ev_epoll.c

/*
 * libev epoll fd activity backend
 *
 * Copyright (c) 2007,2008,2009,2010,2011 Marc Alexander Lehmann <libev@schmorp.de>
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without modifica-
 * tion, are permitted provided that the following conditions are met:
 *
 *   1.  Redistributions of source code must retain the above copyright notice,
 *       this list of conditions and the following disclaimer.
 *
 *   2.  Redistributions in binary form must reproduce the above copyright
 *       notice, this list of conditions and the following disclaimer in the
 *       documentation and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR IMPLIED
 * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MER-
 * CHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO
 * EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPE-
 * CIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS;
 * OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY,
 * WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTH-
 * ERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
 * OF THE POSSIBILITY OF SUCH DAMAGE.
 *
 * Alternatively, the contents of this file may be used under the terms of
 * the GNU General Public License ("GPL") version 2 or any later version,
 * in which case the provisions of the GPL are applicable instead of
 * the above. If you wish to allow the use of your version of this file
 * only under the terms of the GPL and not to allow others to use your
 * version of this file under the BSD license, indicate your decision
 * by deleting the provisions above and replace them with the notice
 * and other provisions required by the GPL. If you do not delete the
 * provisions above, a recipient may use your version of this file under
 * either the BSD or the GPL.
 */

/*
 * general notes about epoll:
 *
 * a) epoll silently removes fds from the fd set. as nothing tells us
 *    that an fd has been removed otherwise, we have to continually
 *    "rearm" fds that we suspect *might* have changed (same
 *    problem with kqueue, but much less costly there).
 * b) the fact that ADD != MOD creates a lot of extra syscalls due to a)
 *    and seems not to have any advantage.
 * c) the inability to handle fork or file descriptors (think dup)
 *    limits the applicability over poll, so this is not a generic
 *    poll replacement.
 * d) epoll doesn't work the same as select with many file descriptors
 *    (such as files). while not critical, no other advanced interface
 *    seems to share this (rather non-unixy) limitation.
 * e) epoll claims to be embeddable, but in practise you never get
 *    a ready event for the epoll fd (broken: <=2.6.26, working: >=2.6.32).
 * f) epoll_ctl returning EPERM means the fd is always ready.
 *
 * lots of "weird code" and complication handling in this file is due
 * to these design problems with epoll, as we try very hard to avoid
 * epoll_ctl syscalls for common usage patterns and handle the breakage
 * ensuing from receiving events for closed and otherwise long gone
 * file descriptors.
 */

#include <sys/epoll.h>

#define EV_EMASK_EPERM 0x80

static void
epoll_modify (EV_P_ int fd, int oev, int nev)
{
  struct epoll_event ev;
  unsigned char oldmask;

  /*
   * we handle EPOLL_CTL_DEL by ignoring it here
   * on the assumption that the fd is gone anyways
   * if that is wrong, we have to handle the spurious
   * event in epoll_poll.
   * if the fd is added again, we try to ADD it, and, if that
   * fails, we assume it still has the same eventmask.
   */
  if (!nev)
    return;

  oldmask = anfds [fd].emask;
  anfds [fd].emask = nev;

  /* store the generation counter in the upper 32 bits, the fd in the lower 32 bits */
  ev.data.u64 = (uint64_t)(uint32_t)fd
              | ((uint64_t)(uint32_t)++anfds [fd].egen << 32);
  ev.events   = (nev & EV_READ  ? EPOLLIN  : 0)
              | (nev & EV_WRITE ? EPOLLOUT : 0);

  if (expect_true (!epoll_ctl (backend_fd, oev && oldmask != nev ? EPOLL_CTL_MOD : EPOLL_CTL_ADD, fd, &ev)))
    return;

  if (expect_true (errno == ENOENT))
    {
      /* if ENOENT then the fd went away, so try to do the right thing */
      if (!nev)
        goto dec_egen;

      if (!epoll_ctl (backend_fd, EPOLL_CTL_ADD, fd, &ev))
        return;
    }
  else if (expect_true (errno == EEXIST))
    {
      /* EEXIST means we ignored a previous DEL, but the fd is still active */
      /* if the kernel mask is the same as the new mask, we assume it hasn't changed */
      if (oldmask == nev)
        goto dec_egen;

      if (!epoll_ctl (backend_fd, EPOLL_CTL_MOD, fd, &ev))
        return;
    }
  else if (expect_true (errno == EPERM))
    {
      /* EPERM means the fd is always ready, but epoll is too snobbish */
      /* to handle it, unlike select or poll. */
      anfds [fd].emask = EV_EMASK_EPERM;

      /* add fd to epoll_eperms, if not already inside */
      if (!(oldmask & EV_EMASK_EPERM))
        {
          array_needsize (int, epoll_eperms, epoll_epermmax, epoll_epermcnt + 1, EMPTY2);
          epoll_eperms [epoll_epermcnt++] = fd;
        }

      return;
    }

  fd_kill (EV_A_ fd);

dec_egen:
  /* we didn't successfully call epoll_ctl, so decrement the generation counter again */
  --anfds [fd].egen;
}

static void
epoll_poll (EV_P_ ev_tstamp timeout)
{
  int i;
  int eventcnt;

  if (expect_false (epoll_epermcnt))
    timeout = 0.;

  /* epoll wait times cannot be larger than (LONG_MAX - 999UL) / HZ msecs, which is below */
  /* the default libev max wait time, however. */
  EV_RELEASE_CB;
  eventcnt = epoll_wait (backend_fd, epoll_events, epoll_eventmax, timeout * 1e3);
  EV_ACQUIRE_CB;

  if (expect_false (eventcnt < 0))
    {
      if (errno != EINTR)
        ev_syserr ("(libev) epoll_wait");

      return;
    }

  for (i = 0; i < eventcnt; ++i)
    {
      struct epoll_event *ev = epoll_events + i;

      int fd = (uint32_t)ev->data.u64; /* mask out the lower 32 bits */
      int want = anfds [fd].events;
      int got  = (ev->events & (EPOLLOUT | EPOLLERR | EPOLLHUP) ? EV_WRITE : 0)
               | (ev->events & (EPOLLIN  | EPOLLERR | EPOLLHUP) ? EV_READ  : 0);

      /*
       * check for spurious notification.
       * this only finds spurious notifications on egen updates
       * other spurious notifications will be found by epoll_ctl, below
       * we assume that fd is always in range, as we never shrink the anfds array
       */
      if (expect_false ((uint32_t)anfds [fd].egen != (uint32_t)(ev->data.u64 >> 32)))
        {
          /* recreate kernel state */
          postfork |= 2;
          continue;
        }

      if (expect_false (got & ~want))
        {
          anfds [fd].emask = want;

          /*
           * we received an event but are not interested in it, try mod or del
           * this often happens because we optimistically do not unregister fds
           * when we are no longer interested in them, but also when we get spurious
           * notifications for fds from another process. this is partially handled
           * above with the gencounter check (== our fd is not the event fd), and
           * partially here, when epoll_ctl returns an error (== a child has the fd
           * but we closed it).
           */
          ev->events = (want & EV_READ  ? EPOLLIN  : 0)
                     | (want & EV_WRITE ? EPOLLOUT : 0);

          /* pre-2.6.9 kernels require a non-null pointer with EPOLL_CTL_DEL, */
          /* which is fortunately easy to do for us. */
          if (epoll_ctl (backend_fd, want ? EPOLL_CTL_MOD : EPOLL_CTL_DEL, fd, ev))
            {
              postfork |= 2; /* an error occurred, recreate kernel state */
              continue;
            }
        }

      fd_event (EV_A_ fd, got);
    }

  /* if the receive array was full, increase its size */
  if (expect_false (eventcnt == epoll_eventmax))
    {
      ev_free (epoll_events);
      epoll_eventmax = array_nextsize (sizeof (struct epoll_event), epoll_eventmax, epoll_eventmax + 1);
      epoll_events = (struct epoll_event *)ev_malloc (sizeof (struct epoll_event) * epoll_eventmax);
    }

  /* now synthesize events for all fds where epoll fails, while select works... */
  for (i = epoll_epermcnt; i--; )
    {
      int fd = epoll_eperms [i];
      unsigned char events = anfds [fd].events & (EV_READ | EV_WRITE);

      if (anfds [fd].emask & EV_EMASK_EPERM && events)
        fd_event (EV_A_ fd, events);
      else
        {
          epoll_eperms [i] = epoll_eperms [--epoll_epermcnt];
          anfds [fd].emask = 0;
        }
    }
}

inline_size
int
epoll_init (EV_P_ int flags)
{
#ifdef EPOLL_CLOEXEC
  backend_fd = epoll_create1 (EPOLL_CLOEXEC);

  if (backend_fd < 0 && (errno == EINVAL || errno == ENOSYS))
#endif
    backend_fd = epoll_create (256);

  if (backend_fd < 0)
    return 0;

  fcntl (backend_fd, F_SETFD, FD_CLOEXEC);

  backend_mintime = 1e-3; /* epoll does sometimes return early, this is just to avoid the worst */
  backend_modify  = epoll_modify;
  backend_poll    = epoll_poll;

  epoll_eventmax = 64; /* initial number of events receivable per poll */
  epoll_events = (struct epoll_event *)ev_malloc (sizeof (struct epoll_event) * epoll_eventmax);

  return EVBACKEND_EPOLL;
}

inline_size
void
epoll_destroy (EV_P)
{
  ev_free (epoll_events);
  array_free (epoll_eperm, EMPTY);
}

inline_size
void
epoll_fork (EV_P)
{
  close (backend_fd);

  while ((backend_fd = epoll_create (256)) < 0)
    ev_syserr ("(libev) epoll_create");

  fcntl (backend_fd, F_SETFD, FD_CLOEXEC);

  fd_rearm_all (EV_A);
}