blob: d46d7b16b734e9c599c507a8dc038d3a66fe306c [file] [log] [blame]
<!DOCTYPE html><html lang="en"><head><meta charset="utf-8"><meta name="viewport" content="width=device-width, initial-scale=1.0"><meta name="generator" content="rustdoc"><meta name="description" content="Source of the Rust file `/root/.cargo/registry/src/github.com-1ecc6299db9ec823/crc32fast-1.3.2/src/specialized/pclmulqdq.rs`."><meta name="keywords" content="rust, rustlang, rust-lang"><title>pclmulqdq.rs - source</title><link rel="preload" as="font" type="font/woff2" crossorigin href="../../../SourceSerif4-Regular.ttf.woff2"><link rel="preload" as="font" type="font/woff2" crossorigin href="../../../FiraSans-Regular.woff2"><link rel="preload" as="font" type="font/woff2" crossorigin href="../../../FiraSans-Medium.woff2"><link rel="preload" as="font" type="font/woff2" crossorigin href="../../../SourceCodePro-Regular.ttf.woff2"><link rel="preload" as="font" type="font/woff2" crossorigin href="../../../SourceSerif4-Bold.ttf.woff2"><link rel="preload" as="font" type="font/woff2" crossorigin href="../../../SourceCodePro-Semibold.ttf.woff2"><link rel="stylesheet" href="../../../normalize.css"><link rel="stylesheet" href="../../../rustdoc.css" id="mainThemeStyle"><link rel="stylesheet" href="../../../ayu.css" disabled><link rel="stylesheet" href="../../../dark.css" disabled><link rel="stylesheet" href="../../../light.css" id="themeStyle"><script id="default-settings" ></script><script src="../../../storage.js"></script><script defer src="../../../source-script.js"></script><script defer src="../../../source-files.js"></script><script defer src="../../../main.js"></script><noscript><link rel="stylesheet" href="../../../noscript.css"></noscript><link rel="alternate icon" type="image/png" href="../../../favicon-16x16.png"><link rel="alternate icon" type="image/png" href="../../../favicon-32x32.png"><link rel="icon" type="image/svg+xml" href="../../../favicon.svg"></head><body class="rustdoc source"><!--[if lte IE 11]><div class="warning">This old browser is unsupported and will most likely display funky things.</div><![endif]--><nav class="sidebar"><a class="sidebar-logo" href="../../../crc32fast/index.html"><div class="logo-container"><img class="rust-logo" src="../../../rust-logo.svg" alt="logo"></div></a></nav><main><div class="width-limiter"><nav class="sub"><a class="sub-logo-container" href="../../../crc32fast/index.html"><img class="rust-logo" src="../../../rust-logo.svg" alt="logo"></a><form class="search-form"><div class="search-container"><span></span><input class="search-input" name="search" autocomplete="off" spellcheck="false" placeholder="Click or press ‘S’ to search, ‘?’ for more options…" type="search"><div id="help-button" title="help" tabindex="-1"><a href="../../../help.html">?</a></div><div id="settings-menu" tabindex="-1"><a href="../../../settings.html" title="settings"><img width="22" height="22" alt="Change settings" src="../../../wheel.svg"></a></div></div></form></nav><section id="main-content" class="content"><div class="example-wrap"><pre class="src-line-numbers"><span id="1">1</span>
<span id="2">2</span>
<span id="3">3</span>
<span id="4">4</span>
<span id="5">5</span>
<span id="6">6</span>
<span id="7">7</span>
<span id="8">8</span>
<span id="9">9</span>
<span id="10">10</span>
<span id="11">11</span>
<span id="12">12</span>
<span id="13">13</span>
<span id="14">14</span>
<span id="15">15</span>
<span id="16">16</span>
<span id="17">17</span>
<span id="18">18</span>
<span id="19">19</span>
<span id="20">20</span>
<span id="21">21</span>
<span id="22">22</span>
<span id="23">23</span>
<span id="24">24</span>
<span id="25">25</span>
<span id="26">26</span>
<span id="27">27</span>
<span id="28">28</span>
<span id="29">29</span>
<span id="30">30</span>
<span id="31">31</span>
<span id="32">32</span>
<span id="33">33</span>
<span id="34">34</span>
<span id="35">35</span>
<span id="36">36</span>
<span id="37">37</span>
<span id="38">38</span>
<span id="39">39</span>
<span id="40">40</span>
<span id="41">41</span>
<span id="42">42</span>
<span id="43">43</span>
<span id="44">44</span>
<span id="45">45</span>
<span id="46">46</span>
<span id="47">47</span>
<span id="48">48</span>
<span id="49">49</span>
<span id="50">50</span>
<span id="51">51</span>
<span id="52">52</span>
<span id="53">53</span>
<span id="54">54</span>
<span id="55">55</span>
<span id="56">56</span>
<span id="57">57</span>
<span id="58">58</span>
<span id="59">59</span>
<span id="60">60</span>
<span id="61">61</span>
<span id="62">62</span>
<span id="63">63</span>
<span id="64">64</span>
<span id="65">65</span>
<span id="66">66</span>
<span id="67">67</span>
<span id="68">68</span>
<span id="69">69</span>
<span id="70">70</span>
<span id="71">71</span>
<span id="72">72</span>
<span id="73">73</span>
<span id="74">74</span>
<span id="75">75</span>
<span id="76">76</span>
<span id="77">77</span>
<span id="78">78</span>
<span id="79">79</span>
<span id="80">80</span>
<span id="81">81</span>
<span id="82">82</span>
<span id="83">83</span>
<span id="84">84</span>
<span id="85">85</span>
<span id="86">86</span>
<span id="87">87</span>
<span id="88">88</span>
<span id="89">89</span>
<span id="90">90</span>
<span id="91">91</span>
<span id="92">92</span>
<span id="93">93</span>
<span id="94">94</span>
<span id="95">95</span>
<span id="96">96</span>
<span id="97">97</span>
<span id="98">98</span>
<span id="99">99</span>
<span id="100">100</span>
<span id="101">101</span>
<span id="102">102</span>
<span id="103">103</span>
<span id="104">104</span>
<span id="105">105</span>
<span id="106">106</span>
<span id="107">107</span>
<span id="108">108</span>
<span id="109">109</span>
<span id="110">110</span>
<span id="111">111</span>
<span id="112">112</span>
<span id="113">113</span>
<span id="114">114</span>
<span id="115">115</span>
<span id="116">116</span>
<span id="117">117</span>
<span id="118">118</span>
<span id="119">119</span>
<span id="120">120</span>
<span id="121">121</span>
<span id="122">122</span>
<span id="123">123</span>
<span id="124">124</span>
<span id="125">125</span>
<span id="126">126</span>
<span id="127">127</span>
<span id="128">128</span>
<span id="129">129</span>
<span id="130">130</span>
<span id="131">131</span>
<span id="132">132</span>
<span id="133">133</span>
<span id="134">134</span>
<span id="135">135</span>
<span id="136">136</span>
<span id="137">137</span>
<span id="138">138</span>
<span id="139">139</span>
<span id="140">140</span>
<span id="141">141</span>
<span id="142">142</span>
<span id="143">143</span>
<span id="144">144</span>
<span id="145">145</span>
<span id="146">146</span>
<span id="147">147</span>
<span id="148">148</span>
<span id="149">149</span>
<span id="150">150</span>
<span id="151">151</span>
<span id="152">152</span>
<span id="153">153</span>
<span id="154">154</span>
<span id="155">155</span>
<span id="156">156</span>
<span id="157">157</span>
<span id="158">158</span>
<span id="159">159</span>
<span id="160">160</span>
<span id="161">161</span>
<span id="162">162</span>
<span id="163">163</span>
<span id="164">164</span>
<span id="165">165</span>
<span id="166">166</span>
<span id="167">167</span>
<span id="168">168</span>
<span id="169">169</span>
<span id="170">170</span>
<span id="171">171</span>
<span id="172">172</span>
<span id="173">173</span>
<span id="174">174</span>
<span id="175">175</span>
<span id="176">176</span>
<span id="177">177</span>
<span id="178">178</span>
<span id="179">179</span>
<span id="180">180</span>
<span id="181">181</span>
<span id="182">182</span>
<span id="183">183</span>
<span id="184">184</span>
<span id="185">185</span>
<span id="186">186</span>
<span id="187">187</span>
<span id="188">188</span>
<span id="189">189</span>
<span id="190">190</span>
<span id="191">191</span>
<span id="192">192</span>
<span id="193">193</span>
<span id="194">194</span>
<span id="195">195</span>
<span id="196">196</span>
<span id="197">197</span>
<span id="198">198</span>
<span id="199">199</span>
<span id="200">200</span>
<span id="201">201</span>
<span id="202">202</span>
<span id="203">203</span>
<span id="204">204</span>
<span id="205">205</span>
<span id="206">206</span>
<span id="207">207</span>
<span id="208">208</span>
<span id="209">209</span>
<span id="210">210</span>
<span id="211">211</span>
<span id="212">212</span>
<span id="213">213</span>
<span id="214">214</span>
<span id="215">215</span>
<span id="216">216</span>
<span id="217">217</span>
<span id="218">218</span>
<span id="219">219</span>
<span id="220">220</span>
<span id="221">221</span>
<span id="222">222</span>
<span id="223">223</span>
<span id="224">224</span>
<span id="225">225</span>
</pre><pre class="rust"><code><span class="attribute">#[cfg(target_arch = <span class="string">&quot;x86&quot;</span>)]
</span><span class="kw">use </span>core::arch::x86 <span class="kw">as </span>arch;
<span class="attribute">#[cfg(target_arch = <span class="string">&quot;x86_64&quot;</span>)]
</span><span class="kw">use </span>core::arch::x86_64 <span class="kw">as </span>arch;
<span class="attribute">#[derive(Clone)]
</span><span class="kw">pub struct </span>State {
state: u32,
}
<span class="kw">impl </span>State {
<span class="attribute">#[cfg(not(feature = <span class="string">&quot;std&quot;</span>))]
</span><span class="kw">pub fn </span>new(state: u32) -&gt; <span class="prelude-ty">Option</span>&lt;<span class="self">Self</span>&gt; {
<span class="kw">if </span><span class="macro">cfg!</span>(target_feature = <span class="string">&quot;pclmulqdq&quot;</span>)
&amp;&amp; <span class="macro">cfg!</span>(target_feature = <span class="string">&quot;sse2&quot;</span>)
&amp;&amp; <span class="macro">cfg!</span>(target_feature = <span class="string">&quot;sse4.1&quot;</span>)
{
<span class="comment">// SAFETY: The conditions above ensure that all
// required instructions are supported by the CPU.
</span><span class="prelude-val">Some</span>(<span class="self">Self </span>{ state })
} <span class="kw">else </span>{
<span class="prelude-val">None
</span>}
}
<span class="attribute">#[cfg(feature = <span class="string">&quot;std&quot;</span>)]
</span><span class="kw">pub fn </span>new(state: u32) -&gt; <span class="prelude-ty">Option</span>&lt;<span class="self">Self</span>&gt; {
<span class="kw">if </span><span class="macro">is_x86_feature_detected!</span>(<span class="string">&quot;pclmulqdq&quot;</span>)
&amp;&amp; <span class="macro">is_x86_feature_detected!</span>(<span class="string">&quot;sse2&quot;</span>)
&amp;&amp; <span class="macro">is_x86_feature_detected!</span>(<span class="string">&quot;sse4.1&quot;</span>)
{
<span class="comment">// SAFETY: The conditions above ensure that all
// required instructions are supported by the CPU.
</span><span class="prelude-val">Some</span>(<span class="self">Self </span>{ state })
} <span class="kw">else </span>{
<span class="prelude-val">None
</span>}
}
<span class="kw">pub fn </span>update(<span class="kw-2">&amp;mut </span><span class="self">self</span>, buf: <span class="kw-2">&amp;</span>[u8]) {
<span class="comment">// SAFETY: The `State::new` constructor ensures that all
// required instructions are supported by the CPU.
</span><span class="self">self</span>.state = <span class="kw">unsafe </span>{ calculate(<span class="self">self</span>.state, buf) }
}
<span class="kw">pub fn </span>finalize(<span class="self">self</span>) -&gt; u32 {
<span class="self">self</span>.state
}
<span class="kw">pub fn </span>reset(<span class="kw-2">&amp;mut </span><span class="self">self</span>) {
<span class="self">self</span>.state = <span class="number">0</span>;
}
<span class="kw">pub fn </span>combine(<span class="kw-2">&amp;mut </span><span class="self">self</span>, other: u32, amount: u64) {
<span class="self">self</span>.state = ::combine::combine(<span class="self">self</span>.state, other, amount);
}
}
<span class="kw">const </span>K1: i64 = <span class="number">0x154442bd4</span>;
<span class="kw">const </span>K2: i64 = <span class="number">0x1c6e41596</span>;
<span class="kw">const </span>K3: i64 = <span class="number">0x1751997d0</span>;
<span class="kw">const </span>K4: i64 = <span class="number">0x0ccaa009e</span>;
<span class="kw">const </span>K5: i64 = <span class="number">0x163cd6124</span>;
<span class="kw">const </span>K6: i64 = <span class="number">0x1db710640</span>;
<span class="kw">const </span>P_X: i64 = <span class="number">0x1DB710641</span>;
<span class="kw">const </span>U_PRIME: i64 = <span class="number">0x1F7011641</span>;
<span class="attribute">#[cfg(feature = <span class="string">&quot;std&quot;</span>)]
</span><span class="kw">unsafe fn </span>debug(s: <span class="kw-2">&amp;</span>str, a: arch::__m128i) -&gt; arch::__m128i {
<span class="kw">if </span><span class="bool-val">false </span>{
<span class="kw">union </span>A {
a: arch::__m128i,
b: [u8; <span class="number">16</span>],
}
<span class="kw">let </span>x = A { a }.b;
<span class="macro">print!</span>(<span class="string">&quot; {:20} | &quot;</span>, s);
<span class="kw">for </span>x <span class="kw">in </span>x.iter() {
<span class="macro">print!</span>(<span class="string">&quot;{:02x} &quot;</span>, x);
}
<span class="macro">println!</span>();
}
<span class="kw">return </span>a;
}
<span class="attribute">#[cfg(not(feature = <span class="string">&quot;std&quot;</span>))]
</span><span class="kw">unsafe fn </span>debug(_s: <span class="kw-2">&amp;</span>str, a: arch::__m128i) -&gt; arch::__m128i {
a
}
<span class="attribute">#[target_feature(enable = <span class="string">&quot;pclmulqdq&quot;</span>, enable = <span class="string">&quot;sse2&quot;</span>, enable = <span class="string">&quot;sse4.1&quot;</span>)]
</span><span class="kw">unsafe fn </span>calculate(crc: u32, <span class="kw-2">mut </span>data: <span class="kw-2">&amp;</span>[u8]) -&gt; u32 {
<span class="comment">// In theory we can accelerate smaller chunks too, but for now just rely on
// the fallback implementation as it&#39;s too much hassle and doesn&#39;t seem too
// beneficial.
</span><span class="kw">if </span>data.len() &lt; <span class="number">128 </span>{
<span class="kw">return </span>::baseline::update_fast_16(crc, data);
}
<span class="comment">// Step 1: fold by 4 loop
</span><span class="kw">let </span><span class="kw-2">mut </span>x3 = get(<span class="kw-2">&amp;mut </span>data);
<span class="kw">let </span><span class="kw-2">mut </span>x2 = get(<span class="kw-2">&amp;mut </span>data);
<span class="kw">let </span><span class="kw-2">mut </span>x1 = get(<span class="kw-2">&amp;mut </span>data);
<span class="kw">let </span><span class="kw-2">mut </span>x0 = get(<span class="kw-2">&amp;mut </span>data);
<span class="comment">// fold in our initial value, part of the incremental crc checksum
</span>x3 = arch::_mm_xor_si128(x3, arch::_mm_cvtsi32_si128(!crc <span class="kw">as </span>i32));
<span class="kw">let </span>k1k2 = arch::_mm_set_epi64x(K2, K1);
<span class="kw">while </span>data.len() &gt;= <span class="number">64 </span>{
x3 = reduce128(x3, get(<span class="kw-2">&amp;mut </span>data), k1k2);
x2 = reduce128(x2, get(<span class="kw-2">&amp;mut </span>data), k1k2);
x1 = reduce128(x1, get(<span class="kw-2">&amp;mut </span>data), k1k2);
x0 = reduce128(x0, get(<span class="kw-2">&amp;mut </span>data), k1k2);
}
<span class="kw">let </span>k3k4 = arch::_mm_set_epi64x(K4, K3);
<span class="kw">let </span><span class="kw-2">mut </span>x = reduce128(x3, x2, k3k4);
x = reduce128(x, x1, k3k4);
x = reduce128(x, x0, k3k4);
<span class="comment">// Step 2: fold by 1 loop
</span><span class="kw">while </span>data.len() &gt;= <span class="number">16 </span>{
x = reduce128(x, get(<span class="kw-2">&amp;mut </span>data), k3k4);
}
debug(<span class="string">&quot;128 &gt; 64 init&quot;</span>, x);
<span class="comment">// Perform step 3, reduction from 128 bits to 64 bits. This is
// significantly different from the paper and basically doesn&#39;t follow it
// at all. It&#39;s not really clear why, but implementations of this algorithm
// in Chrome/Linux diverge in the same way. It is beyond me why this is
// different than the paper, maybe the paper has like errata or something?
// Unclear.
//
// It&#39;s also not clear to me what&#39;s actually happening here and/or why, but
// algebraically what&#39;s happening is:
//
// x = (x[0:63] • K4) ^ x[64:127] // 96 bit result
// x = ((x[0:31] as u64) • K5) ^ x[32:95] // 64 bit result
//
// It&#39;s... not clear to me what&#39;s going on here. The paper itself is pretty
// vague on this part but definitely uses different constants at least.
// It&#39;s not clear to me, reading the paper, where the xor operations are
// happening or why things are shifting around. This implementation...
// appears to work though!
</span>drop(K6);
<span class="kw">let </span>x = arch::_mm_xor_si128(
arch::_mm_clmulepi64_si128(x, k3k4, <span class="number">0x10</span>),
arch::_mm_srli_si128(x, <span class="number">8</span>),
);
<span class="kw">let </span>x = arch::_mm_xor_si128(
arch::_mm_clmulepi64_si128(
arch::_mm_and_si128(x, arch::_mm_set_epi32(<span class="number">0</span>, <span class="number">0</span>, <span class="number">0</span>, !<span class="number">0</span>)),
arch::_mm_set_epi64x(<span class="number">0</span>, K5),
<span class="number">0x00</span>,
),
arch::_mm_srli_si128(x, <span class="number">4</span>),
);
debug(<span class="string">&quot;128 &gt; 64 xx&quot;</span>, x);
<span class="comment">// Perform a Barrett reduction from our now 64 bits to 32 bits. The
// algorithm for this is described at the end of the paper, and note that
// this also implements the &quot;bit reflected input&quot; variant.
</span><span class="kw">let </span>pu = arch::_mm_set_epi64x(U_PRIME, P_X);
<span class="comment">// T1(x) = ⌊(R(x) % x^32)⌋ • μ
</span><span class="kw">let </span>t1 = arch::_mm_clmulepi64_si128(
arch::_mm_and_si128(x, arch::_mm_set_epi32(<span class="number">0</span>, <span class="number">0</span>, <span class="number">0</span>, !<span class="number">0</span>)),
pu,
<span class="number">0x10</span>,
);
<span class="comment">// T2(x) = ⌊(T1(x) % x^32)⌋ • P(x)
</span><span class="kw">let </span>t2 = arch::_mm_clmulepi64_si128(
arch::_mm_and_si128(t1, arch::_mm_set_epi32(<span class="number">0</span>, <span class="number">0</span>, <span class="number">0</span>, !<span class="number">0</span>)),
pu,
<span class="number">0x00</span>,
);
<span class="comment">// We&#39;re doing the bit-reflected variant, so get the upper 32-bits of the
// 64-bit result instead of the lower 32-bits.
//
// C(x) = R(x) ^ T2(x) / x^32
</span><span class="kw">let </span>c = arch::_mm_extract_epi32(arch::_mm_xor_si128(x, t2), <span class="number">1</span>) <span class="kw">as </span>u32;
<span class="kw">if </span>!data.is_empty() {
::baseline::update_fast_16(!c, data)
} <span class="kw">else </span>{
!c
}
}
<span class="kw">unsafe fn </span>reduce128(a: arch::__m128i, b: arch::__m128i, keys: arch::__m128i) -&gt; arch::__m128i {
<span class="kw">let </span>t1 = arch::_mm_clmulepi64_si128(a, keys, <span class="number">0x00</span>);
<span class="kw">let </span>t2 = arch::_mm_clmulepi64_si128(a, keys, <span class="number">0x11</span>);
arch::_mm_xor_si128(arch::_mm_xor_si128(b, t1), t2)
}
<span class="kw">unsafe fn </span>get(a: <span class="kw-2">&amp;mut &amp;</span>[u8]) -&gt; arch::__m128i {
<span class="macro">debug_assert!</span>(a.len() &gt;= <span class="number">16</span>);
<span class="kw">let </span>r = arch::_mm_loadu_si128(a.as_ptr() <span class="kw">as </span><span class="kw-2">*const </span>arch::__m128i);
<span class="kw-2">*</span>a = <span class="kw-2">&amp;</span>a[<span class="number">16</span>..];
<span class="kw">return </span>r;
}
<span class="attribute">#[cfg(test)]
</span><span class="kw">mod </span>test {
<span class="macro">quickcheck! </span>{
<span class="kw">fn </span>check_against_baseline(init: u32, chunks: Vec&lt;(Vec&lt;u8&gt;, usize)&gt;) -&gt; bool {
<span class="kw">let </span><span class="kw-2">mut </span>baseline = <span class="kw">super</span>::<span class="kw">super</span>::<span class="kw">super</span>::baseline::State::new(init);
<span class="kw">let </span><span class="kw-2">mut </span>pclmulqdq = <span class="kw">super</span>::State::new(init).expect(<span class="string">&quot;not supported&quot;</span>);
<span class="kw">for </span>(chunk, <span class="kw-2">mut </span>offset) <span class="kw">in </span>chunks {
<span class="comment">// simulate random alignments by offsetting the slice by up to 15 bytes
</span>offset &amp;= <span class="number">0xF</span>;
<span class="kw">if </span>chunk.len() &lt;= offset {
baseline.update(<span class="kw-2">&amp;</span>chunk);
pclmulqdq.update(<span class="kw-2">&amp;</span>chunk);
} <span class="kw">else </span>{
baseline.update(<span class="kw-2">&amp;</span>chunk[offset..]);
pclmulqdq.update(<span class="kw-2">&amp;</span>chunk[offset..]);
}
}
pclmulqdq.finalize() == baseline.finalize()
}
}
}
</code></pre></div>
</section></div></main><div id="rustdoc-vars" data-root-path="../../../" data-current-crate="crc32fast" data-themes="ayu,dark,light" data-resource-suffix="" data-rustdoc-version="1.66.0-nightly (5c8bff74b 2022-10-21)" ></div></body></html>