You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Trac3r-rust/doc/matrixmultiply/index.html

52 lines
7.1 KiB

This file contains ambiguous Unicode characters!

This file contains ambiguous Unicode characters that may be confused with others in your current locale. If your use case is intentional and legitimate, you can safely ignore this warning. Use the Escape button to highlight these characters.

<!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="API documentation for the Rust `matrixmultiply` crate."><meta name="keywords" content="rust, rustlang, rust-lang, matrixmultiply"><title>matrixmultiply - Rust</title><link rel="stylesheet" type="text/css" href="../normalize.css"><link rel="stylesheet" type="text/css" href="../rustdoc.css" id="mainThemeStyle"><link rel="stylesheet" type="text/css" href="../dark.css"><link rel="stylesheet" type="text/css" href="../light.css" id="themeStyle"><script src="../storage.js"></script><noscript><link rel="stylesheet" href="../noscript.css"></noscript><link rel="shortcut icon" href="../favicon.ico"><style type="text/css">#crate-search{background-image:url("../down-arrow.svg");}</style></head><body class="rustdoc mod"><!--[if lte IE 8]><div class="warning">This old browser is unsupported and will most likely display funky things.</div><![endif]--><nav class="sidebar"><div class="sidebar-menu">&#9776;</div><a href='../matrixmultiply/index.html'><div class='logo-container'><img src='../rust-logo.png' alt='logo'></div></a><p class='location'>Crate matrixmultiply</p><div class="sidebar-elems"><a id='all-types' href='all.html'><p>See all matrixmultiply's items</p></a><div class="block items"><ul><li><a href="#functions">Functions</a></li></ul></div><p class='location'></p><script>window.sidebarCurrent = {name: 'matrixmultiply', ty: 'mod', relpath: '../'};</script></div></nav><div class="theme-picker"><button id="theme-picker" aria-label="Pick another theme!"><img src="../brush.svg" width="18" alt="Pick another theme!"></button><div id="theme-choices"></div></div><script src="../theme.js"></script><nav class="sub"><form class="search-form js-only"><div class="search-container"><div><select id="crate-search"><option value="All crates">All crates</option></select><input class="search-input" name="search" autocomplete="off" spellcheck="false" placeholder="Click or press S to search, ? for more options…" type="search"></div><a id="settings-menu" href="../settings.html"><img src="../wheel.svg" width="18" alt="Change settings"></a></div></form></nav><section id="main" class="content"><h1 class='fqn'><span class='out-of-band'><span id='render-detail'><a id="toggle-all-docs" href="javascript:void(0)" title="collapse all docs">[<span class='inner'>&#x2212;</span>]</a></span><a class='srclink' href='../src/matrixmultiply/lib.rs.html#8-86' title='goto source code'>[src]</a></span><span class='in-band'>Crate <a class="mod" href=''>matrixmultiply</a></span></h1><div class='docblock'><p>General matrix multiplication for f32, f64 matrices. Operates on matrices
with general layout (they can use arbitrary row and column stride).</p>
<p>This crate uses the same macro/microkernel approach to matrix multiplication as
the <a href="https://github.com/flame/blis">BLIS</a> project.</p>
<p>We presently provide a few good microkernels, portable and for x86-64, and
only one operation: the general matrix-matrix multiplication (“gemm”).</p>
<h2 id="matrix-representation" class="section-header"><a href="#matrix-representation">Matrix Representation</a></h2>
<p><strong>matrixmultiply</strong> supports matrices with general stride, so a matrix
is passed using a pointer and four integers:</p>
<ul>
<li><code>a: *const f32</code>, pointer to the first element in the matrix</li>
<li><code>m: usize</code>, number of rows</li>
<li><code>k: usize</code>, number of columns</li>
<li><code>rsa: isize</code>, row stride</li>
<li><code>csa: isize</code>, column stride</li>
</ul>
<p>In this example, A is a m by k matrix. <code>a</code> is a pointer to the element at
index <em>0, 0</em>.</p>
<p>The <em>row stride</em> is the pointer offset (in number of elements) to the
element on the next row. Its the distance from element <em>i, j</em> to <em>i + 1,
j</em>.</p>
<p>The <em>column stride</em> is the pointer offset (in number of elements) to the
element in the next column. Its the distance from element <em>i, j</em> to <em>i,
j + 1</em>.</p>
<p>For example for a contiguous matrix, row major strides are <em>rsa=k,
csa=1</em> and column major strides are <em>rsa=1, csa=m</em>.</p>
<p>Stides can be negative or even zero, but for a mutable matrix elements
may not alias each other.</p>
<h2 id="portability-and-performance" class="section-header"><a href="#portability-and-performance">Portability and Performance</a></h2>
<ul>
<li>
<p>The default kernels are written in portable Rust and available
on all targets. These may depend on autovectorization to perform well.</p>
</li>
<li>
<p><em>x86</em> and <em>x86-64</em> features can be detected at runtime by default or
compile time (if enabled), and the crate following kernel variants are
implemented:</p>
<ul>
<li><code>fma</code></li>
<li><code>avx</code></li>
<li><code>sse2</code></li>
</ul>
</li>
</ul>
<h2 id="other-notes" class="section-header"><a href="#other-notes">Other Notes</a></h2>
<p>The functions in this crate are thread safe, as long as the destination
matrix is distinct.</p>
</div><h2 id='functions' class='section-header'><a href="#functions">Functions</a></h2>
<table><tr class='module-item'><td><a class="fn" href="fn.dgemm.html" title='matrixmultiply::dgemm fn'>dgemm</a><a title='unsafe function' href='#'><sup></sup></a></td><td class='docblock-short'><p>General matrix multiplication (f64)</p>
</td></tr><tr class='module-item'><td><a class="fn" href="fn.sgemm.html" title='matrixmultiply::sgemm fn'>sgemm</a><a title='unsafe function' href='#'><sup></sup></a></td><td class='docblock-short'><p>General matrix multiplication (f32)</p>
</td></tr></table></section><section id="search" class="content hidden"></section><section class="footer"></section><aside id="help" class="hidden"><div><h1 class="hidden">Help</h1><div class="shortcuts"><h2>Keyboard Shortcuts</h2><dl><dt><kbd>?</kbd></dt><dd>Show this help dialog</dd><dt><kbd>S</kbd></dt><dd>Focus the search field</dd><dt><kbd></kbd></dt><dd>Move up in search results</dd><dt><kbd></kbd></dt><dd>Move down in search results</dd><dt><kbd></kbd></dt><dd>Switch tab</dd><dt><kbd>&#9166;</kbd></dt><dd>Go to active search result</dd><dt><kbd>+</kbd></dt><dd>Expand all sections</dd><dt><kbd>-</kbd></dt><dd>Collapse all sections</dd></dl></div><div class="infos"><h2>Search Tricks</h2><p>Prefix searches with a type followed by a colon (e.g., <code>fn:</code>) to restrict the search to a given type.</p><p>Accepted types are: <code>fn</code>, <code>mod</code>, <code>struct</code>, <code>enum</code>, <code>trait</code>, <code>type</code>, <code>macro</code>, and <code>const</code>.</p><p>Search functions by type signature (e.g., <code>vec -> usize</code> or <code>* -> vec</code>)</p><p>Search multiple things at once by splitting your query with comma (e.g., <code>str,u8</code> or <code>String,struct:Vec,test</code>)</p></div></div></aside><script>window.rootPath = "../";window.currentCrate = "matrixmultiply";</script><script src="../aliases.js"></script><script src="../main.js"></script><script defer src="../search-index.js"></script></body></html>