feat(docs): rebuild whitepaper site and content

.github/workflows/pages.yml

@@ -21,7 +21,7 @@ concurrency:
 jobs:
   deploy:
     # Only run on the original repository, not on forks
-    if: github.repository == 'LessUp/gpu-spmv'
+    if: github.repository == 'AICL-Lab/gpu-spmv'
     runs-on: ubuntu-latest
     environment:
       name: github-pages
@@ -68,6 +68,10 @@ jobs:
         working-directory: docs
         run: npm run sync
 
+      - name: Verify docs site
+        working-directory: docs
+        run: npm run verify:site
+
       - name: Build docs
         working-directory: docs
         env:

CHANGELOG.md

@@ -42,7 +42,7 @@ This is the first stable release of GPU SpMV, featuring complete CSR and ELL for
 - Doxygen-compatible documentation
 
 #### Documentation
-- Full documentation site at https://lessup.github.io/gpu-spmv/
+- Full documentation site at https://aicl-lab.github.io/gpu-spmv/
 - Bilingual README (English and Chinese)
 - API reference, performance guide, and code examples
 - Architecture documentation and design decision records
@@ -140,5 +140,5 @@ No breaking changes from pre-release versions. The API is now stable.
 
 ---
 
-[1.0.0]: https://github.com/LessUp/gpu-spmv/releases/tag/v1.0.0
-[0.1.0]: https://github.com/LessUp/gpu-spmv/tree/7d6dd0c
+[1.0.0]: https://github.com/AICL-Lab/gpu-spmv/releases/tag/v1.0.0
+[0.1.0]: https://github.com/AICL-Lab/gpu-spmv/tree/7d6dd0c

README.md

@@ -16,16 +16,16 @@
 </p>
 
 <p align="center">
-  <a href="https://github.com/LessUp/gpu-spmv/actions/workflows/ci.yml">
-    <img src="https://github.com/LessUp/gpu-spmv/actions/workflows/ci.yml/badge.svg" alt="CI">
+  <a href="https://github.com/AICL-Lab/gpu-spmv/actions/workflows/ci.yml">
+    <img src="https://github.com/AICL-Lab/gpu-spmv/actions/workflows/ci.yml/badge.svg" alt="CI">
   </a>
-  <a href="https://lessup.github.io/gpu-spmv/">
+  <a href="https://aicl-lab.github.io/gpu-spmv/">
     <img src="https://img.shields.io/badge/Docs-GitHub%20Pages-2EA44F?logo=github" alt="Documentation">
   </a>
-  <a href="https://github.com/LessUp/gpu-spmv/releases">
-    <img src="https://img.shields.io/github/v/release/LessUp/gpu-spmv?color=blue" alt="Release">
+  <a href="https://github.com/AICL-Lab/gpu-spmv/releases">
+    <img src="https://img.shields.io/github/v/release/AICL-Lab/gpu-spmv?color=blue" alt="Release">
   </a>
-  <a href="https://github.com/LessUp/gpu-spmv/blob/main/LICENSE">
+  <a href="https://github.com/AICL-Lab/gpu-spmv/blob/main/LICENSE">
     <img src="https://img.shields.io/badge/License-MIT-green" alt="License">
   </a>
 </p>
@@ -105,7 +105,7 @@ if (result.error != SpMVError::SUCCESS) {
 
 ```bash
 # 1. Clone
-git clone https://github.com/LessUp/gpu-spmv.git && cd gpu-spmv
+git clone https://github.com/AICL-Lab/gpu-spmv.git && cd gpu-spmv
 
 # 2. Build
 cmake --preset release && cmake --build --preset release
@@ -143,7 +143,7 @@ int main() {
 }
 ```
 
-📚 **More examples**: [Documentation Site](https://lessup.github.io/gpu-spmv/examples)
+📚 **More examples**: [Documentation Site](https://aicl-lab.github.io/gpu-spmv/en/examples/basic-spmv)
 
 ---
 
@@ -167,7 +167,7 @@ Benchmark on **NVIDIA RTX 3090** (Ampere, 936 GB/s peak):
 # Avg time: 23.5 ms | Bandwidth: 69.8 GB/s (71.5% of peak)
 ```
 
-📈 **Full performance guide**: [Performance Optimization](https://lessup.github.io/gpu-spmv/performance)
+📈 **Full performance guide**: [Performance Optimization](https://aicl-lab.github.io/gpu-spmv/en/performance/optimization-guide)
 
 ---
 
@@ -194,16 +194,16 @@ gpu-spmv/
 
 ## 📚 Documentation
 
-Complete documentation is available at **[https://lessup.github.io/gpu-spmv/](https://lessup.github.io/gpu-spmv/)**:
+Complete documentation is available at **[https://aicl-lab.github.io/gpu-spmv/](https://aicl-lab.github.io/gpu-spmv/)**:
 
 | Document | Description |
 |:---------|:------------|
-| [📦 Installation Guide](https://lessup.github.io/gpu-spmv/installation) | System requirements, detailed installation |
-| [📚 API Reference](https://lessup.github.io/gpu-spmv/api) | Complete API documentation, data structures |
-| [📝 Examples](https://lessup.github.io/gpu-spmv/examples) | 7 complete code examples (basic → advanced) |
-| [🚀 Performance Guide](https://lessup.github.io/gpu-spmv/performance) | Tuning strategies, benchmark data |
-| [🏗️ Architecture](https://lessup.github.io/gpu-spmv/architecture) | System design, kernel selection |
-| [📋 Changelog](https://lessup.github.io/gpu-spmv/changelog) | Version history, migration guide |
+| [📦 Installation Guide](https://aicl-lab.github.io/gpu-spmv/en/quickstart) | System requirements, detailed installation |
+| [📚 API Reference](https://aicl-lab.github.io/gpu-spmv/en/api/spmv) | Complete API documentation, data structures |
+| [📝 Examples](https://aicl-lab.github.io/gpu-spmv/en/examples/basic-spmv) | End-to-end code example and walkthrough |
+| [🚀 Performance Guide](https://aicl-lab.github.io/gpu-spmv/en/performance/optimization-guide) | Tuning strategies, benchmark data |
+| [🏗️ Architecture](https://aicl-lab.github.io/gpu-spmv/en/architecture/overview) | System design, kernel selection |
+| [📋 Changelog](https://aicl-lab.github.io/gpu-spmv/en/changelog) | Version history, migration guide |
 
 ---
 

docs/.vitepress/config.ts

@@ -26,11 +26,11 @@ export default withMermaid(
           content: 'High-Performance CUDA Sparse Matrix-Vector Multiplication Library'
         }
       ],
-      ['meta', { property: 'og:image', content: `${base}images/og-image.svg` }],
+      ['meta', { property: 'og:image', content: `${base}images/social/og-dark.svg` }],
       ['meta', { name: 'twitter:card', content: 'summary_large_image' }],
       ['meta', { name: 'twitter:title', content: 'GPU SpMV' }],
       ['meta', { name: 'twitter:description', content: 'High-Performance CUDA Sparse Matrix-Vector Multiplication Library' }],
-      ['meta', { name: 'twitter:image', content: `${base}images/og-image.svg` }],
+      ['meta', { name: 'twitter:image', content: `${base}images/social/og-dark.svg` }],
       ['link', { rel: 'icon', href: `${base}images/favicon.svg`, type: 'image/svg+xml' }],
       ['link', { rel: 'preconnect', href: 'https://fonts.googleapis.com' }],
       ['link', { rel: 'preconnect', href: 'https://fonts.gstatic.com', crossorigin: '' }],
@@ -59,10 +59,10 @@ export default withMermaid(
         themeConfig: {
           nav: [
             { text: '技术白皮书', link: '/zh/whitepaper/', activeMatch: '/zh/whitepaper/' },
-            { text: '快速开始', link: '/zh/quickstart', activeMatch: '/zh/(quickstart|examples)/' },
             { text: '架构设计', link: '/zh/architecture/overview', activeMatch: '/zh/architecture/' },
+            { text: '性能测试', link: '/zh/performance/benchmarks', activeMatch: '/zh/performance/' },
             { text: 'API 参考', link: '/zh/api/spmv', activeMatch: '/zh/api/' },
-            { text: '性能测试', link: '/zh/performance/benchmarks', activeMatch: '/zh/performance/' }
+            { text: '学术引用', link: '/zh/references', activeMatch: '/zh/(references|citation)/' }
           ],
           sidebar: {
             '/zh/': [
@@ -88,15 +88,18 @@ export default withMermaid(
                 collapsed: false,
                 items: [
                   { text: '系统概览', link: '/zh/architecture/overview' },
+                  { text: '执行流水线', link: '/zh/architecture/execution-pipeline' },
                   { text: 'Kernel 选择策略', link: '/zh/architecture/kernel-selection' },
                   { text: '内存布局', link: '/zh/architecture/memory-layout' },
+                  { text: '可靠性约束', link: '/zh/architecture/reliability' },
                   { text: 'Spec-Driven 开发', link: '/zh/architecture/spec-driven' }
                 ]
               },
               {
                 text: '性能优化',
                 collapsed: false,
                 items: [
+                  { text: '性能方法学', link: '/zh/performance/methodology' },
                   { text: '基准测试', link: '/zh/performance/benchmarks' },
                   { text: '优化指南', link: '/zh/performance/optimization-guide' }
                 ]
@@ -124,7 +127,7 @@ export default withMermaid(
             ]
           },
           editLink: {
-            pattern: 'https://github.com/LessUp/gpu-spmv/edit/main/docs/:path',
+            pattern: 'https://github.com/AICL-Lab/gpu-spmv/edit/main/docs/:path',
             text: '在 GitHub 上编辑此页'
           },
           docFooter: { prev: '上一页', next: '下一页' },
@@ -145,10 +148,10 @@ export default withMermaid(
         themeConfig: {
           nav: [
             { text: 'Whitepaper', link: '/en/whitepaper/', activeMatch: '/en/whitepaper/' },
-            { text: 'Getting Started', link: '/en/quickstart', activeMatch: '/en/(quickstart|examples)/' },
             { text: 'Architecture', link: '/en/architecture/overview', activeMatch: '/en/architecture/' },
+            { text: 'Benchmarks', link: '/en/performance/benchmarks', activeMatch: '/en/performance/' },
             { text: 'API Reference', link: '/en/api/spmv', activeMatch: '/en/api/' },
-            { text: 'Benchmarks', link: '/en/performance/benchmarks', activeMatch: '/en/performance/' }
+            { text: 'References', link: '/en/references', activeMatch: '/en/(references|citation)/' }
           ],
           sidebar: {
             '/en/': [
@@ -174,15 +177,18 @@ export default withMermaid(
                 collapsed: false,
                 items: [
                   { text: 'System Overview', link: '/en/architecture/overview' },
+                  { text: 'Execution Pipeline', link: '/en/architecture/execution-pipeline' },
                   { text: 'Kernel Selection', link: '/en/architecture/kernel-selection' },
                   { text: 'Memory Layout', link: '/en/architecture/memory-layout' },
+                  { text: 'Reliability Constraints', link: '/en/architecture/reliability' },
                   { text: 'Spec-Driven Dev', link: '/en/architecture/spec-driven' }
                 ]
               },
               {
                 text: 'Performance',
                 collapsed: false,
                 items: [
+                  { text: 'Methodology', link: '/en/performance/methodology' },
                   { text: 'Benchmarks', link: '/en/performance/benchmarks' },
                   { text: 'Optimization Guide', link: '/en/performance/optimization-guide' }
                 ]
@@ -210,7 +216,7 @@ export default withMermaid(
             ]
           },
           editLink: {
-            pattern: 'https://github.com/LessUp/gpu-spmv/edit/main/docs/:path',
+            pattern: 'https://github.com/AICL-Lab/gpu-spmv/edit/main/docs/:path',
             text: 'Edit this page on GitHub'
           },
           outline: { label: 'On This Page', level: [2, 3] }
@@ -219,15 +225,18 @@ export default withMermaid(
     },
 
     themeConfig: {
-      logo: '/images/logo.svg',
+      logo: {
+        light: '/images/brand/logo-mark-light.svg',
+        dark: '/images/brand/logo-mark-dark.svg'
+      },
       siteTitle: 'GPU SpMV',
       socialLinks: [
-        { icon: 'github', link: 'https://github.com/LessUp/gpu-spmv' }
+        { icon: 'github', link: 'https://github.com/AICL-Lab/gpu-spmv' }
       ],
       search: { provider: 'local' },
       footer: {
         message: 'MIT License',
-        copyright: '© 2024-2026 LessUp'
+        copyright: '© 2024-2026 AICL-Lab'
       },
       outline: [2, 3]
     },

docs/.vitepress/data/benchmarks.ts

@@ -0,0 +1,20 @@
+export const benchmarkData = {
+  environment: [
+    { label: 'GPU', value: 'RTX 3090' },
+    { label: 'Peak BW', value: '936 GB/s' },
+    { label: 'CUDA', value: '12.0' },
+    { label: 'CPU', value: 'Ryzen 9 5950X' }
+  ],
+  summary: [
+    { label: 'Typical Utilization', value: '70%+' },
+    { label: 'Best Kernel Family', value: 'Merge Path' },
+    { label: 'Best Regular Pattern', value: 'ELL' },
+    { label: 'Selector Accuracy', value: '100%' }
+  ],
+  scenarios: [
+    { label: 'Very sparse', value: 'Scalar CSR', description: 'avg_nnz_per_row < 4' },
+    { label: 'Uniform rows', value: 'Vector CSR', description: 'Low skewness, good warp utilization' },
+    { label: 'High skew', value: 'Merge Path', description: 'Irregular row lengths with better balancing' },
+    { label: 'ELL-friendly', value: 'ELL Kernel', description: 'Uniform row width, coalesced memory access' }
+  ]
+}

docs/.vitepress/data/references.ts

@@ -0,0 +1,48 @@
+export const references = {
+  papers: [
+    {
+      key: 'bell-garland-2009',
+      title: 'Implementing Sparse Matrix-Vector Multiplication on Throughput-Oriented Processors',
+      meta: 'Nathan Bell, Michael Garland · SC 2009',
+      url: 'https://doi.org/10.1145/1654059.1654121'
+    },
+    {
+      key: 'merrill-garland-2016',
+      title: 'Merge-based Parallel Sparse Matrix-Vector Multiplication',
+      meta: 'Duane Merrill, Michael Garland · SC 2016',
+      url: 'https://doi.org/10.1145/3016078.2851141'
+    },
+    {
+      key: 'vazquez-ellrt-2011',
+      title: 'Automatic Tuning of the Sparse Matrix Vector Product on GPUs Based on the ELL-R-T Format',
+      meta: 'Fernando Vázquez et al. · Concurrency and Computation 2011',
+      url: 'https://doi.org/10.1002/cpe.1761'
+    }
+  ],
+  projects: [
+    {
+      key: 'cusparse',
+      title: 'NVIDIA cuSPARSE',
+      meta: 'Vendor baseline and API reference for sparse GPU primitives',
+      url: 'https://developer.nvidia.com/cusparse'
+    },
+    {
+      key: 'ginkgo',
+      title: 'Ginkgo',
+      meta: 'Production-quality sparse linear algebra library with strong docs',
+      url: 'https://github.com/ginkgo-project/ginkgo'
+    },
+    {
+      key: 'moderngpu',
+      title: 'ModernGPU',
+      meta: 'Useful for understanding scan / merge / partitioning techniques on GPU',
+      url: 'https://github.com/moderngpu/moderngpu'
+    },
+    {
+      key: 'suitesparse',
+      title: 'SuiteSparse Matrix Collection',
+      meta: 'Representative real-world sparse matrices for benchmark reasoning',
+      url: 'https://github.com/DrTimothyAldenDavis/SuiteSparse'
+    }
+  ]
+}

docs/.vitepress/data/site.ts

@@ -0,0 +1,29 @@
+export type SiteMetric = {
+  label: string
+  value: string
+  description?: string
+}
+
+export const siteData = {
+  repo: 'https://github.com/AICL-Lab/gpu-spmv',
+  zh: {
+    heroTitle: 'GPU SpMV：技术白皮书与架构展示站',
+    heroLead: '把 CUDA 稀疏矩阵向量乘法项目打造成可读、可证、可展示的工程作品。',
+    metrics: [
+      { label: 'Bandwidth Utilization', value: '70%+' },
+      { label: 'Adaptive Kernels', value: '4' },
+      { label: 'Sparse Formats', value: 'CSR + ELL' },
+      { label: 'Property Tests', value: '100+' }
+    ] satisfies SiteMetric[]
+  },
+  en: {
+    heroTitle: 'GPU SpMV: Technical Whitepaper and Architecture Showcase',
+    heroLead: 'Present the CUDA sparse matrix-vector multiplication project as a serious engineering artifact.',
+    metrics: [
+      { label: 'Bandwidth Utilization', value: '70%+' },
+      { label: 'Adaptive Kernels', value: '4' },
+      { label: 'Sparse Formats', value: 'CSR + ELL' },
+      { label: 'Property Tests', value: '100+' }
+    ] satisfies SiteMetric[]
+  }
+}

docs/.vitepress/theme/Layout.vue

@@ -0,0 +1,13 @@
+<script setup lang="ts">
+import DefaultTheme from 'vitepress/theme'
+
+const { Layout } = DefaultTheme
+</script>
+
+<template>
+  <Layout>
+    <template #layout-top>
+      <div class="spmv-layout-top" aria-hidden="true"></div>
+    </template>
+  </Layout>
+</template>