diff options
Diffstat (limited to '2024/05/01/Rust-虚表布局规则介绍/index.html')
| -rw-r--r-- | 2024/05/01/Rust-虚表布局规则介绍/index.html | 309 |
1 files changed, 0 insertions, 309 deletions
diff --git a/2024/05/01/Rust-虚表布局规则介绍/index.html b/2024/05/01/Rust-虚表布局规则介绍/index.html deleted file mode 100644 index bb83b67c..00000000 --- a/2024/05/01/Rust-虚表布局规则介绍/index.html +++ /dev/null @@ -1,309 +0,0 @@ -<!DOCTYPE html> -<html lang="en"> - <head> - <meta charset="UTF-8"> -<meta name="viewport" - content="width=device-width, initial-scale=1.0, maximum-scale=1.0, minimum-scale=1.0"> -<meta http-equiv="X-UA-Compatible" content="ie=edge"> - - <meta name="author" content="韩暮秋"> - - - <meta name="subtitle" content="暮秋小屋"> - - - <meta name="description" content="这里是暮秋小屋,思念和灵感的寄存处"> - - - <meta name="keywords" content="韩暮秋,MuqiuHan,'Muqiu Han', 'muqiu han', muqiuhan"> - - - - -<title>Rust 虚表布局规则介绍 | 暮秋小屋</title> - - - - <link rel="icon" href="/favicon.ico"> - - - -<style> - @import url('https://fonts.googleapis.com/css2?family=Inter:wght@300;400;500;600;700&family=Noto+Sans+SC:wght@300;400;500;700&family=Roboto+Mono&display=swap'); -</style> - - - - <!-- stylesheets list from _config.yml --> - - <link rel="stylesheet" href="/css/style.css"> - - - - - - <!-- scripts list from _config.yml --> - - <script src="/js/frame.js"></script> - - - - - - <script src="https://polyfill.io/v3/polyfill.min.js?features=es6"></script> - <script id="MathJax-script" async src="https://cdn.jsdelivr.net/npm/mathjax@3/es5/tex-mml-chtml.js"></script> - - - - - - - - <meta name="generator" content="Hexo 6.3.0"></head> - <body> - <div class="mask-border"> - </div> - - <div class="wrapper"> - - <div class="header"> - <div class="flex-container"> - <div class="header-inner"> - <div class="site-brand-container"> - <a href="/"> - - 暮秋小屋 - - </a> - </div> - <div id="menu-btn" class="menu-btn" onclick="toggleMenu()"> - Menu - </div> - <nav class="site-nav"> - <ul class="menu-list"> - - - <li class="menu-item"> - <a href="/">主页</a> - </li> - - - - <li class="menu-item"> - <a href="/categories/gallery/">日记本</a> - </li> - - - - <li class="menu-item"> - <a href="/tags/Medicine/">医学</a> - </li> - - - - <li class="menu-item"> - <a href="/tags/Technique/">计算机科学</a> - </li> - - - - <li class="menu-item"> - <a href="/tags/Life/">生活</a> - </li> - - - - <li class="menu-item"> - <a href="/about">关于</a> - </li> - - - - <li class="menu-item search-btn"> - <a href="#">Search</a> - </li> - - </ul> - </nav> - </div> - </div> -</div> - - - <div class="main"> - <div class="flex-container"> - <article id="post"> - - - <div class="post-head"> - <div class="post-info"> - <div class="tag-list"> - - - <span class="post-tag"> - <a href="/tags/Technique/"> - Technique - </a> - </span> - - <span class="post-tag"> - <a href="/tags/Rust/"> - Rust - </a> - </span> - - - </div> - <div class="post-title"> - - - Rust 虚表布局规则介绍 - - - </div> - <span class="post-date"> - May 1, 2024 - </span> - </div> - <div class="post-img"> - - <div class="h-line-primary"></div> - - </div> -</div> - <div class="post-content"> - <p>在 Rust 中,一个指向未知大小对象(!Sized)的引用或指针被实现为一个由两个 usize 大小的域构成的胖指针。这两个域中,其中一个域保存了被引用或被指向的对象的地址,另一个域保存了一个名为 metadata 的数据。对于 slice 的引用或指针来说,其 metadata 为 slice 的长度。对于 trait object 的引用或指针来说,其 metadata 为虚表(vtable)地址。与 C++ 虚表类似,Rust 虚表的存在使得诸多动态语言特性得以实现,例如动态派发(dynamic dispatch)、向上转换(upcasting)、向下转换(downcast)等。本文将对 Rust 中虚表的布局规则进行简要介绍,并在此过程中对 Rust 中若干动态特性的实现方法进行简要介绍。</p> -<blockquote> -<p>注意:Rust 虚表及其结构属于 Rust 语言的内部实现细节,不保证稳定性。本文所介绍的虚表布局仅反映本文创作时最新的 Rust 虚表结构[1],在将来 Rust 虚表结构可能会发生变化。一个 Rust 程序的正确性不应该以任何方式依赖于 Rust 虚表的结构。</p> -</blockquote> -<h2 id="基本结构"><a href="#基本结构" class="headerlink" title="基本结构"></a>基本结构</h2><p>Rust 程序中的所有虚表均以一个固定结构的 header 开头。Header 中按顺序包含三个usize 大小的字段:drop_in_place ,size 和 align 。在 header 之后是一系列的 usize 大小字段,其数量以及含义在每个虚表中可能都不同。</p> -<figure class="highlight plaintext"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br></pre></td><td class="code"><pre><span class="line">+---------------+</span><br><span class="line">| drop_in_place |</span><br><span class="line">+---------------+</span><br><span class="line">| size |</span><br><span class="line">+---------------+</span><br><span class="line">| align |</span><br><span class="line">+---------------+</span><br><span class="line">| entry1 |</span><br><span class="line">+---------------+</span><br><span class="line">| entry2 |</span><br><span class="line">+---------------+</span><br><span class="line">| entry3 |</span><br><span class="line">+---------------+</span><br></pre></td></tr></table></figure> -<p>虚表 header 中的drop_in_place 是一个函数指针,其指向的函数能够原地 drop 当前胖指针所引用的对象。size 和 align 两个域分别给出对象的大小和内存对齐,这两个域共同构成一个 std::alloc::Layout 结构,可用于释放当前胖指针所引用的对象所占据的内存。虚表 header 的存在使得 trait object 总是能被销毁和释放。例如当销毁一个 <code>Box<dyn Trait></code> 时,<code>Box::<dyn Trait>::drop</code> 会首先调用虚表中的 drop_in_place 函数原地销毁 Box 所引用的对象,然后再调用 dealloc 函数并传递虚表中的 size 和 align 释放堆空间。</p> -<p>在虚表 header 之后是一系列的字段。在最普遍的情况下,每个字段代表一个指向 trait 所定义的函数的指针。例如,对于下列 object safe 的 trait:</p> -<figure class="highlight rust"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br></pre></td><td class="code"><pre><span class="line"><span class="keyword">pub</span> <span class="keyword">trait</span> <span class="title class_">Trait</span> {</span><br><span class="line"> <span class="keyword">fn</span> <span class="title function_">fun1</span>(&<span class="keyword">self</span>);</span><br><span class="line"> <span class="keyword">fn</span> <span class="title function_">fun2</span>(&<span class="keyword">self</span>);</span><br><span class="line"> <span class="keyword">fn</span> <span class="title function_">fun3</span>(&<span class="keyword">self</span>);</span><br><span class="line">}</span><br></pre></td></tr></table></figure> - -<p>如果类型T 实现了 Trait,那么为 T 生成的 Trait 虚表的结构为:</p> -<figure class="highlight plaintext"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br></pre></td><td class="code"><pre><span class="line">+--------------------------+</span><br><span class="line">| fn drop_in_place(*mut T) |</span><br><span class="line">+--------------------------+</span><br><span class="line">| size of T |</span><br><span class="line">+--------------------------+</span><br><span class="line">| align of T |</span><br><span class="line">+--------------------------+</span><br><span class="line">| fn <T as Trait>::fun1 |</span><br><span class="line">+--------------------------+</span><br><span class="line">| fn <T as Trait>::fun2 |</span><br><span class="line">+--------------------------+</span><br><span class="line">| fn <T as Trait>::fun3 |</span><br><span class="line">+--------------------------+</span><br></pre></td></tr></table></figure> - -<p>Trait 中的函数按照声明顺序依次排列在虚表 header 之后。当通过一个指向 T 对象的 &dyn Trait 调用 fun2 函数时,程序会先从虚表的第 5 个域中得到为 T 实现的 Trait::fun2 函数的地址,然后再调用之。</p> -<h2 id="Super-Trait"><a href="#Super-Trait" class="headerlink" title="Super Trait"></a>Super Trait</h2><p>Object safe 的 trait 可以有 super trait。例如:</p> -<figure class="highlight plaintext"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br></pre></td><td class="code"><pre><span class="line">pub trait Grand {</span><br><span class="line"> fn grand_fun1(&self);</span><br><span class="line"> fn grand_fun2(&self);</span><br><span class="line">}</span><br><span class="line"></span><br><span class="line">pub trait Parent : Grand {</span><br><span class="line"> fn parent_fun1(&self);</span><br><span class="line"> fn parent_fun2(&self);</span><br><span class="line">}</span><br><span class="line"></span><br><span class="line">pub trait Trait : Parent {</span><br><span class="line"> fn fun(&self);</span><br><span class="line">}</span><br></pre></td></tr></table></figure> - -<p>如果类型T 实现了 Trait,那么此时为 T 生成的 Trait 虚表的结构为:</p> -<figure class="highlight plaintext"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br><span class="line">16</span><br><span class="line">17</span><br></pre></td><td class="code"><pre><span class="line">+-------------------------------+</span><br><span class="line">| fn drop_in_place(*mut T) |</span><br><span class="line">+-------------------------------+</span><br><span class="line">| size of T |</span><br><span class="line">+-------------------------------+</span><br><span class="line">| align of T |</span><br><span class="line">+-------------------------------+</span><br><span class="line">| fn <T as Grand>::grand_fun1 |</span><br><span class="line">+-------------------------------+</span><br><span class="line">| fn <T as Grand>::grand_fun2 |</span><br><span class="line">+-------------------------------+</span><br><span class="line">| fn <T as Parent>::parent_fun1 |</span><br><span class="line">+-------------------------------+</span><br><span class="line">| fn <T as Parent>::parent_fun2 |</span><br><span class="line">+-------------------------------+</span><br><span class="line">| fn <T as Trait>::fun |</span><br><span class="line">+-------------------------------+</span><br></pre></td></tr></table></figure> - -<p>可以看到,此时Trait 以及 Trait 的所有直接或间接父 trait 所定义的所有函数均包含在虚表 header 之后,且顺序为后序(即先排布 Trait 的父 trait 所定义的所有函数,最后再排布 Trait 所定义的所有函数)。这样的排布方式使得在得到 T 类型的 Trait 虚表的同时也同时得到了 T 类型的 Parent 虚表和 Grand 虚表。T 类型的 Grand 虚表恰好由 Trait 虚表的前五个域构成,T 类型的 Parent 虚表恰好由 Trait 虚表的前七项构成。这使得向上转换变得非常简单。</p> -<p>所谓向上转换,即 Rust 允许将&dyn Trait 转换为 &dyn Parent 或 &dyn Grand 。在向上转换的过程中,胖指针的对象地址域保持不变,但 metadata 域可能需要进行调整,因为不同的 trait 可能具有不同的虚表地址。但在当前示例中,向上转换不需要调整 metadata 域,因为一个指向 Trait 虚表的指针同时也指向 Parent 虚表和 Grand 虚表。在后文中我们会进一步介绍需要调整 metadata 域的向上转换的情况。</p> -<blockquote> -<p>注意:目前 stable Rust 暂不支持向上转换。要使用向上转换特性,必须使用 nightly 工具链,并向源文件中添加 #![feature(trait_upcasting)] 特性开关。</p> -</blockquote> -<h2 id="多重继承"><a href="#多重继承" class="headerlink" title="多重继承"></a>多重继承</h2><p>Trait 可以有多个 super trait。例如:</p> -<figure class="highlight plaintext"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br><span class="line">16</span><br><span class="line">17</span><br><span class="line">18</span><br></pre></td><td class="code"><pre><span class="line">pub trait Base {</span><br><span class="line"> fn base_fun1(&self);</span><br><span class="line"> fn base_fun2(&self);</span><br><span class="line">}</span><br><span class="line"></span><br><span class="line">pub trait Left : Base {</span><br><span class="line"> fn left_fun1(&self);</span><br><span class="line"> fn left_fun2(&self);</span><br><span class="line">}</span><br><span class="line"></span><br><span class="line">pub trait Right : Base {</span><br><span class="line"> fn right_fun1(&self);</span><br><span class="line"> fn right_fun2(&self);</span><br><span class="line">}</span><br><span class="line"></span><br><span class="line">pub trait Trait : Left + Right {</span><br><span class="line"> fn fun(&self);</span><br><span class="line">}</span><br></pre></td></tr></table></figure> - -<p>如果类型T 实现了 Trait,那么此时为 T 生成的 Trait 虚表的结构为:</p> -<figure class="highlight plaintext"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br><span class="line">10</span><br><span class="line">11</span><br><span class="line">12</span><br><span class="line">13</span><br><span class="line">14</span><br><span class="line">15</span><br><span class="line">16</span><br><span class="line">17</span><br><span class="line">18</span><br><span class="line">19</span><br><span class="line">20</span><br><span class="line">21</span><br><span class="line">22</span><br><span class="line">23</span><br></pre></td><td class="code"><pre><span class="line">+-----------------------------+</span><br><span class="line">| fn drop_in_place(*mut T) |</span><br><span class="line">+-----------------------------+</span><br><span class="line">| size of T |</span><br><span class="line">+-----------------------------+</span><br><span class="line">| align of T |</span><br><span class="line">+-----------------------------+</span><br><span class="line">| fn <T as Base>::base_fun1 |</span><br><span class="line">+-----------------------------+</span><br><span class="line">| fn <T as Base>::base_fun2 |</span><br><span class="line">+-----------------------------+</span><br><span class="line">| fn <T as Left>::left_fun1 |</span><br><span class="line">+-----------------------------+</span><br><span class="line">| fn <T as Left>::left_fun2 |</span><br><span class="line">+-----------------------------+</span><br><span class="line">| fn <T as Right>::right_fun1 |</span><br><span class="line">+-----------------------------+</span><br><span class="line">| fn <T as Right>::right_fun2 |</span><br><span class="line">+-----------------------------+</span><br><span class="line">| ptr to <T as Right>::vtable |</span><br><span class="line">+-----------------------------+</span><br><span class="line">| fn <T as Trait>::fun |</span><br><span class="line">+-----------------------------+</span><br></pre></td></tr></table></figure> - -<p>可以看到,此时Trait 及其所有直接或间接父 trait 所定义的所有函数仍然包含在虚表内,因此通过 &dyn Trait 调用的函数仍然可以直接从虚表内得到其实际目标函数的地址。另外,Trait 虚表内仍然包含有效的 Base 虚表和 Left 虚表。因此,将 &dyn Trait 向上转换为 &dyn Left 或 &dyn Base 仍然是极其简单的,不需要调整胖指针的 metadata 域。但是,将 &dyn Trait 向上转换为 &dyn Right 就需要调整 metadata 域了,因为 Trait 虚表内并不包含一个有效的 Right 虚表。这也是 Trait 虚表中 ptr to <code><T as Right>::vtable</code> 域的作用:在执行向上转换时,程序会读取 Trait 虚表的这个域作为得到的 &dyn Right 胖指针的 metadata 。这也是 <em>Rust 向上转换与 C++ 向上转换的一个很大不同:在 C++ 中的向上转换通常并不需要访问虚表(除非需要执行跨虚继承边界的转换),但在 Rust 中向上转换可能需要访问虚表。</em></p> -<p>更加一般地,对于一个 object safe 的 traitTr,将其第一个父 trait、第一个父 trait 的第一个父 trait、…… 这一系列直接或间接父 trait 记为这个 trait 的 PrefixTrait 集合。在将 &dyn Tr 向上转换时,如果转换到的目标 trait 包含在 PrefixTrait 集合内,那么这个向上转换是平凡的:不需要调整胖指针的 metadata 域。否则,这个向上转换需要在 Tr 的虚表内读取目标 trait 的虚表指针作为转换结果的 metadata 。在 Tr 的虚表结构中,位于 PrefixTrait 集合中的父 trait 只需要排布他们所定义的函数即可;对于其他父 trait 还需要额外在虚表内排布一个指向其虚表的指针用于向上转换。</p> -<h2 id="向下转换"><a href="#向下转换" class="headerlink" title="向下转换"></a>向下转换</h2><p>Rust 提供了一个特殊的 trait:std::any::Any 。该 trait 支持向下转换,即可以将 &dyn Any 转换为 T 。转换过程中会对胖指针所指向的对象的实际类型进行检查,确认其确实是一个 T 类型的对象。Any trait 的虚表结构有一些特殊;在虚表 header 之后,Any 虚表仅包含一个域,这个域直接给出胖指针指向的对象的类型标识(由一个 std::any::TypeId 类型的值表示)。例如,对于任意的 T: ‘static,编译器为其生成的 Any 虚表为:</p> -<figure class="highlight plaintext"><table><tr><td class="gutter"><pre><span class="line">1</span><br><span class="line">2</span><br><span class="line">3</span><br><span class="line">4</span><br><span class="line">5</span><br><span class="line">6</span><br><span class="line">7</span><br><span class="line">8</span><br><span class="line">9</span><br></pre></td><td class="code"><pre><span class="line">+--------------------------+</span><br><span class="line">| fn drop_in_place(*mut T) |</span><br><span class="line">+--------------------------+</span><br><span class="line">| size of T |</span><br><span class="line">+--------------------------+</span><br><span class="line">| align of T |</span><br><span class="line">+--------------------------+</span><br><span class="line">| TypeId of T |</span><br><span class="line">+--------------------------+</span><br></pre></td></tr></table></figure> -<p>在执行向下转换时,程序首先检查转换到的类型是否与虚表中给出的TypeId 所标识的类型一致。若类型检查通过,向下转换操作可以直接返回胖指针中的指针域作为转换结果。</p> -<h3 id="REFERENCE"><a href="#REFERENCE" class="headerlink" title="REFERENCE"></a>REFERENCE</h3><ol> -<li>Vtable format to support dyn upcasting coercion <a target="_blank" rel="noopener" href="https://rust-lang.github.io/dyn-upcasting-coercion-initiative/design-discussions/vtable-layout.*html">https://rust-lang.github.io/dyn-upcasting-coercion-initiative/design-discussions/vtable-layout.*html</a>*</li> -</ol> - -</div> - -<script> - window.onload = detectors(); -</script> - <div class="post-footer"> - <div class="h-line-primary"></div> - <nav class="post-nav"> - <div class="prev-item"> - - <div class="icon arrow-left"></div> - <div class="post-link"> - <a href="/2024/05/02/Rust-Partial-%E8%AF%AD%E4%B9%89/">Prev</a> - </div> - - </div> - <div class="next-item"> - - <div class="icon arrow-right"></div> - <div class="post-link"> - <a href="/2024/04/30/TypeScript-With-Rust-Errors-No-Try-Catch-Heresy/">Next</a> - </div> - - </div> - </nav> -</div> - - - <div class="post-comment"> - - - - - - - -</div> - - -</article> - </div> - </div> - - <div class="footer"> - <div class="flex-container"> - <div class="footer-text"> - - - | - - - 希望路过的人可以添点柴火让这里暖和点 - - </div> - </div> -</div> - - </div> - - - <div class="search-popup"> - <div class="search-popup-overlay"> - </div> - <div class="search-popup-window" > - <div class="search-header"> - <div class="search-input-container"> - <input autocomplete="off" autocapitalize="off" maxlength="80" - placeholder="Search Anything" spellcheck="false" - type="search" class="search-input"> - </div> - <div class="search-close-btn"> - <div class="icon close-btn"></div> - </div> - </div> - <div class="search-result-container"> - </div> - </div> -</div> - -<script> - const searchConfig = { - path : "/search.xml", - top_n_per_article: "1", - unescape : "false", - trigger: "auto", - preload: "false" - } -</script> -<script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/search.js"></script> -<script src="/js/search.js"></script> - - - - </body> -</html> |
