Blame view
fs/select.c
20.5 KB
1da177e4c Linux-2.6.12-rc2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 |
/* * 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). */ #include <linux/syscalls.h> #include <linux/module.h> #include <linux/slab.h> #include <linux/smp_lock.h> #include <linux/poll.h> #include <linux/personality.h> /* for STICKY_TIMEOUTS */ #include <linux/file.h> #include <linux/fs.h> |
b835996f6 [PATCH] files: lo... |
25 |
#include <linux/rcupdate.h> |
1da177e4c Linux-2.6.12-rc2 |
26 27 28 29 30 |
#include <asm/uaccess.h> #define ROUND_UP(x,y) (((x)+(y)-1)/(y)) #define DEFAULT_POLLMASK (POLLIN | POLLOUT | POLLRDNORM | POLLWRNORM) |
1da177e4c Linux-2.6.12-rc2 |
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 |
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... |
52 53 |
static void __pollwait(struct file *filp, wait_queue_head_t *wait_address, poll_table *p); |
1da177e4c Linux-2.6.12-rc2 |
54 55 56 57 58 59 |
void poll_initwait(struct poll_wqueues *pwq) { init_poll_funcptr(&pwq->pt, __pollwait); pwq->error = 0; pwq->table = NULL; |
70674f95c [PATCH] Optimize ... |
60 |
pwq->inline_index = 0; |
1da177e4c Linux-2.6.12-rc2 |
61 62 63 |
} EXPORT_SYMBOL(poll_initwait); |
70674f95c [PATCH] Optimize ... |
64 65 66 67 68 |
static void free_poll_entry(struct poll_table_entry *entry) { remove_wait_queue(entry->wait_address,&entry->wait); fput(entry->filp); } |
1da177e4c Linux-2.6.12-rc2 |
69 70 71 |
void poll_freewait(struct poll_wqueues *pwq) { struct poll_table_page * p = pwq->table; |
70674f95c [PATCH] Optimize ... |
72 73 74 |
int i; for (i = 0; i < pwq->inline_index; i++) free_poll_entry(pwq->inline_entries + i); |
1da177e4c Linux-2.6.12-rc2 |
75 76 77 78 79 80 81 |
while (p) { struct poll_table_entry * entry; struct poll_table_page *old; entry = p->entry; do { entry--; |
70674f95c [PATCH] Optimize ... |
82 |
free_poll_entry(entry); |
1da177e4c Linux-2.6.12-rc2 |
83 84 85 86 87 88 89 90 |
} while (entry > p->entries); old = p; p = p->next; free_page((unsigned long) old); } } EXPORT_SYMBOL(poll_freewait); |
70674f95c [PATCH] Optimize ... |
91 |
static struct poll_table_entry *poll_get_entry(poll_table *_p) |
1da177e4c Linux-2.6.12-rc2 |
92 93 94 |
{ struct poll_wqueues *p = container_of(_p, struct poll_wqueues, pt); struct poll_table_page *table = p->table; |
70674f95c [PATCH] Optimize ... |
95 96 |
if (p->inline_index < N_INLINE_POLL_ENTRIES) return p->inline_entries + p->inline_index++; |
1da177e4c Linux-2.6.12-rc2 |
97 98 99 100 101 102 103 |
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; __set_current_state(TASK_RUNNING); |
70674f95c [PATCH] Optimize ... |
104 |
return NULL; |
1da177e4c Linux-2.6.12-rc2 |
105 106 107 108 109 110 |
} new_table->entry = new_table->entries; new_table->next = table; p->table = new_table; table = new_table; } |
70674f95c [PATCH] Optimize ... |
111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 |
return table->entry++; } /* Add a new entry */ static void __pollwait(struct file *filp, wait_queue_head_t *wait_address, poll_table *p) { struct poll_table_entry *entry = poll_get_entry(p); if (!entry) return; get_file(filp); entry->filp = filp; entry->wait_address = wait_address; init_waitqueue_entry(&entry->wait, current); add_wait_queue(wait_address,&entry->wait); |
1da177e4c Linux-2.6.12-rc2 |
126 127 128 129 130 131 132 133 134 135 136 137 138 |
} #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... |
139 |
struct fdtable *fdt; |
1da177e4c Linux-2.6.12-rc2 |
140 141 142 143 |
/* handle last in-complete long-word first */ set = ~(~0UL << (n & (__NFDBITS-1))); n /= __NFDBITS; |
badf16621 [PATCH] files: br... |
144 145 |
fdt = files_fdtable(current->files); open_fds = fdt->open_fds->fds_bits+n; |
1da177e4c Linux-2.6.12-rc2 |
146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 |
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); max += n * __NFDBITS; } return max; } #define BIT(i) (1UL << ((i)&(__NFDBITS-1))) #define MEM(i,m) ((m)+(unsigned)(i)/__NFDBITS) #define ISSET(i,m) (((i)&*(m)) != 0) #define SET(i,m) (*(m) |= (i)) #define POLLIN_SET (POLLRDNORM | POLLRDBAND | POLLIN | POLLHUP | POLLERR) #define POLLOUT_SET (POLLWRBAND | POLLWRNORM | POLLOUT | POLLERR) #define POLLEX_SET (POLLPRI) |
9f72949f6 [PATCH] Add psele... |
184 |
int do_select(int n, fd_set_bits *fds, s64 *timeout) |
1da177e4c Linux-2.6.12-rc2 |
185 186 187 188 |
{ struct poll_wqueues table; poll_table *wait; int retval, i; |
1da177e4c Linux-2.6.12-rc2 |
189 |
|
b835996f6 [PATCH] files: lo... |
190 |
rcu_read_lock(); |
1da177e4c Linux-2.6.12-rc2 |
191 |
retval = max_select_fd(n, fds); |
b835996f6 [PATCH] files: lo... |
192 |
rcu_read_unlock(); |
1da177e4c Linux-2.6.12-rc2 |
193 194 195 196 197 198 199 |
if (retval < 0) return retval; n = retval; poll_initwait(&table); wait = &table.pt; |
9f72949f6 [PATCH] Add psele... |
200 |
if (!*timeout) |
1da177e4c Linux-2.6.12-rc2 |
201 202 203 204 |
wait = NULL; retval = 0; for (;;) { unsigned long *rinp, *routp, *rexp, *inp, *outp, *exp; |
9f72949f6 [PATCH] Add psele... |
205 |
long __timeout; |
1da177e4c Linux-2.6.12-rc2 |
206 207 208 209 210 211 212 213 214 |
set_current_state(TASK_INTERRUPTIBLE); 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; |
99ac48f54 [PATCH] mark f_op... |
215 |
const struct file_operations *f_op = NULL; |
1da177e4c Linux-2.6.12-rc2 |
216 217 218 219 220 221 222 223 224 225 |
struct file *file = NULL; in = *inp++; out = *outp++; ex = *exp++; all_bits = in | out | ex; if (all_bits == 0) { i += __NFDBITS; continue; } for (j = 0; j < __NFDBITS; ++j, ++i, bit <<= 1) { |
e4a1f129f [PATCH] use fget_... |
226 |
int fput_needed; |
1da177e4c Linux-2.6.12-rc2 |
227 228 229 230 |
if (i >= n) break; if (!(bit & all_bits)) continue; |
e4a1f129f [PATCH] use fget_... |
231 |
file = fget_light(i, &fput_needed); |
1da177e4c Linux-2.6.12-rc2 |
232 233 234 235 236 |
if (file) { f_op = file->f_op; mask = DEFAULT_POLLMASK; if (f_op && f_op->poll) mask = (*f_op->poll)(file, retval ? NULL : wait); |
e4a1f129f [PATCH] use fget_... |
237 |
fput_light(file, fput_needed); |
1da177e4c Linux-2.6.12-rc2 |
238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 |
if ((mask & POLLIN_SET) && (in & bit)) { res_in |= bit; retval++; } if ((mask & POLLOUT_SET) && (out & bit)) { res_out |= bit; retval++; } if ((mask & POLLEX_SET) && (ex & bit)) { res_ex |= bit; retval++; } } cond_resched(); } if (res_in) *rinp = res_in; if (res_out) *routp = res_out; if (res_ex) *rexp = res_ex; } wait = NULL; |
9f72949f6 [PATCH] Add psele... |
261 |
if (retval || !*timeout || signal_pending(current)) |
1da177e4c Linux-2.6.12-rc2 |
262 263 264 265 266 |
break; if(table.error) { retval = table.error; break; } |
9f72949f6 [PATCH] Add psele... |
267 268 269 270 271 272 273 274 275 276 277 278 |
if (*timeout < 0) { /* Wait indefinitely */ __timeout = MAX_SCHEDULE_TIMEOUT; } else if (unlikely(*timeout >= (s64)MAX_SCHEDULE_TIMEOUT - 1)) { /* Wait for longer than MAX_SCHEDULE_TIMEOUT. Do it in a loop */ __timeout = MAX_SCHEDULE_TIMEOUT - 1; *timeout -= __timeout; } else { __timeout = *timeout; *timeout = 0; } |
1da177e4c Linux-2.6.12-rc2 |
279 |
__timeout = schedule_timeout(__timeout); |
9f72949f6 [PATCH] Add psele... |
280 281 |
if (*timeout >= 0) *timeout += __timeout; |
1da177e4c Linux-2.6.12-rc2 |
282 283 284 285 |
} __set_current_state(TASK_RUNNING); poll_freewait(&table); |
1da177e4c Linux-2.6.12-rc2 |
286 287 |
return retval; } |
1da177e4c Linux-2.6.12-rc2 |
288 289 290 291 292 293 294 295 296 297 |
/* * 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. */ #define MAX_SELECT_SECONDS \ ((unsigned long) (MAX_SCHEDULE_TIMEOUT / HZ)-1) |
9f72949f6 [PATCH] Add psele... |
298 299 |
static int core_sys_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, s64 *timeout) |
1da177e4c Linux-2.6.12-rc2 |
300 301 |
{ fd_set_bits fds; |
29ff2db55 [PATCH] select() ... |
302 |
void *bits; |
bbea9f696 [PATCH] fdtable: ... |
303 |
int ret, max_fds; |
b04eb6aa0 [PATCH] select: d... |
304 |
unsigned int size; |
badf16621 [PATCH] files: br... |
305 |
struct fdtable *fdt; |
70674f95c [PATCH] Optimize ... |
306 |
/* Allocate small arguments on the stack to save memory and be faster */ |
30c14e40e [PATCH] avoid una... |
307 |
long stack_fds[SELECT_STACK_ALLOC/sizeof(long)]; |
1da177e4c Linux-2.6.12-rc2 |
308 |
|
1da177e4c Linux-2.6.12-rc2 |
309 310 311 |
ret = -EINVAL; if (n < 0) goto out_nofds; |
bbea9f696 [PATCH] fdtable: ... |
312 |
/* max_fds can increase, so grab it once to avoid race */ |
b835996f6 [PATCH] files: lo... |
313 |
rcu_read_lock(); |
badf16621 [PATCH] files: br... |
314 |
fdt = files_fdtable(current->files); |
bbea9f696 [PATCH] fdtable: ... |
315 |
max_fds = fdt->max_fds; |
b835996f6 [PATCH] files: lo... |
316 |
rcu_read_unlock(); |
bbea9f696 [PATCH] fdtable: ... |
317 318 |
if (n > max_fds) n = max_fds; |
1da177e4c Linux-2.6.12-rc2 |
319 320 321 322 323 324 |
/* * 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 |
325 |
size = FDS_BYTES(n); |
b04eb6aa0 [PATCH] select: d... |
326 327 328 329 |
bits = stack_fds; if (size > sizeof(stack_fds) / 6) { /* Not enough space in on-stack array; must use kmalloc */ ret = -ENOMEM; |
70674f95c [PATCH] Optimize ... |
330 |
bits = kmalloc(6 * size, GFP_KERNEL); |
b04eb6aa0 [PATCH] select: d... |
331 332 333 |
if (!bits) goto out_nofds; } |
29ff2db55 [PATCH] select() ... |
334 335 336 337 338 339 |
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 |
340 341 342 343 344 345 346 347 |
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); |
9f72949f6 [PATCH] Add psele... |
348 |
ret = do_select(n, &fds, timeout); |
1da177e4c Linux-2.6.12-rc2 |
349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 |
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 ... |
365 366 |
if (bits != stack_fds) kfree(bits); |
1da177e4c Linux-2.6.12-rc2 |
367 368 369 |
out_nofds: return ret; } |
9f72949f6 [PATCH] Add psele... |
370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 |
asmlinkage long sys_select(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, struct timeval __user *tvp) { s64 timeout = -1; struct timeval tv; int ret; if (tvp) { if (copy_from_user(&tv, tvp, sizeof(tv))) return -EFAULT; if (tv.tv_sec < 0 || tv.tv_usec < 0) return -EINVAL; /* Cast to u64 to make GCC stop complaining */ if ((u64)tv.tv_sec >= (u64)MAX_INT64_SECONDS) timeout = -1; /* infinite */ else { timeout = ROUND_UP(tv.tv_usec, USEC_PER_SEC/HZ); timeout += tv.tv_sec * HZ; } } ret = core_sys_select(n, inp, outp, exp, &timeout); if (tvp) { |
643a65454 [PATCH] select: f... |
396 |
struct timeval rtv; |
9f72949f6 [PATCH] Add psele... |
397 398 |
if (current->personality & STICKY_TIMEOUTS) goto sticky; |
643a65454 [PATCH] select: f... |
399 400 |
rtv.tv_usec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)); rtv.tv_sec = timeout; |
74910e6c7 [PATCH] select: t... |
401 |
if (timeval_compare(&rtv, &tv) >= 0) |
643a65454 [PATCH] select: f... |
402 403 |
rtv = tv; if (copy_to_user(tvp, &rtv, sizeof(rtv))) { |
9f72949f6 [PATCH] Add psele... |
404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 |
sticky: /* * 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. */ if (ret == -ERESTARTNOHAND) ret = -EINTR; } } return ret; } #ifdef TIF_RESTORE_SIGMASK asmlinkage long sys_pselect7(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) { s64 timeout = MAX_SCHEDULE_TIMEOUT; sigset_t ksigmask, sigsaved; struct timespec ts; int ret; if (tsp) { if (copy_from_user(&ts, tsp, sizeof(ts))) return -EFAULT; if (ts.tv_sec < 0 || ts.tv_nsec < 0) return -EINVAL; /* Cast to u64 to make GCC stop complaining */ if ((u64)ts.tv_sec >= (u64)MAX_INT64_SECONDS) timeout = -1; /* infinite */ else { timeout = ROUND_UP(ts.tv_nsec, NSEC_PER_SEC/HZ); timeout += ts.tv_sec * HZ; } } 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); } ret = core_sys_select(n, inp, outp, exp, &timeout); if (tsp) { |
643a65454 [PATCH] select: f... |
461 |
struct timespec rts; |
9f72949f6 [PATCH] Add psele... |
462 463 |
if (current->personality & STICKY_TIMEOUTS) goto sticky; |
643a65454 [PATCH] select: f... |
464 465 466 |
rts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) * 1000; rts.tv_sec = timeout; |
74910e6c7 [PATCH] select: t... |
467 |
if (timespec_compare(&rts, &ts) >= 0) |
643a65454 [PATCH] select: f... |
468 469 |
rts = ts; if (copy_to_user(tsp, &rts, sizeof(rts))) { |
9f72949f6 [PATCH] Add psele... |
470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 |
sticky: /* * 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. */ if (ret == -ERESTARTNOHAND) ret = -EINTR; } } 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_thread_flag(TIF_RESTORE_SIGMASK); } } 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. */ asmlinkage long sys_pselect6(int n, fd_set __user *inp, fd_set __user *outp, fd_set __user *exp, struct timespec __user *tsp, void __user *sig) { 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... |
515 |
|| __get_user(up, (sigset_t __user * __user *)sig) |
9f72949f6 [PATCH] Add psele... |
516 |
|| __get_user(sigsetsize, |
e110ab94e [PATCH] fix __use... |
517 |
(size_t __user *)(sig+sizeof(void *)))) |
9f72949f6 [PATCH] Add psele... |
518 519 520 521 522 523 |
return -EFAULT; } return sys_pselect7(n, inp, outp, exp, tsp, up, sigsetsize); } #endif /* TIF_RESTORE_SIGMASK */ |
1da177e4c Linux-2.6.12-rc2 |
524 525 526 527 528 529 530 |
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... |
531 532 533 534 535 536 537 538 |
/* * 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, * if non-NULL. */ static inline unsigned int do_pollfd(struct pollfd *pollfd, poll_table *pwait) |
1da177e4c Linux-2.6.12-rc2 |
539 |
{ |
4a4b69f79 [PATCH] Poll clea... |
540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 |
unsigned int mask; int fd; mask = 0; fd = pollfd->fd; if (fd >= 0) { int fput_needed; struct file * file; file = fget_light(fd, &fput_needed); mask = POLLNVAL; if (file != NULL) { mask = DEFAULT_POLLMASK; if (file->f_op && file->f_op->poll) mask = file->f_op->poll(file, pwait); /* Mask out unneeded events. */ mask &= pollfd->events | POLLERR | POLLHUP; fput_light(file, fput_needed); |
1da177e4c Linux-2.6.12-rc2 |
558 |
} |
1da177e4c Linux-2.6.12-rc2 |
559 |
} |
4a4b69f79 [PATCH] Poll clea... |
560 561 562 |
pollfd->revents = mask; return mask; |
1da177e4c Linux-2.6.12-rc2 |
563 564 565 |
} static int do_poll(unsigned int nfds, struct poll_list *list, |
9f72949f6 [PATCH] Add psele... |
566 |
struct poll_wqueues *wait, s64 *timeout) |
1da177e4c Linux-2.6.12-rc2 |
567 568 569 |
{ int count = 0; poll_table* pt = &wait->pt; |
9f72949f6 [PATCH] Add psele... |
570 571 |
/* Optimise the no-wait case */ if (!(*timeout)) |
1da177e4c Linux-2.6.12-rc2 |
572 573 574 575 |
pt = NULL; for (;;) { struct poll_list *walk; |
9f72949f6 [PATCH] Add psele... |
576 |
long __timeout; |
1da177e4c Linux-2.6.12-rc2 |
577 |
set_current_state(TASK_INTERRUPTIBLE); |
4a4b69f79 [PATCH] Poll clea... |
578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 |
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 * and kill the poll_table, so we don't * needlessly register any other waiters after * this. They'll get immediately deregistered * when we break out and return. */ if (do_pollfd(pfd, pt)) { count++; pt = NULL; } } |
1da177e4c Linux-2.6.12-rc2 |
596 |
} |
4a4b69f79 [PATCH] Poll clea... |
597 598 599 600 |
/* * All waiters have already been registered, so don't provide * a poll_table to them on the next loop iteration. */ |
1da177e4c Linux-2.6.12-rc2 |
601 |
pt = NULL; |
9f72949f6 [PATCH] Add psele... |
602 |
if (count || !*timeout || signal_pending(current)) |
1da177e4c Linux-2.6.12-rc2 |
603 604 605 606 |
break; count = wait->error; if (count) break; |
9f72949f6 [PATCH] Add psele... |
607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 |
if (*timeout < 0) { /* Wait indefinitely */ __timeout = MAX_SCHEDULE_TIMEOUT; } else if (unlikely(*timeout >= (s64)MAX_SCHEDULE_TIMEOUT-1)) { /* * Wait for longer than MAX_SCHEDULE_TIMEOUT. Do it in * a loop */ __timeout = MAX_SCHEDULE_TIMEOUT - 1; *timeout -= __timeout; } else { __timeout = *timeout; *timeout = 0; } __timeout = schedule_timeout(__timeout); if (*timeout >= 0) *timeout += __timeout; |
1da177e4c Linux-2.6.12-rc2 |
626 627 628 629 |
} __set_current_state(TASK_RUNNING); return count; } |
70674f95c [PATCH] Optimize ... |
630 631 |
#define N_STACK_PPS ((sizeof(stack_pps) - sizeof(struct poll_list)) / \ sizeof(struct pollfd)) |
9f72949f6 [PATCH] Add psele... |
632 |
int do_sys_poll(struct pollfd __user *ufds, unsigned int nfds, s64 *timeout) |
1da177e4c Linux-2.6.12-rc2 |
633 634 635 636 637 638 |
{ struct poll_wqueues table; int fdcount, err; unsigned int i; struct poll_list *head; struct poll_list *walk; |
30c14e40e [PATCH] avoid una... |
639 640 641 642 |
/* 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)]; |
70674f95c [PATCH] Optimize ... |
643 |
struct poll_list *stack_pp = NULL; |
1da177e4c Linux-2.6.12-rc2 |
644 645 |
/* Do a sanity check on nfds ... */ |
4e6fd33b7 [PATCH] enforce R... |
646 |
if (nfds > current->signal->rlim[RLIMIT_NOFILE].rlim_cur) |
1da177e4c Linux-2.6.12-rc2 |
647 |
return -EINVAL; |
1da177e4c Linux-2.6.12-rc2 |
648 649 650 651 652 653 654 655 |
poll_initwait(&table); head = NULL; walk = NULL; i = nfds; err = -ENOMEM; while(i!=0) { struct poll_list *pp; |
70674f95c [PATCH] Optimize ... |
656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 |
int num, size; if (stack_pp == NULL) num = N_STACK_PPS; else num = POLLFD_PER_PAGE; if (num > i) num = i; size = sizeof(struct poll_list) + sizeof(struct pollfd)*num; if (!stack_pp) stack_pp = pp = (struct poll_list *)stack_pps; else { pp = kmalloc(size, GFP_KERNEL); if (!pp) goto out_fds; } |
1da177e4c Linux-2.6.12-rc2 |
671 |
pp->next=NULL; |
70674f95c [PATCH] Optimize ... |
672 |
pp->len = num; |
1da177e4c Linux-2.6.12-rc2 |
673 674 675 676 677 678 679 |
if (head == NULL) head = pp; else walk->next = pp; walk = pp; if (copy_from_user(pp->entries, ufds + nfds-i, |
70674f95c [PATCH] Optimize ... |
680 |
sizeof(struct pollfd)*num)) { |
1da177e4c Linux-2.6.12-rc2 |
681 682 683 684 685 |
err = -EFAULT; goto out_fds; } i -= pp->len; } |
9f72949f6 [PATCH] Add psele... |
686 |
|
1da177e4c Linux-2.6.12-rc2 |
687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 |
fdcount = do_poll(nfds, head, &table, timeout); /* OK, now copy the revents fields back to user space. */ walk = head; err = -EFAULT; while(walk != NULL) { struct pollfd *fds = walk->entries; int j; for (j=0; j < walk->len; j++, ufds++) { if(__put_user(fds[j].revents, &ufds->revents)) goto out_fds; } walk = walk->next; } err = fdcount; if (!fdcount && signal_pending(current)) err = -EINTR; out_fds: walk = head; while(walk!=NULL) { struct poll_list *pp = walk->next; |
70674f95c [PATCH] Optimize ... |
709 710 |
if (walk != stack_pp) kfree(walk); |
1da177e4c Linux-2.6.12-rc2 |
711 712 713 714 715 |
walk = pp; } poll_freewait(&table); return err; } |
9f72949f6 [PATCH] Add psele... |
716 717 718 719 |
asmlinkage long sys_poll(struct pollfd __user *ufds, unsigned int nfds, long timeout_msecs) { |
04a3446c9 [PATCH] fs: sys_p... |
720 |
s64 timeout_jiffies; |
9f72949f6 [PATCH] Add psele... |
721 |
|
04a3446c9 [PATCH] fs: sys_p... |
722 |
if (timeout_msecs > 0) { |
9f72949f6 [PATCH] Add psele... |
723 724 725 726 727 728 729 |
#if HZ > 1000 /* We can only overflow if HZ > 1000 */ if (timeout_msecs / 1000 > (s64)0x7fffffffffffffffULL / (s64)HZ) timeout_jiffies = -1; else #endif timeout_jiffies = msecs_to_jiffies(timeout_msecs); |
04a3446c9 [PATCH] fs: sys_p... |
730 731 732 |
} else { /* Infinite (< 0) or no (0) timeout */ timeout_jiffies = timeout_msecs; |
9f72949f6 [PATCH] Add psele... |
733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 |
} return do_sys_poll(ufds, nfds, &timeout_jiffies); } #ifdef TIF_RESTORE_SIGMASK asmlinkage long sys_ppoll(struct pollfd __user *ufds, unsigned int nfds, struct timespec __user *tsp, const sigset_t __user *sigmask, size_t sigsetsize) { sigset_t ksigmask, sigsaved; struct timespec ts; s64 timeout = -1; int ret; if (tsp) { if (copy_from_user(&ts, tsp, sizeof(ts))) return -EFAULT; /* Cast to u64 to make GCC stop complaining */ if ((u64)ts.tv_sec >= (u64)MAX_INT64_SECONDS) timeout = -1; /* infinite */ else { timeout = ROUND_UP(ts.tv_nsec, NSEC_PER_SEC/HZ); timeout += ts.tv_sec * HZ; } } 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); } ret = do_sys_poll(ufds, nfds, &timeout); /* 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_thread_flag(TIF_RESTORE_SIGMASK); } ret = -ERESTARTNOHAND; } else if (sigmask) sigprocmask(SIG_SETMASK, &sigsaved, NULL); if (tsp && timeout >= 0) { |
643a65454 [PATCH] select: f... |
791 |
struct timespec rts; |
9f72949f6 [PATCH] Add psele... |
792 793 794 |
if (current->personality & STICKY_TIMEOUTS) goto sticky; /* Yes, we know it's actually an s64, but it's also positive. */ |
643a65454 [PATCH] select: f... |
795 796 797 |
rts.tv_nsec = jiffies_to_usecs(do_div((*(u64*)&timeout), HZ)) * 1000; rts.tv_sec = timeout; |
74910e6c7 [PATCH] select: t... |
798 |
if (timespec_compare(&rts, &ts) >= 0) |
643a65454 [PATCH] select: f... |
799 800 |
rts = ts; if (copy_to_user(tsp, &rts, sizeof(rts))) { |
9f72949f6 [PATCH] Add psele... |
801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 |
sticky: /* * 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. */ if (ret == -ERESTARTNOHAND && timeout >= 0) ret = -EINTR; } } return ret; } #endif /* TIF_RESTORE_SIGMASK */ |