diff --git a/src/poll.rs b/src/poll.rs
index 47d7d22201..60c31aceed 100644
--- a/src/poll.rs
+++ b/src/poll.rs
@@ -2,10 +2,19 @@
 use sys::time::TimeSpec;
 #[cfg(any(target_os = "android", target_os = "dragonfly", target_os = "freebsd", target_os = "linux"))]
 use sys::signal::SigSet;
+use std::os::unix::io::RawFd;
 
 use libc;
 use {Errno, Result};
 
+/// This is a wrapper around `libc::pollfd`.
+///
+/// It's meant to be used as an argument to the [`poll`](fn.poll.html) and
+/// [`ppoll`](fn.ppoll.html) functions to specify the events of interest
+/// for a specific file descriptor.
+///
+/// After a call to `poll` or `ppoll`, the events that occured can be
+/// retrieved by calling [`revents()`](#method.revents) on the `PollFd`.
 #[repr(C)]
 #[derive(Clone, Copy)]
 pub struct PollFd {
@@ -13,7 +22,9 @@ pub struct PollFd {
 }
 
 impl PollFd {
-    pub fn new(fd: libc::c_int, events: EventFlags) -> PollFd {
+    /// Creates a new `PollFd` specifying the events of interest
+    /// for a given file descriptor.
+    pub fn new(fd: RawFd, events: EventFlags) -> PollFd {
         PollFd {
             pollfd: libc::pollfd {
                 fd: fd,
@@ -23,26 +34,82 @@ impl PollFd {
         }
     }
 
+    /// Returns the events that occured in the last call to `poll` or `ppoll`.
     pub fn revents(&self) -> Option<EventFlags> {
         EventFlags::from_bits(self.pollfd.revents)
     }
 }
 
 libc_bitflags! {
+    /// These flags define the different events that can be monitored by `poll` and `ppoll`
     pub flags EventFlags: libc::c_short {
+        /// There is data to read.
         POLLIN,
+        /// There is some exceptional condition on the file descriptor.
+        ///
+        /// Possibilities include:
+        ///
+        /// *  There is out-of-band data on a TCP socket (see
+        ///    [tcp(7)](http://man7.org/linux/man-pages/man7/tcp.7.html)).
+        /// *  A pseudoterminal master in packet mode has seen a state
+        ///    change on the slave (see
+        ///    [ioctl_tty(2)](http://man7.org/linux/man-pages/man2/ioctl_tty.2.html)).
+        /// *  A cgroup.events file has been modified (see
+        ///    [cgroups(7)](http://man7.org/linux/man-pages/man7/cgroups.7.html)).
         POLLPRI,
+        /// Writing is now possible, though a write larger that the
+        /// available space in a socket or pipe will still block (unless
+        /// `O_NONBLOCK` is set).
         POLLOUT,
+        /// Equivalent to [`POLLIN`](constant.POLLIN.html)
         POLLRDNORM,
+        /// Equivalent to [`POLLOUT`](constant.POLLOUT.html)
         POLLWRNORM,
+        /// Priority band data can be read (generally unused on Linux).
         POLLRDBAND,
+        /// Priority data may be written.
         POLLWRBAND,
+        /// Error condition (only returned in
+        /// [`PollFd::revents`](struct.PollFd.html#method.revents);
+        /// ignored in [`PollFd::new`](struct.PollFd.html#method.new)).
+        /// This bit is also set for a file descriptor referring to the
+        /// write end of a pipe when the read end has been closed.
         POLLERR,
+        /// Hang up (only returned in [`PollFd::revents`](struct.PollFd.html#method.revents);
+        /// ignored in [`PollFd::new`](struct.PollFd.html#method.new)).
+        /// Note that when reading from a channel such as a pipe or a stream
+        /// socket, this event merely indicates that the peer closed its
+        /// end of the channel.  Subsequent reads from the channel will
+        /// return 0 (end of file) only after all outstanding data in the
+        /// channel has been consumed.
         POLLHUP,
+        /// Invalid request: `fd` not open (only returned in
+        /// [`PollFd::revents`](struct.PollFd.html#method.revents);
+        /// ignored in [`PollFd::new`](struct.PollFd.html#method.new)).
         POLLNVAL,
     }
 }
 
+/// `poll` waits for one of a set of file descriptors to become ready to perform I/O.
+/// ([`poll(2)`](http://man7.org/linux/man-pages/man2/poll.2.html))
+///
+/// `fds` contains all [`PollFd`](struct.PollFd.html) to poll.
+/// The function will return as soon as any event occur for any of these `PollFd`s.
+///
+/// The `timeout` argument specifies the number of milliseconds that `poll()`
+/// should block waiting for a file descriptor to become ready.  The call
+/// will block until either:
+///
+/// *  a file descriptor becomes ready;
+/// *  the call is interrupted by a signal handler; or
+/// *  the timeout expires.
+///
+/// Note that the timeout interval will be rounded up to the system clock
+/// granularity, and kernel scheduling delays mean that the blocking
+/// interval may overrun by a small amount.  Specifying a negative value
+/// in timeout means an infinite timeout.  Specifying a timeout of zero
+/// causes `poll()` to return immediately, even if no file descriptors are
+/// ready.
 pub fn poll(fds: &mut [PollFd], timeout: libc::c_int) -> Result<libc::c_int> {
     let res = unsafe {
         libc::poll(fds.as_mut_ptr() as *mut libc::pollfd,
@@ -53,6 +120,13 @@ pub fn poll(fds: &mut [PollFd], timeout: libc::c_int) -> Result<libc::c_int> {
     Errno::result(res)
 }
 
+/// `ppoll()` allows an application to safely wait until either a file
+/// descriptor becomes ready or until a signal is caught.
+/// ([`poll(2)`](http://man7.org/linux/man-pages/man2/poll.2.html))
+///
+/// `ppoll` behaves like `poll`, but let you specify what signals may interrupt it
+/// with the `sigmask` argument.
+///
 #[cfg(any(target_os = "android", target_os = "dragonfly", target_os = "freebsd", target_os = "linux"))]
 pub fn ppoll(fds: &mut [PollFd], timeout: TimeSpec, sigmask: SigSet) -> Result<libc::c_int> {