646c8b39da
This bugfix implements correct way to calculate source sha1 hash, before it will changed, by propagation of unchanged sources. This commit will also include regenerated static files and *.hash files.
244 lines
8.4 KiB
Plaintext
Generated
244 lines
8.4 KiB
Plaintext
Generated
<!DOCTYPE html>
|
|
<html>
|
|
<head>
|
|
<meta charset="utf-8">
|
|
<title>Go by Example: Atomic Counters</title>
|
|
<link rel=stylesheet href="site.css">
|
|
</head>
|
|
<script>
|
|
onkeydown = (e) => {
|
|
|
|
if (e.key == "ArrowLeft") {
|
|
window.location.href = 'rate-limiting';
|
|
}
|
|
|
|
|
|
if (e.key == "ArrowRight") {
|
|
window.location.href = 'mutexes';
|
|
}
|
|
|
|
}
|
|
</script>
|
|
<body>
|
|
<div class="example" id="atomic-counters">
|
|
<h2><a href="./">Go by Example</a>: Atomic Counters</h2>
|
|
|
|
<table>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
<p>The primary mechanism for managing state in Go is
|
|
communication over channels. We saw this for example
|
|
with <a href="worker-pools">worker pools</a>. There are a few other
|
|
options for managing state though. Here we’ll
|
|
look at using the <code>sync/atomic</code> package for <em>atomic
|
|
counters</em> accessed by multiple goroutines.</p>
|
|
|
|
</td>
|
|
<td class="code empty leading">
|
|
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
|
|
</td>
|
|
<td class="code leading">
|
|
<a href="http://play.golang.org/p/j-14agntvEO"><img title="Run code" src="play.png" class="run" /></a><img title="Copy code" src="clipboard.png" class="copy" />
|
|
<div class="highlight"><pre><span class="kn">package</span> <span class="nx">main</span>
|
|
</pre></div>
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
|
|
</td>
|
|
<td class="code leading">
|
|
|
|
<div class="highlight"><pre><span class="kn">import</span> <span class="p">(</span>
|
|
<span class="s">"fmt"</span>
|
|
<span class="s">"sync"</span>
|
|
<span class="s">"sync/atomic"</span>
|
|
<span class="p">)</span>
|
|
</pre></div>
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
|
|
</td>
|
|
<td class="code leading">
|
|
|
|
<div class="highlight"><pre><span class="kd">func</span> <span class="nx">main</span><span class="p">()</span> <span class="p">{</span>
|
|
</pre></div>
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
<p>We’ll use an unsigned integer to represent our
|
|
(always-positive) counter.</p>
|
|
|
|
</td>
|
|
<td class="code leading">
|
|
|
|
<div class="highlight"><pre> <span class="kd">var</span> <span class="nx">ops</span> <span class="kt">uint64</span>
|
|
</pre></div>
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
<p>A WaitGroup will help us wait for all goroutines
|
|
to finish their work.</p>
|
|
|
|
</td>
|
|
<td class="code leading">
|
|
|
|
<div class="highlight"><pre> <span class="kd">var</span> <span class="nx">wg</span> <span class="nx">sync</span><span class="p">.</span><span class="nx">WaitGroup</span>
|
|
</pre></div>
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
<p>We’ll start 50 goroutines that each increment the
|
|
counter exactly 1000 times.</p>
|
|
|
|
</td>
|
|
<td class="code leading">
|
|
|
|
<div class="highlight"><pre> <span class="k">for</span> <span class="nx">i</span> <span class="o">:=</span> <span class="mi">0</span><span class="p">;</span> <span class="nx">i</span> <span class="p"><</span> <span class="mi">50</span><span class="p">;</span> <span class="nx">i</span><span class="o">++</span> <span class="p">{</span>
|
|
<span class="nx">wg</span><span class="p">.</span><span class="nx">Add</span><span class="p">(</span><span class="mi">1</span><span class="p">)</span>
|
|
</pre></div>
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
<p>To atomically increment the counter we
|
|
use <code>AddUint64</code>, giving it the memory
|
|
address of our <code>ops</code> counter with the
|
|
<code>&</code> syntax.</p>
|
|
|
|
</td>
|
|
<td class="code leading">
|
|
|
|
<div class="highlight"><pre> <span class="k">go</span> <span class="kd">func</span><span class="p">()</span> <span class="p">{</span>
|
|
<span class="k">for</span> <span class="nx">c</span> <span class="o">:=</span> <span class="mi">0</span><span class="p">;</span> <span class="nx">c</span> <span class="p"><</span> <span class="mi">1000</span><span class="p">;</span> <span class="nx">c</span><span class="o">++</span> <span class="p">{</span>
|
|
</pre></div>
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
|
|
</td>
|
|
<td class="code leading">
|
|
|
|
<div class="highlight"><pre> <span class="nx">atomic</span><span class="p">.</span><span class="nx">AddUint64</span><span class="p">(</span><span class="o">&</span><span class="nx">ops</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
|
|
<span class="p">}</span>
|
|
<span class="nx">wg</span><span class="p">.</span><span class="nx">Done</span><span class="p">()</span>
|
|
<span class="p">}()</span>
|
|
<span class="p">}</span>
|
|
</pre></div>
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
<p>Wait until all the goroutines are done.</p>
|
|
|
|
</td>
|
|
<td class="code leading">
|
|
|
|
<div class="highlight"><pre> <span class="nx">wg</span><span class="p">.</span><span class="nx">Wait</span><span class="p">()</span>
|
|
</pre></div>
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
<p>It’s safe to access <code>ops</code> now because we know
|
|
no other goroutine is writing to it. Reading
|
|
atomics safely while they are being updated is
|
|
also possible, using functions like
|
|
<code>atomic.LoadUint64</code>.</p>
|
|
|
|
</td>
|
|
<td class="code">
|
|
|
|
<div class="highlight"><pre> <span class="nx">fmt</span><span class="p">.</span><span class="nx">Println</span><span class="p">(</span><span class="s">"ops:"</span><span class="p">,</span> <span class="nx">ops</span><span class="p">)</span>
|
|
<span class="p">}</span>
|
|
</pre></div>
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
</table>
|
|
|
|
<table>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
<p>We expect to get exactly 50,000 operations. Had we
|
|
used the non-atomic <code>ops++</code> to increment the counter,
|
|
we’d likely get a different number, changing between
|
|
runs, because the goroutines would interfere with
|
|
each other. Moreover, we’d get data race failures
|
|
when running with the <code>-race</code> flag.</p>
|
|
|
|
</td>
|
|
<td class="code leading">
|
|
|
|
<div class="highlight"><pre><span class="gp">$</span> go run atomic-counters.go
|
|
<span class="go">ops: 50000</span>
|
|
</pre></div>
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
<tr>
|
|
<td class="docs">
|
|
<p>Next we’ll look at mutexes, another tool for managing
|
|
state.</p>
|
|
|
|
</td>
|
|
<td class="code empty">
|
|
|
|
|
|
</td>
|
|
</tr>
|
|
|
|
</table>
|
|
|
|
|
|
<p class="next">
|
|
Next example: <a href="mutexes">Mutexes</a>.
|
|
</p>
|
|
|
|
<p class="footer">
|
|
by <a href="https://markmcgranaghan.com">Mark McGranaghan</a> | <a href="https://github.com/mmcgrana/gobyexample/blob/master/examples/atomic-counters">source</a> | <a href="https://github.com/mmcgrana/gobyexample#license">license</a>
|
|
</p>
|
|
</div>
|
|
<script>
|
|
var codeLines = [];
|
|
codeLines.push('');codeLines.push('package main\u000A');codeLines.push('import (\u000A \"fmt\"\u000A \"sync\"\u000A \"sync/atomic\"\u000A)\u000A');codeLines.push('func main() {\u000A');codeLines.push(' var ops uint64\u000A');codeLines.push(' var wg sync.WaitGroup\u000A');codeLines.push(' for i := 0; i \x3C 50; i++ {\u000A wg.Add(1)\u000A');codeLines.push(' go func() {\u000A for c := 0; c \x3C 1000; c++ {\u000A');codeLines.push(' atomic.AddUint64(&ops, 1)\u000A }\u000A wg.Done()\u000A }()\u000A }\u000A');codeLines.push(' wg.Wait()\u000A');codeLines.push(' fmt.Println(\"ops:\", ops)\u000A}\u000A');codeLines.push('');codeLines.push('');
|
|
</script>
|
|
<script src="site.js" async></script>
|
|
</body>
|
|
</html>
|