Update docs

Author: LeiWang1999

api/tilelang.engine.html

@@ -476,6 +476,7 @@ <h2>Submodules<a class="headerlink" href="#submodules" title="Permalink to this
 <li class="toctree-l2"><a class="reference internal" href="tilelang.engine.phase.html#tilelang.engine.phase.allow_tma_and_warp_specialized"><code class="docutils literal notranslate"><span class="pre">allow_tma_and_warp_specialized()</span></code></a></li>
 <li class="toctree-l2"><a class="reference internal" href="tilelang.engine.phase.html#tilelang.engine.phase.allow_vectorize"><code class="docutils literal notranslate"><span class="pre">allow_vectorize()</span></code></a></li>
 <li class="toctree-l2"><a class="reference internal" href="tilelang.engine.phase.html#tilelang.engine.phase.allow_warp_specialized"><code class="docutils literal notranslate"><span class="pre">allow_warp_specialized()</span></code></a></li>
+<li class="toctree-l2"><a class="reference internal" href="tilelang.engine.phase.html#tilelang.engine.phase.should_enable_aggressive_merge"><code class="docutils literal notranslate"><span class="pre">should_enable_aggressive_merge()</span></code></a></li>
 </ul>
 </li>
 </ul>

api/tilelang.engine.phase.html

@@ -455,6 +455,11 @@
 <span class="sig-prename descclassname"><span class="pre">tilelang.engine.phase.</span></span><span class="sig-name descname"><span class="pre">allow_warp_specialized</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pass_ctx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">PassContext</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">Target</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">bool</span></span></span><a class="headerlink" href="#tilelang.engine.phase.allow_warp_specialized" title="Permalink to this definition">#</a></dt>
 <dd></dd></dl>
 
+<dl class="py function">
+<dt class="sig sig-object py" id="tilelang.engine.phase.should_enable_aggressive_merge">
+<span class="sig-prename descclassname"><span class="pre">tilelang.engine.phase.</span></span><span class="sig-name descname"><span class="pre">should_enable_aggressive_merge</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pass_ctx</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">PassContext</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">target</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Optional</span><span class="p"><span class="pre">[</span></span><span class="pre">Target</span><span class="p"><span class="pre">]</span></span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">None</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">bool</span></span></span><a class="headerlink" href="#tilelang.engine.phase.should_enable_aggressive_merge" title="Permalink to this definition">#</a></dt>
+<dd></dd></dl>
+
 </section>
 
         </article>
@@ -520,6 +525,7 @@
 <li><a class="reference internal" href="#tilelang.engine.phase.allow_tma_and_warp_specialized"><code class="docutils literal notranslate"><span class="pre">allow_tma_and_warp_specialized()</span></code></a></li>
 <li><a class="reference internal" href="#tilelang.engine.phase.allow_vectorize"><code class="docutils literal notranslate"><span class="pre">allow_vectorize()</span></code></a></li>
 <li><a class="reference internal" href="#tilelang.engine.phase.allow_warp_specialized"><code class="docutils literal notranslate"><span class="pre">allow_warp_specialized()</span></code></a></li>
+<li><a class="reference internal" href="#tilelang.engine.phase.should_enable_aggressive_merge"><code class="docutils literal notranslate"><span class="pre">should_enable_aggressive_merge()</span></code></a></li>
 </ul>
 </li>
 </ul>

api/tilelang.html

@@ -659,6 +659,7 @@ <h2>Subpackages<a class="headerlink" href="#subpackages" title="Permalink to thi
 <li class="toctree-l4"><a class="reference internal" href="tilelang.engine.phase.html#tilelang.engine.phase.allow_tma_and_warp_specialized"><code class="docutils literal notranslate"><span class="pre">allow_tma_and_warp_specialized()</span></code></a></li>
 <li class="toctree-l4"><a class="reference internal" href="tilelang.engine.phase.html#tilelang.engine.phase.allow_vectorize"><code class="docutils literal notranslate"><span class="pre">allow_vectorize()</span></code></a></li>
 <li class="toctree-l4"><a class="reference internal" href="tilelang.engine.phase.html#tilelang.engine.phase.allow_warp_specialized"><code class="docutils literal notranslate"><span class="pre">allow_warp_specialized()</span></code></a></li>
+<li class="toctree-l4"><a class="reference internal" href="tilelang.engine.phase.html#tilelang.engine.phase.should_enable_aggressive_merge"><code class="docutils literal notranslate"><span class="pre">should_enable_aggressive_merge()</span></code></a></li>
 </ul>
 </li>
 </ul>

api/tilelang.language.logical.html

@@ -423,7 +423,7 @@
 <p>The language interface for tl programs.</p>
 <dl class="py function">
 <dt class="sig sig-object py" id="tilelang.language.logical.all_of">
-<span class="sig-prename descclassname"><span class="pre">tilelang.language.logical.</span></span><span class="sig-name descname"><span class="pre">all_of</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="pre">buffer:</span> <span class="pre">~typing.Union[&lt;tilelang.language.proxy.TensorProxy</span> <span class="pre">object</span> <span class="pre">at</span> <span class="pre">0x7f666d4564d0&gt;,</span> <span class="pre">~tvm.tir.stmt.BufferRegion]</span></em><span class="sig-paren">)</span><a class="headerlink" href="#tilelang.language.logical.all_of" title="Permalink to this definition">#</a></dt>
+<span class="sig-prename descclassname"><span class="pre">tilelang.language.logical.</span></span><span class="sig-name descname"><span class="pre">all_of</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="pre">buffer:</span> <span class="pre">~typing.Union[&lt;tilelang.language.proxy.TensorProxy</span> <span class="pre">object</span> <span class="pre">at</span> <span class="pre">0x7f31fa56f2b0&gt;,</span> <span class="pre">~tvm.tir.stmt.BufferRegion]</span></em><span class="sig-paren">)</span><a class="headerlink" href="#tilelang.language.logical.all_of" title="Permalink to this definition">#</a></dt>
 <dd><p>Check if all elements in the buffer are true.</p>
 <dl class="field-list simple">
 <dt class="field-odd">Parameters<span class="colon">:</span></dt>
@@ -437,7 +437,7 @@
 
 <dl class="py function">
 <dt class="sig sig-object py" id="tilelang.language.logical.any_of">
-<span class="sig-prename descclassname"><span class="pre">tilelang.language.logical.</span></span><span class="sig-name descname"><span class="pre">any_of</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="pre">buffer:</span> <span class="pre">~typing.Union[&lt;tilelang.language.proxy.TensorProxy</span> <span class="pre">object</span> <span class="pre">at</span> <span class="pre">0x7f666d4564d0&gt;,</span> <span class="pre">~tvm.tir.stmt.BufferRegion]</span></em><span class="sig-paren">)</span><a class="headerlink" href="#tilelang.language.logical.any_of" title="Permalink to this definition">#</a></dt>
+<span class="sig-prename descclassname"><span class="pre">tilelang.language.logical.</span></span><span class="sig-name descname"><span class="pre">any_of</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="pre">buffer:</span> <span class="pre">~typing.Union[&lt;tilelang.language.proxy.TensorProxy</span> <span class="pre">object</span> <span class="pre">at</span> <span class="pre">0x7f31fa56f2b0&gt;,</span> <span class="pre">~tvm.tir.stmt.BufferRegion]</span></em><span class="sig-paren">)</span><a class="headerlink" href="#tilelang.language.logical.any_of" title="Permalink to this definition">#</a></dt>
 <dd><p>Check if any element in the buffer is true.</p>
 <dl class="field-list simple">
 <dt class="field-odd">Parameters<span class="colon">:</span></dt>

api/tilelang.language.proxy.html

@@ -471,7 +471,7 @@
 <p>Buffer proxy class for constructing tir buffer.</p>
 <dl class="py method">
 <dt class="sig sig-object py" id="tilelang.language.proxy.BufferProxy.from_ptr">
-<span class="sig-name descname"><span class="pre">from_ptr</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pointer_var</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Var</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">shape</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">tvm.ir.expr.PrimExpr</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="p"><span class="pre">...</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'float32'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">&lt;tilelang.language.proxy.BufferProxy</span> <span class="pre">object</span> <span class="pre">at</span> <span class="pre">0x7f666d4564a0&gt;</span></span></span><a class="headerlink" href="#tilelang.language.proxy.BufferProxy.from_ptr" title="Permalink to this definition">#</a></dt>
+<span class="sig-name descname"><span class="pre">from_ptr</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">pointer_var</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">Var</span></span></em>, <em class="sig-param"><span class="n"><span class="pre">shape</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">tuple</span><span class="p"><span class="pre">[</span></span><span class="pre">tvm.ir.expr.PrimExpr</span><span class="p"><span class="pre">,</span></span><span class="w"> </span><span class="p"><span class="pre">...</span></span><span class="p"><span class="pre">]</span></span></span></em>, <em class="sig-param"><span class="n"><span class="pre">dtype</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">str</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">'float32'</span></span></em><span class="sig-paren">)</span> <span class="sig-return"><span class="sig-return-icon">&#x2192;</span> <span class="sig-return-typehint"><span class="pre">&lt;tilelang.language.proxy.BufferProxy</span> <span class="pre">object</span> <span class="pre">at</span> <span class="pre">0x7f31fa56fe80&gt;</span></span></span><a class="headerlink" href="#tilelang.language.proxy.BufferProxy.from_ptr" title="Permalink to this definition">#</a></dt>
 <dd><p>Create a buffer from a pointer, shape, and data type.</p>
 <dl class="field-list simple">
 <dt class="field-odd">Parameters<span class="colon">:</span></dt>

api/tilelang.transform.html

@@ -444,6 +444,7 @@ <h2>Submodules<a class="headerlink" href="#submodules" title="Permalink to this
 <li class="toctree-l3"><a class="reference internal" href="tilelang.transform.pass_config.html#tilelang.transform.pass_config.PassConfigKey.TL_DISABLE_TMA_LOWER"><code class="docutils literal notranslate"><span class="pre">PassConfigKey.TL_DISABLE_TMA_LOWER</span></code></a></li>
 <li class="toctree-l3"><a class="reference internal" href="tilelang.transform.pass_config.html#tilelang.transform.pass_config.PassConfigKey.TL_DISABLE_WARP_SPECIALIZED"><code class="docutils literal notranslate"><span class="pre">PassConfigKey.TL_DISABLE_WARP_SPECIALIZED</span></code></a></li>
 <li class="toctree-l3"><a class="reference internal" href="tilelang.transform.pass_config.html#tilelang.transform.pass_config.PassConfigKey.TL_DYNAMIC_ALIGNMENT"><code class="docutils literal notranslate"><span class="pre">PassConfigKey.TL_DYNAMIC_ALIGNMENT</span></code></a></li>
+<li class="toctree-l3"><a class="reference internal" href="tilelang.transform.pass_config.html#tilelang.transform.pass_config.PassConfigKey.TL_ENABLE_AGGRESSIVE_SHARED_MEMORY_MERGE"><code class="docutils literal notranslate"><span class="pre">PassConfigKey.TL_ENABLE_AGGRESSIVE_SHARED_MEMORY_MERGE</span></code></a></li>
 <li class="toctree-l3"><a class="reference internal" href="tilelang.transform.pass_config.html#tilelang.transform.pass_config.PassConfigKey.TL_SIMPLIFY"><code class="docutils literal notranslate"><span class="pre">PassConfigKey.TL_SIMPLIFY</span></code></a></li>
 </ul>
 </li>
@@ -742,7 +743,7 @@ <h2>Submodules<a class="headerlink" href="#submodules" title="Permalink to this
 
 <dl class="py function">
 <dt class="sig sig-object py" id="tilelang.transform.MergeSharedMemoryAllocations">
-<span class="sig-prename descclassname"><span class="pre">tilelang.transform.</span></span><span class="sig-name descname"><span class="pre">MergeSharedMemoryAllocations</span></span><span class="sig-paren">(</span><span class="sig-paren">)</span><a class="headerlink" href="#tilelang.transform.MergeSharedMemoryAllocations" title="Permalink to this definition">#</a></dt>
+<span class="sig-prename descclassname"><span class="pre">tilelang.transform.</span></span><span class="sig-name descname"><span class="pre">MergeSharedMemoryAllocations</span></span><span class="sig-paren">(</span><em class="sig-param"><span class="n"><span class="pre">enable_aggressive_merge</span></span><span class="p"><span class="pre">:</span></span><span class="w"> </span><span class="n"><span class="pre">bool</span></span><span class="w"> </span><span class="o"><span class="pre">=</span></span><span class="w"> </span><span class="default_value"><span class="pre">False</span></span></em><span class="sig-paren">)</span><a class="headerlink" href="#tilelang.transform.MergeSharedMemoryAllocations" title="Permalink to this definition">#</a></dt>
 <dd><dl class="field-list simple">
 <dt class="field-odd">Returns<span class="colon">:</span></dt>
 <dd class="field-odd"><p><strong>fpass</strong> – The result pass</p>

api/tilelang.transform.pass_config.html

@@ -623,6 +623,17 @@
 </dl>
 </dd></dl>
 
+<dl class="py attribute">
+<dt class="sig sig-object py" id="tilelang.transform.pass_config.PassConfigKey.TL_ENABLE_AGGRESSIVE_SHARED_MEMORY_MERGE">
+<span class="sig-name descname"><span class="pre">TL_ENABLE_AGGRESSIVE_SHARED_MEMORY_MERGE</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'tl.enable_aggressive_shared_memory_merge'</span></em><a class="headerlink" href="#tilelang.transform.pass_config.PassConfigKey.TL_ENABLE_AGGRESSIVE_SHARED_MEMORY_MERGE" title="Permalink to this definition">#</a></dt>
+<dd><p>False</p>
+<dl class="field-list simple">
+<dt class="field-odd">Type<span class="colon">:</span></dt>
+<dd class="field-odd"><p>Enable aggressive merge of shared memory allocations. Default</p>
+</dd>
+</dl>
+</dd></dl>
+
 <dl class="py attribute">
 <dt class="sig sig-object py" id="tilelang.transform.pass_config.PassConfigKey.TL_SIMPLIFY">
 <span class="sig-name descname"><span class="pre">TL_SIMPLIFY</span></span><em class="property"><span class="w"> </span><span class="p"><span class="pre">=</span></span><span class="w"> </span><span class="pre">'tl.Simplify'</span></em><a class="headerlink" href="#tilelang.transform.pass_config.PassConfigKey.TL_SIMPLIFY" title="Permalink to this definition">#</a></dt>
@@ -713,6 +724,7 @@
 <li><a class="reference internal" href="#tilelang.transform.pass_config.PassConfigKey.TL_DISABLE_TMA_LOWER"><code class="docutils literal notranslate"><span class="pre">PassConfigKey.TL_DISABLE_TMA_LOWER</span></code></a></li>
 <li><a class="reference internal" href="#tilelang.transform.pass_config.PassConfigKey.TL_DISABLE_WARP_SPECIALIZED"><code class="docutils literal notranslate"><span class="pre">PassConfigKey.TL_DISABLE_WARP_SPECIALIZED</span></code></a></li>
 <li><a class="reference internal" href="#tilelang.transform.pass_config.PassConfigKey.TL_DYNAMIC_ALIGNMENT"><code class="docutils literal notranslate"><span class="pre">PassConfigKey.TL_DYNAMIC_ALIGNMENT</span></code></a></li>
+<li><a class="reference internal" href="#tilelang.transform.pass_config.PassConfigKey.TL_ENABLE_AGGRESSIVE_SHARED_MEMORY_MERGE"><code class="docutils literal notranslate"><span class="pre">PassConfigKey.TL_ENABLE_AGGRESSIVE_SHARED_MEMORY_MERGE</span></code></a></li>
 <li><a class="reference internal" href="#tilelang.transform.pass_config.PassConfigKey.TL_SIMPLIFY"><code class="docutils literal notranslate"><span class="pre">PassConfigKey.TL_SIMPLIFY</span></code></a></li>
 </ul>
 </li>