rebase

2025-11-05 11:28:12 +08:00 · 2025-01-09 21:56:20 +00:00
parent 830f99716d
commit 231caa94a6
2642 changed files with 137687 additions and 70861 deletions
--- a/docs/build/html/steelgemmmasked_8h.html
+++ b/docs/build/html/steelgemmmasked_8h.html
@@ -3,7 +3,7 @@
 <head>
 <meta http-equiv="Content-Type" content="text/xhtml;charset=UTF-8"/>
 <meta http-equiv="X-UA-Compatible" content="IE=11"/>
-<meta name="generator" content="Doxygen 1.12.0"/>
+<meta name="generator" content="Doxygen 1.13.1"/>
 <meta name="viewport" content="width=device-width, initial-scale=1"/>
 <title>MLX: mlx/backend/metal/kernels/steel/gemm/kernels/steel_gemm_masked.h File Reference</title>
 <link href="tabs.css" rel="stylesheet" type="text/css"/>
@@ -11,11 +11,18 @@
 <script type="text/javascript" src="dynsections.js"></script>
 <script type="text/javascript" src="clipboard.js"></script>
 <link href="navtree.css" rel="stylesheet" type="text/css"/>
+<script type="text/javascript" src="navtreedata.js"></script>
+<script type="text/javascript" src="navtree.js"></script>
 <script type="text/javascript" src="resize.js"></script>
 <script type="text/javascript" src="cookie.js"></script>
 <link href="search/search.css" rel="stylesheet" type="text/css"/>
 <script type="text/javascript" src="search/searchdata.js"></script>
 <script type="text/javascript" src="search/search.js"></script>
+<script type="text/javascript">
+/* @license magnet:?xt=urn:btih:d3d9a9a6595521f9666a5e94cc830dab83b65699&amp;dn=expat.txt MIT */
+  $(function() { init_search(); });
+/* @license-end */
+</script>
 <link href="doxygen.css" rel="stylesheet" type="text/css" />
 </head>
 <body>
@@ -28,12 +35,24 @@
   <div id="projectname">MLX
   </div>
  </td>
+    <td>        <div id="MSearchBox" class="MSearchBoxInactive">
+        <span class="left">
+          <span id="MSearchSelect"                onmouseover="return searchBox.OnSearchSelectShow()"                onmouseout="return searchBox.OnSearchSelectHide()">&#160;</span>
+          <input type="text" id="MSearchField" value="" placeholder="Search" accesskey="S"
+               onfocus="searchBox.OnSearchFieldFocus(true)" 
+               onblur="searchBox.OnSearchFieldFocus(false)" 
+               onkeyup="searchBox.OnSearchFieldChange(event)"/>
+          </span><span class="right">
+            <a id="MSearchClose" href="javascript:searchBox.CloseResultsWindow()"><img id="MSearchCloseImg" border="0" src="search/close.svg" alt=""/></a>
+          </span>
+        </div>
+</td>
 </tr>
 </tbody>
 </table>
 </div>
 <!-- end header part -->
-<!-- Generated by Doxygen 1.12.0 -->
+<!-- Generated by Doxygen 1.13.1 -->
 <script type="text/javascript">
 /* @license magnet:?xt=urn:btih:d3d9a9a6595521f9666a5e94cc830dab83b65699&amp;dn=expat.txt MIT */
 var searchBox = new SearchBox("searchBox", "search/",'.html');
@@ -44,22 +63,23 @@ var searchBox = new SearchBox("searchBox", "search/",'.html');
 $(function() { codefold.init(0); });
 /* @license-end */
 </script>
-<script type="text/javascript" src="menudata.js"></script>
-<script type="text/javascript" src="menu.js"></script>
+</div><!-- top -->
+<div id="side-nav" class="ui-resizable side-nav-resizable">
+  <div id="nav-tree">
+    <div id="nav-tree-contents">
+      <div id="nav-sync" class="sync"></div>
+    </div>
+  </div>
+  <div id="splitbar" style="-moz-user-select:none;" 
+       class="ui-resizable-handle">
+  </div>
+</div>
 <script type="text/javascript">
 /* @license magnet:?xt=urn:btih:d3d9a9a6595521f9666a5e94cc830dab83b65699&amp;dn=expat.txt MIT */
-$(function() {
-  initMenu('',true,false,'search.php','Search',false);
-  $(function() { init_search(); });
-});
-/* @license-end */
-</script>
-<div id="main-nav"></div>
-<script type="text/javascript">
-/* @license magnet:?xt=urn:btih:d3d9a9a6595521f9666a5e94cc830dab83b65699&amp;dn=expat.txt MIT */
-$(function(){ initResizable(false); });
+$(function(){initNavTree('steel__gemm__masked_8h.html',''); initResizable(true); });
 /* @license-end */
 </script>
+<div id="doc-content">
 <!-- window showing the filter options -->
 <div id="MSearchSelectWindow"
     onmouseover="return searchBox.OnSearchSelectShow()"
@@ -81,12 +101,6 @@ $(function(){ initResizable(false); });
 </div>
 </div>

-<div id="nav-path" class="navpath">
-  <ul>
-<li class="navelem"><a class="el" href="dir_938ab0ecf10b8b860ff766c820f665fd.html">mlx</a></li><li class="navelem"><a class="el" href="dir_1d446c9bd3c99228254c9484e0bc5c06.html">backend</a></li><li class="navelem"><a class="el" href="dir_d0c977ea65824390717cdb7efc36c157.html">metal</a></li><li class="navelem"><a class="el" href="dir_70a37effa88bcbd6b791977fa1e64356.html">kernels</a></li><li class="navelem"><a class="el" href="dir_76215a6c54e2b67053e723fc2395583c.html">steel</a></li><li class="navelem"><a class="el" href="dir_6768c99e6145fb9510ccdb40db8ede25.html">gemm</a></li><li class="navelem"><a class="el" href="dir_9c555e3d0f5b8c3fb3a7397c81fd5bf9.html">kernels</a></li>  </ul>
-</div>
-</div><!-- top -->
-<div id="doc-content">
 <div class="header">
  <div class="summary">
 <a href="#nested-classes">Classes</a> &#124;
@@ -107,21 +121,21 @@ Classes</h2></td></tr>
 </table><table class="memberdecls">
 <tr class="heading"><td colspan="2"><h2 class="groupheader"><a id="func-members" name="func-members"></a>
 Functions</h2></td></tr>
-<tr class="memitem:af805e998b2046ee30c2b4be813e3af97" id="r_af805e998b2046ee30c2b4be813e3af97"><td class="memTemplParams" colspan="2">template&lt;typename T , typename out_mask_t , typename op_mask_t , int BM, int BN, int BK, int WM, int WN, bool transpose_a, bool transpose_b, bool MN_aligned, bool K_aligned&gt; </td></tr>
-<tr class="memitem:af805e998b2046ee30c2b4be813e3af97"><td class="memTemplItemLeft" align="right" valign="top">void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="#af805e998b2046ee30c2b4be813e3af97">block_masked_gemm</a> (const device T *A, const device T *B, device T *D, const constant <a class="el" href="structmlx_1_1steel_1_1_g_e_m_m_params.html">GEMMParams</a> *params, const constant int *batch_shape, const constant size_t *batch_strides, const device out_mask_t *out_mask, const device op_mask_t *lhs_mask, const device op_mask_t *rhs_mask, const constant int *mask_strides, uint simd_lane_id, uint simd_group_id, uint3 tid, uint3 lid)</td></tr>
-<tr class="separator:af805e998b2046ee30c2b4be813e3af97"><td class="memSeparator" colspan="2">&#160;</td></tr>
-<tr class="memitem:a477932e2ae9d49366f7ede6db63f9cac" id="r_a477932e2ae9d49366f7ede6db63f9cac"><td class="memTemplParams" colspan="2">template&lt;typename T , int BM, int BN, int BK, int WM, int WN, bool transpose_a, bool transpose_b, bool MN_aligned, bool K_aligned, bool has_operand_mask = false&gt; </td></tr>
-<tr class="memitem:a477932e2ae9d49366f7ede6db63f9cac"><td class="memTemplItemLeft" align="right" valign="top">void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="#a477932e2ae9d49366f7ede6db63f9cac">block_masked_gemm</a> (const device T *A, const device T *B, device T *D, const constant <a class="el" href="structmlx_1_1steel_1_1_g_e_m_m_params.html">GEMMParams</a> *params, const constant int *batch_shape, const constant size_t *batch_strides, const device bool *out_mask, const device bool *lhs_mask, const device bool *rhs_mask, const constant int *mask_strides, uint simd_lane_id, uint simd_group_id, uint3 tid, uint3 lid)</td></tr>
-<tr class="separator:a477932e2ae9d49366f7ede6db63f9cac"><td class="memSeparator" colspan="2">&#160;</td></tr>
+<tr class="memitem:a3642df61cba8eff47a08a20834137eac" id="r_a3642df61cba8eff47a08a20834137eac"><td class="memTemplParams" colspan="2">template&lt;typename T, typename out_mask_t, typename op_mask_t, int BM, int BN, int BK, int WM, int WN, bool transpose_a, bool transpose_b, bool MN_aligned, bool K_aligned&gt; </td></tr>
+<tr class="memitem:a3642df61cba8eff47a08a20834137eac"><td class="memTemplItemLeft" align="right" valign="top">void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="#a3642df61cba8eff47a08a20834137eac">block_masked_gemm</a> (const device T *A, const device T *B, device T *D, const constant <a class="el" href="structmlx_1_1steel_1_1_g_e_m_m_params.html">GEMMParams</a> *params, const constant int *batch_shape, const constant int64_t *batch_strides, const device out_mask_t *out_mask, const device op_mask_t *lhs_mask, const device op_mask_t *rhs_mask, const constant int *mask_strides, uint simd_lane_id, uint simd_group_id, uint3 tid, uint3 lid)</td></tr>
+<tr class="separator:a3642df61cba8eff47a08a20834137eac"><td class="memSeparator" colspan="2">&#160;</td></tr>
+<tr class="memitem:ac59086cee8cf396751da691fe80cda7b" id="r_ac59086cee8cf396751da691fe80cda7b"><td class="memTemplParams" colspan="2">template&lt;typename T, int BM, int BN, int BK, int WM, int WN, bool transpose_a, bool transpose_b, bool MN_aligned, bool K_aligned, bool has_operand_mask = false&gt; </td></tr>
+<tr class="memitem:ac59086cee8cf396751da691fe80cda7b"><td class="memTemplItemLeft" align="right" valign="top">void&#160;</td><td class="memTemplItemRight" valign="bottom"><a class="el" href="#ac59086cee8cf396751da691fe80cda7b">block_masked_gemm</a> (const device T *A, const device T *B, device T *D, const constant <a class="el" href="structmlx_1_1steel_1_1_g_e_m_m_params.html">GEMMParams</a> *params, const constant int *batch_shape, const constant int64_t *batch_strides, const device bool *out_mask, const device bool *lhs_mask, const device bool *rhs_mask, const constant int *mask_strides, uint simd_lane_id, uint simd_group_id, uint3 tid, uint3 lid)</td></tr>
+<tr class="separator:ac59086cee8cf396751da691fe80cda7b"><td class="memSeparator" colspan="2">&#160;</td></tr>
 </table>
 <h2 class="groupheader">Function Documentation</h2>
-<a id="a477932e2ae9d49366f7ede6db63f9cac" name="a477932e2ae9d49366f7ede6db63f9cac"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#a477932e2ae9d49366f7ede6db63f9cac">&#9670;&#160;</a></span>block_masked_gemm() <span class="overload">[1/2]</span></h2>
+<a id="ac59086cee8cf396751da691fe80cda7b" name="ac59086cee8cf396751da691fe80cda7b"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#ac59086cee8cf396751da691fe80cda7b">&#9670;&#160;</a></span>block_masked_gemm() <span class="overload">[1/2]</span></h2>

 <div class="memitem">
 <div class="memproto">
 <div class="memtemplate">
-template&lt;typename T , int BM, int BN, int BK, int WM, int WN, bool transpose_a, bool transpose_b, bool MN_aligned, bool K_aligned, bool has_operand_mask = false&gt; </div>
+template&lt;typename T, int BM, int BN, int BK, int WM, int WN, bool transpose_a, bool transpose_b, bool MN_aligned, bool K_aligned, bool has_operand_mask = false&gt; </div>
      <table class="memname">
        <tr>
          <td class="memname">void block_masked_gemm </td>
@@ -151,7 +165,7 @@ template&lt;typename T , int BM, int BN, int BK, int WM, int WN, bool transpose_
        <tr>
          <td class="paramkey"></td>
          <td></td>
-          <td class="paramtype">const constant size_t *</td>          <td class="paramname"><span class="paramname"><em>batch_strides</em></span>, </td>
+          <td class="paramtype">const constant int64_t *</td>          <td class="paramname"><span class="paramname"><em>batch_strides</em></span>, </td>
        </tr>
        <tr>
          <td class="paramkey"></td>
@@ -198,13 +212,13 @@ template&lt;typename T , int BM, int BN, int BK, int WM, int WN, bool transpose_

 </div>
 </div>
-<a id="af805e998b2046ee30c2b4be813e3af97" name="af805e998b2046ee30c2b4be813e3af97"></a>
-<h2 class="memtitle"><span class="permalink"><a href="#af805e998b2046ee30c2b4be813e3af97">&#9670;&#160;</a></span>block_masked_gemm() <span class="overload">[2/2]</span></h2>
+<a id="a3642df61cba8eff47a08a20834137eac" name="a3642df61cba8eff47a08a20834137eac"></a>
+<h2 class="memtitle"><span class="permalink"><a href="#a3642df61cba8eff47a08a20834137eac">&#9670;&#160;</a></span>block_masked_gemm() <span class="overload">[2/2]</span></h2>

 <div class="memitem">
 <div class="memproto">
 <div class="memtemplate">
-template&lt;typename T , typename out_mask_t , typename op_mask_t , int BM, int BN, int BK, int WM, int WN, bool transpose_a, bool transpose_b, bool MN_aligned, bool K_aligned&gt; </div>
+template&lt;typename T, typename out_mask_t, typename op_mask_t, int BM, int BN, int BK, int WM, int WN, bool transpose_a, bool transpose_b, bool MN_aligned, bool K_aligned&gt; </div>
      <table class="memname">
        <tr>
          <td class="memname">void block_masked_gemm </td>
@@ -234,7 +248,7 @@ template&lt;typename T , typename out_mask_t , typename op_mask_t , int BM, int
        <tr>
          <td class="paramkey"></td>
          <td></td>
-          <td class="paramtype">const constant size_t *</td>          <td class="paramname"><span class="paramname"><em>batch_strides</em></span>, </td>
+          <td class="paramtype">const constant int64_t *</td>          <td class="paramname"><span class="paramname"><em>batch_strides</em></span>, </td>
        </tr>
        <tr>
          <td class="paramkey"></td>
@@ -282,10 +296,13 @@ template&lt;typename T , typename out_mask_t , typename op_mask_t , int BM, int
 </div>
 </div>
 </div><!-- contents -->
-<!-- start footer part -->
-<hr class="footer"/><address class="footer"><small>
-Generated by&#160;<a href="https://www.doxygen.org/index.html"><img class="footer" src="doxygen.svg" width="104" height="31" alt="doxygen"/></a> 1.12.0
-</small></address>
 </div><!-- doc-content -->
+<!-- start footer part -->
+<div id="nav-path" class="navpath"><!-- id is needed for treeview function! -->
+  <ul>
+    <li class="navelem"><a class="el" href="dir_938ab0ecf10b8b860ff766c820f665fd.html">mlx</a></li><li class="navelem"><a class="el" href="dir_1d446c9bd3c99228254c9484e0bc5c06.html">backend</a></li><li class="navelem"><a class="el" href="dir_d0c977ea65824390717cdb7efc36c157.html">metal</a></li><li class="navelem"><a class="el" href="dir_70a37effa88bcbd6b791977fa1e64356.html">kernels</a></li><li class="navelem"><a class="el" href="dir_76215a6c54e2b67053e723fc2395583c.html">steel</a></li><li class="navelem"><a class="el" href="dir_6768c99e6145fb9510ccdb40db8ede25.html">gemm</a></li><li class="navelem"><a class="el" href="dir_9c555e3d0f5b8c3fb3a7397c81fd5bf9.html">kernels</a></li><li class="navelem"><a class="el" href="steel__gemm__masked_8h.html">steel_gemm_masked.h</a></li>
+    <li class="footer">Generated by <a href="https://www.doxygen.org/index.html"><img class="footer" src="doxygen.svg" width="104" height="31" alt="doxygen"/></a> 1.13.1 </li>
+  </ul>
+</div>
 </body>
 </html>