<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="3.10.0">Jekyll</generator><link href="https://sizezheng.github.io/feed.xml" rel="self" type="application/atom+xml" /><link href="https://sizezheng.github.io/" rel="alternate" type="text/html" /><updated>2026-07-09T07:39:40-07:00</updated><id>https://sizezheng.github.io/feed.xml</id><title type="html">Si-Ze Zheng</title><subtitle>PhD@PKU</subtitle><author><name>Si-Ze Zheng</name><email>“zheng.size@outlook.com&quot;</email></author><entry><title type="html">Implementing High-performance GEMM on GPUs</title><link href="https://sizezheng.github.io/posts/2023/08/cuda-gemm/" rel="alternate" type="text/html" title="Implementing High-performance GEMM on GPUs" /><published>2023-08-01T00:00:00-07:00</published><updated>2023-08-01T00:00:00-07:00</updated><id>https://sizezheng.github.io/posts/2023/08/gemm</id><content type="html" xml:base="https://sizezheng.github.io/posts/2023/08/cuda-gemm/"><![CDATA[<p>Post available on Zhihu. For Ampere GPUs (<a href="https://zhuanlan.zhihu.com/p/631227862">link</a>). For Hopper GPUs (<a href="https://zhuanlan.zhihu.com/p/695589046">link</a>). And how to calculate multi-stage numbers (<a href="https://zhuanlan.zhihu.com/p/714353243">link</a>).</p>]]></content><author><name>Si-Ze Zheng</name><email>“zheng.size@outlook.com&quot;</email></author><category term="cuda" /><category term="high-performance" /><category term="ampere" /><category term="hopper" /><summary type="html"><![CDATA[Post available on Zhihu. For Ampere GPUs (link). For Hopper GPUs (link). And how to calculate multi-stage numbers (link).]]></summary></entry><entry><title type="html">[Paper Reading] Alpa@OSDI 2022</title><link href="https://sizezheng.github.io/posts/2022/05/osdi/" rel="alternate" type="text/html" title="[Paper Reading] Alpa@OSDI 2022" /><published>2022-05-28T00:00:00-07:00</published><updated>2022-05-28T00:00:00-07:00</updated><id>https://sizezheng.github.io/posts/2022/05/alpa</id><content type="html" xml:base="https://sizezheng.github.io/posts/2022/05/osdi/"><![CDATA[<p>The post is available on Zhihu (<a href="https://zhuanlan.zhihu.com/p/521211578">link</a>).</p>]]></content><author><name>Si-Ze Zheng</name><email>“zheng.size@outlook.com&quot;</email></author><category term="paper_reading" /><category term="osdi" /><summary type="html"><![CDATA[The post is available on Zhihu (link).]]></summary></entry><entry><title type="html">[Paper Reading] Rake@ASPLOS 2022</title><link href="https://sizezheng.github.io/posts/2022/05/asplos/" rel="alternate" type="text/html" title="[Paper Reading] Rake@ASPLOS 2022" /><published>2022-05-16T00:00:00-07:00</published><updated>2022-05-16T00:00:00-07:00</updated><id>https://sizezheng.github.io/posts/2022/05/rake</id><content type="html" xml:base="https://sizezheng.github.io/posts/2022/05/asplos/"><![CDATA[<p>The post is available on Zhihu (<a href="https://zhuanlan.zhihu.com/p/511381790">link</a>).</p>]]></content><author><name>Si-Ze Zheng</name><email>“zheng.size@outlook.com&quot;</email></author><category term="paper_reading" /><category term="asplos" /><summary type="html"><![CDATA[The post is available on Zhihu (link).]]></summary></entry><entry><title type="html">[Paper Reading] BOLT@MLSys 2022</title><link href="https://sizezheng.github.io/posts/2022/05/mlsys/" rel="alternate" type="text/html" title="[Paper Reading] BOLT@MLSys 2022" /><published>2022-05-13T00:00:00-07:00</published><updated>2022-05-13T00:00:00-07:00</updated><id>https://sizezheng.github.io/posts/2022/05/bolt</id><content type="html" xml:base="https://sizezheng.github.io/posts/2022/05/mlsys/"><![CDATA[<p>The post is available on Zhihu (<a href="https://zhuanlan.zhihu.com/p/514032549">link</a>).</p>]]></content><author><name>Si-Ze Zheng</name><email>“zheng.size@outlook.com&quot;</email></author><category term="paper_reading" /><category term="mlsys" /><summary type="html"><![CDATA[The post is available on Zhihu (link).]]></summary></entry><entry><title type="html">[Paper Reading] NASA@ISCA 2021</title><link href="https://sizezheng.github.io/posts/2022/05/isca/" rel="alternate" type="text/html" title="[Paper Reading] NASA@ISCA 2021" /><published>2022-05-13T00:00:00-07:00</published><updated>2022-05-13T00:00:00-07:00</updated><id>https://sizezheng.github.io/posts/2022/05/nasa</id><content type="html" xml:base="https://sizezheng.github.io/posts/2022/05/isca/"><![CDATA[<p>The post is available on Zhihu (<a href="https://zhuanlan.zhihu.com/p/513464183">link</a>).</p>]]></content><author><name>Si-Ze Zheng</name><email>“zheng.size@outlook.com&quot;</email></author><category term="paper_reading" /><category term="isca" /><summary type="html"><![CDATA[The post is available on Zhihu (link).]]></summary></entry><entry><title type="html">[Paper Reading] Astitch@ASPLOS 2022</title><link href="https://sizezheng.github.io/posts/2022/03/asplos/" rel="alternate" type="text/html" title="[Paper Reading] Astitch@ASPLOS 2022" /><published>2022-03-09T00:00:00-08:00</published><updated>2022-03-09T00:00:00-08:00</updated><id>https://sizezheng.github.io/posts/2022/03/astitch</id><content type="html" xml:base="https://sizezheng.github.io/posts/2022/03/asplos/"><![CDATA[<p>The post is available on Zhihu (<a href="https://zhuanlan.zhihu.com/p/477984880">link</a>).</p>]]></content><author><name>Si-Ze Zheng</name><email>“zheng.size@outlook.com&quot;</email></author><category term="paper_reading" /><category term="asplos" /><summary type="html"><![CDATA[The post is available on Zhihu (link).]]></summary></entry><entry><title type="html">Add PTX Tensor Core Codegen in TVM</title><link href="https://sizezheng.github.io/posts/2022/01/tvm/" rel="alternate" type="text/html" title="Add PTX Tensor Core Codegen in TVM" /><published>2022-01-14T00:00:00-08:00</published><updated>2022-01-14T00:00:00-08:00</updated><id>https://sizezheng.github.io/posts/2022/01/tvm-ptx</id><content type="html" xml:base="https://sizezheng.github.io/posts/2022/01/tvm/"><![CDATA[<p>The post is available on Zhihu (<a href="https://zhuanlan.zhihu.com/p/455166274">post 1</a> and <a href="https://zhuanlan.zhihu.com/p/456242751">post 2</a>).</p>]]></content><author><name>Si-Ze Zheng</name><email>“zheng.size@outlook.com&quot;</email></author><category term="technique" /><category term="tvm" /><category term="tensor_core" /><summary type="html"><![CDATA[The post is available on Zhihu (post 1 and post 2).]]></summary></entry><entry><title type="html">Install OpenCL driver on Hikey960</title><link href="https://sizezheng.github.io/posts/2020/10/hikey960/" rel="alternate" type="text/html" title="Install OpenCL driver on Hikey960" /><published>2020-10-08T00:00:00-07:00</published><updated>2020-10-08T00:00:00-07:00</updated><id>https://sizezheng.github.io/posts/2020/10/hikey960</id><content type="html" xml:base="https://sizezheng.github.io/posts/2020/10/hikey960/"><![CDATA[<p>The post is available on Zhihu (<a href="https://zhuanlan.zhihu.com/p/263051079">link</a>).</p>]]></content><author><name>Si-Ze Zheng</name><email>“zheng.size@outlook.com&quot;</email></author><category term="technique" /><category term="hikey960" /><category term="opencl" /><summary type="html"><![CDATA[The post is available on Zhihu (link).]]></summary></entry><entry><title type="html">FlexTensor@ASPLOS 2020</title><link href="https://sizezheng.github.io/posts/2020/03/flextensor/" rel="alternate" type="text/html" title="FlexTensor@ASPLOS 2020" /><published>2020-03-15T00:00:00-07:00</published><updated>2020-03-15T00:00:00-07:00</updated><id>https://sizezheng.github.io/posts/2020/03/flextensor</id><content type="html" xml:base="https://sizezheng.github.io/posts/2020/03/flextensor/"><![CDATA[<p>Post available on Zhihu (<a href="https://zhuanlan.zhihu.com/p/113340891">link</a>).</p>]]></content><author><name>Si-Ze Zheng</name><email>“zheng.size@outlook.com&quot;</email></author><category term="publication" /><category term="tvm" /><category term="compiler" /><category term="asplos" /><summary type="html"><![CDATA[Post available on Zhihu (link).]]></summary></entry></feed>