1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
|
.. _design:
Design overview
===============
libuv is cross-platform support library which was originally written for NodeJS. It's designed
around the event-driven asynchronous I/O model.
The library provides much more than a simple abstraction over different I/O polling mechanisms:
'handles' and 'streams' provide a high level abstraction for sockets and other entities;
cross-platform file I/O and threading functionality is also provided, amongst other things.
Here is a diagram illustrating the different parts that compose libuv and what subsystem they
relate to:
.. image:: static/architecture.png
:scale: 75%
:align: center
Handles and requests
^^^^^^^^^^^^^^^^^^^^
libuv provides users with 2 abstractions to work with, in combination with the event loop:
handles and requests.
Handles represent long-lived objects capable of performing certain operations while active. Some examples:
- A prepare handle gets its callback called once every loop iteration when active.
- A TCP server handle that gets its connection callback called every time there is a new connection.
Requests represent (typically) short-lived operations. These operations can be performed over a
handle: write requests are used to write data on a handle; or standalone: getaddrinfo requests
don't need a handle they run directly on the loop.
The I/O loop
^^^^^^^^^^^^
The I/O (or event) loop is the central part of libuv. It establishes the content for all I/O
operations, and it's meant to be tied to a single thread. One can run multiple event loops
as long as each runs in a different thread. The libuv event loop (or any other API involving
the loop or handles, for that matter) **is not thread-safe** except where stated otherwise.
The event loop follows the rather usual single threaded asynchronous I/O approach: all (network)
I/O is performed on non-blocking sockets which are polled using the best mechanism available
on the given platform: epoll on Linux, kqueue on OSX and other BSDs, event ports on SunOS and IOCP
on Windows. As part of a loop iteration the loop will block waiting for I/O activity on sockets
which have been added to the poller and callbacks will be fired indicating socket conditions
(readable, writable hangup) so handles can read, write or perform the desired I/O operation.
In order to better understand how the event loop operates, the following diagram illustrates all
stages of a loop iteration:
.. image:: static/loop_iteration.png
:scale: 75%
:align: center
#. The loop concept of 'now' is updated. The event loop caches the current time at the start of
the event loop tick in order to reduce the number of time-related system calls.
#. If the loop is *alive* an iteration is started, otherwise the loop will exit immediately. So,
when is a loop considered to be *alive*? If a loop has active and ref'd handles, active
requests or closing handles it's considered to be *alive*.
#. Due timers are run. All active timers scheduled for a time before the loop's concept of *now*
get their callbacks called.
#. Pending callbacks are called. All I/O callbacks are called right after polling for I/O, for the
most part. There are cases, however, in which calling such a callback is deferred for the next
loop iteration. If the previous iteration deferred any I/O callback it will be run at this point.
#. Idle handle callbacks are called. Despite the unfortunate name, idle handles are run on every
loop iteration, if they are active.
#. Prepare handle callbacks are called. Prepare handles get their callbacks called right before
the loop will block for I/O.
#. Poll timeout is calculated. Before blocking for I/O the loop calculates for how long it should
block. These are the rules when calculating the timeout:
* If the loop was run with the ``UV_RUN_NOWAIT`` flag, the timeout is 0.
* If the loop is going to be stopped (:c:func:`uv_stop` was called), the timeout is 0.
* If there are no active handles or requests, the timeout is 0.
* If there are any idle handles active, the timeout is 0.
* If there are any handles pending to be closed, the timeout is 0.
* If none of the above cases matches, the timeout of the closest timer is taken, or
if there are no active timers, infinity.
#. The loop blocks for I/O. At this point the loop will block for I/O for the duration calculated
in the previous step. All I/O related handles that were monitoring a given file descriptor
for a read or write operation get their callbacks called at this point.
#. Check handle callbacks are called. Check handles get their callbacks called right after the
loop has blocked for I/O. Check handles are essentially the counterpart of prepare handles.
#. Close callbacks are called. If a handle was closed by calling :c:func:`uv_close` it will
get the close callback called.
#. Special case in case the loop was run with ``UV_RUN_ONCE``, as it implies forward progress.
It's possible that no I/O callbacks were fired after blocking for I/O, but some time has passed
so there might be timers which are due, those timers get their callbacks called.
#. Iteration ends. If the loop was run with ``UV_RUN_NOWAIT`` or ``UV_RUN_ONCE`` modes the
iteration ends and :c:func:`uv_run` will return. If the loop was run with ``UV_RUN_DEFAULT``
it will continue from the start if it's still *alive*, otherwise it will also end.
.. important::
libuv uses a thread pool to make asynchronous file I/O operations possible, but
network I/O is **always** performed in a single thread, each loop's thread.
.. note::
While the polling mechanism is different, libuv makes the execution model consistent
across Unix systems and Windows.
File I/O
^^^^^^^^
Unlike network I/O, there are no platform-specific file I/O primitives libuv could rely on,
so the current approach is to run blocking file I/O operations in a thread pool.
For a thorough explanation of the cross-platform file I/O landscape, checkout
`this post <http://blog.libtorrent.org/2012/10/asynchronous-disk-io/>`_.
libuv currently uses a global thread pool on which all loops can queue work. 3 types of
operations are currently run on this pool:
* File system operations
* DNS functions (getaddrinfo and getnameinfo)
* User specified code via :c:func:`uv_queue_work`
.. warning::
See the :c:ref:`threadpool` section for more details, but keep in mind the thread pool size
is quite limited.
|