sigset_from_compat(&ksigmask, &ss32); sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP)); sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); } ret = compat_core_sys_select(n, inp, outp, exp, to); ret = compat_poll_select_copy_remaining(&end_time, tsp, 0, ret); if (ret == -ERESTARTNOHAND) { /* * Don't restore the signal mask yet. Let do_signal() deliver * the signal on the way back to userspace, before the signal * mask is restored. */ if (sigmask) { memcpy(¤t->saved_sigmask, &sigsaved, sizeof(sigsaved)); set_restore_sigmask(); } } else if (sigmask) sigprocmask(SIG_SETMASK, &sigsaved, NULL); return ret; } COMPAT_SYSCALL_DEFINE6(pselect6, int, n, compat_ulong_t __user *, inp, compat_ulong_t __user *, outp, compat_ulong_t __user *, exp, struct compat_timespec __user *, tsp, void __user *, sig) { compat_size_t sigsetsize = 0; compat_uptr_t up = 0; if (sig) { if (!access_ok(VERIFY_READ, sig, sizeof(compat_uptr_t)+sizeof(compat_size_t)) || __get_user(up, (compat_uptr_t __user *)sig) || __get_user(sigsetsize, (compat_size_t __user *)(sig+sizeof(up)))) return -EFAULT; } return do_compat_pselect(n, inp, outp, exp, tsp, compat_ptr(up), sigsetsize); } COMPAT_SYSCALL_DEFINE5(ppoll, struct pollfd __user *, ufds, unsigned int, nfds, struct compat_timespec __user *, tsp, const compat_sigset_t __user *, sigmask, compat_size_t, sigsetsize) { compat_sigset_t ss32; sigset_t ksigmask, sigsaved; struct compat_timespec ts; struct timespec end_time, *to = NULL; int ret; if (tsp) { if (copy_from_user(&ts, tsp, sizeof(ts))) return -EFAULT; to = &end_time; if (poll_select_set_timeout(to, ts.tv_sec, ts.tv_nsec)) return -EINVAL; } if (sigmask) { if (sigsetsize != sizeof(compat_sigset_t)) return -EINVAL; if (copy_from_user(&ss32, sigmask, sizeof(ss32))) return -EFAULT; sigset_from_compat(&ksigmask, &ss32); sigdelsetmask(&ksigmask, sigmask(SIGKILL)|sigmask(SIGSTOP)); sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved); } ret = do_sys_poll(ufds, nfds, to); /* We can restart this syscall, usually */ if (ret == -EINTR) { /* * Don't restore the signal mask yet. Let do_signal() deliver * the signal on the way back to userspace, before the signal * mask is restored. */ if (sigmask) { memcpy(¤t->saved_sigmask, &sigsaved, sizeof(sigsaved)); set_restore_sigmask(); } ret = -ERESTARTNOHAND; } else if (sigmask) sigprocmask(SIG_SETMASK, &sigsaved, NULL); ret = compat_poll_select_copy_remaining(&end_time, tsp, 0, ret); return ret; } #endif |