https://blog.zhenwei.site/gpu/cuda/cuda-cta-%E4%B8%8E-thread-block-cluster-%E5%85%A5%E9%97%A8 2026-06-14T00:00:00.000Z https://blog.zhenwei.site/gpu/cuda/cuda-gemm-%E7%9F%A9%E9%98%B5%E4%B9%98%E6%B3%95%E4%BC%98%E5%8C%96%E6%8C%87%E5%8D%97 2026-06-14T00:00:00.000Z https://blog.zhenwei.site/gpu/cuda/cuda-kernel-%E7%A4%BA%E4%BE%8B%EF%BC%9A%E5%90%91%E9%87%8F%E5%8A%A0%E6%B3%95 2026-05-16T00:00:00.000Z https://blog.zhenwei.site/gpu/cuda/cuda-kernel-%E7%A4%BA%E4%BE%8B%EF%BC%9A%E5%BD%92%E7%BA%A6%E6%B1%82%E5%92%8C 2026-05-20T00:00:00.000Z https://blog.zhenwei.site/gpu/cuda/cuda-kernel-%E7%A4%BA%E4%BE%8B%EF%BC%9A%E7%9F%A9%E9%98%B5%E4%B9%98%E6%B3%95 2026-05-16T00:00:00.000Z https://blog.zhenwei.site/gpu/cuda/cuda-shared-memory-%E4%B8%8E-bank-conflict 2026-05-30T00:00:00.000Z https://blog.zhenwei.site/gpu/cuda/cuda-%E7%BA%BF%E7%A8%8B%E9%85%8D%E7%BD%AE%E4%B8%8E%E5%8D%A0%E7%94%A8%E7%8E%87 2026-05-30T00:00:00.000Z https://blog.zhenwei.site/gpu/cuda/cuda-%E7%BC%96%E7%A8%8B%E5%9F%BA%E7%A1%80 2026-06-14T00:00:00.000Z https://blog.zhenwei.site/gpu/gpu-kernel-%E5%AD%A6%E4%B9%A0%E8%B7%AF%E7%BA%BF 2026-05-19T00:00:00.000Z https://blog.zhenwei.site/gpu/gpu-%E5%88%9D%E5%AD%A6%E8%80%85%E6%9C%AF%E8%AF%AD%E8%A1%A8 2026-06-14T00:00:00.000Z https://blog.zhenwei.site/gpu/gpu-%E5%AD%A6%E4%B9%A0%E4%BB%BB%E5%8A%A1 2026-05-17T00:00:00.000Z https://blog.zhenwei.site/gpu/gpu-%E7%9F%A5%E8%AF%86%E5%BA%93%E7%B4%A2%E5%BC%95 2026-06-14T00:00:00.000Z https://blog.zhenwei.site/gpu/gpu-kernel-learning/readme 2026-05-19T00:00:00.000Z https://blog.zhenwei.site/gpu/gpu-kernel-learning/%E7%8E%AF%E5%A2%83%E9%85%8D%E7%BD%AE%E6%8C%87%E5%8D%97 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/gpu-kernel-learning/%E7%AC%AC1%E7%AB%A0-%E5%9F%BA%E7%A1%80%E6%A6%82%E5%BF%B5 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/gpu-kernel-learning/%E7%AC%AC2%E7%AB%A0-cuda%E5%85%A5%E9%97%A8 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/gpu-kernel-learning/%E7%AC%AC3%E7%AB%A0-gpu%E7%A1%AC%E4%BB%B6 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/gpu-kernel-learning/%E7%AC%AC4%E7%AB%A0-%E4%BC%98%E5%8C%96%E6%8A%80%E5%B7%A7 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/gpu-kernel-learning/%E7%AC%AC5%E7%AB%A0-%E5%9F%BA%E7%A1%80kernel 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/gpu-kernel-learning/%E7%AC%AC6%E7%AB%A0-%E5%A4%A7%E6%A8%A1%E5%9E%8Bkernel 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/gpu-kernel-learning/%E7%AC%AC7%E7%AB%A0-%E9%AB%98%E7%BA%A7%E4%B8%BB%E9%A2%98 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/hardware/blackwell-%E6%9E%B6%E6%9E%84%E6%96%B0%E7%89%B9%E6%80%A7%E4%B8%8E-kernel-%E7%BC%96%E7%A8%8B 2026-06-03T00:00:00.000Z https://blog.zhenwei.site/gpu/hardware/gpu-%E7%A1%AC%E4%BB%B6%E6%9E%B6%E6%9E%84%E8%83%8C%E6%99%AF%E4%B8%8E%E7%BC%96%E7%A8%8B%E8%8C%83%E5%BC%8F 2026-05-30T00:00:00.000Z https://blog.zhenwei.site/gpu/hardware/gpu-%E7%A1%AC%E4%BB%B6%E8%83%8C%E6%99%AF%E5%9C%B0%E5%9B%BE 2026-05-30T00:00:00.000Z https://blog.zhenwei.site/gpu/hardware/gpu-%E9%80%9A%E4%BF%A1%E6%9C%BA%E5%88%B6%E4%B8%8E%E4%BC%98%E5%8C%96%E6%8A%80%E6%9C%AF 2026-06-15T00:00:00.000Z https://blog.zhenwei.site/gpu/hardware/nvidia-gpu-%E6%9E%B6%E6%9E%84%E4%B8%8E%E8%A7%84%E6%A0%BC 2026-06-03T00:00:00.000Z https://blog.zhenwei.site/gpu/profiling/cuda-kernel-%E6%80%A7%E8%83%BD%E7%93%B6%E9%A2%88%E5%AE%9A%E4%BD%8D%E6%B5%81%E7%A8%8B 2026-05-30T00:00:00.000Z https://blog.zhenwei.site/gpu/profiling/ncu_analysis 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/profiling/nsight-compute-ncu-%E5%88%86%E6%9E%90%E6%96%B9%E6%B3%95%E4%B8%8E%E4%BC%98%E5%8C%96%E6%80%9D%E8%B7%AF 2026-05-30T00:00:00.000Z https://blog.zhenwei.site/gpu/profiling/%E5%A4%8D%E6%9D%82-python-%E8%BF%9B%E7%A8%8B%E9%80%89%E6%8B%A9%E6%80%A7-ncu-profiling-%E6%93%8D%E4%BD%9C%E6%89%8B%E5%86%8C 2026-05-30T00:00:00.000Z https://blog.zhenwei.site/gpu/runtime/cuda-jit%E3%80%81aot-%E4%B8%8E-kernel-%E9%80%89%E6%8B%A9%E6%9C%BA%E5%88%B6 2026-05-19T00:00:00.000Z https://blog.zhenwei.site/gpu/runtime/cuda-pdl-programmatic-dependent-launch 2026-05-31T00:00:00.000Z https://blog.zhenwei.site/gpu/runtime/cuda-stream-%E4%B8%8E%E5%BC%82%E6%AD%A5%E6%89%A7%E8%A1%8C 2026-06-14T00:00:00.000Z https://blog.zhenwei.site/gpu/runtime/nvjet_kernel_naming 2026-05-30T00:00:00.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/interactive-demos 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/readme 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/source 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-1-gpu/01-gpu-execution-model 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-1-gpu/02-what-makes-a-kernel-fast 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-1-gpu/03-data-layout-and-its-notation 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-1-gpu/04-the-evolution-of-tensor-core-data-layouts 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-1-gpu/05-async-data-movement---tma 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-1-gpu/06-blackwell-tensor-core---tcgen05-mma 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-1-gpu/07-tensor-memory---tmem 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-1-gpu/08-async-coordination---mbarrier 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-1-gpu/09-advanced-scheduling---cluster-launch-control 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-2-tirx/10-introduction-to-tirx 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-2-tirx/11-tirx-layout-api 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-3-gemm/12-building-a-tiled-gemm 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-3-gemm/13-pipelining-gemm-with-tma 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-3-gemm/14-scaling-gemm-with-warp-specialization-and-clusters 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-4-flash-attention/15-flash-attention-4 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/reference/compiler-internals 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/reference/debugging-warp-specialized-kernels 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/reference/ 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/reference/tirx-language-reference 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/reference/tirx-lowering-pipeline 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/reference/tirx-language/buffers-and-memory 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/reference/tirx-language/cuda-and-ptx-intrinsics 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/reference/tirx-language/control-flow 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/reference/tirx-language/data-types-and-expressions 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/reference/tirx-language/parser-utilities 2026-07-21T09:55:38.000Z https://blog.zhenwei.site/ 2026-07-22T08:51:41.000Z https://blog.zhenwei.site/gpu/cuda/ 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/gpu/ 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/gpu/gpu-kernel-learning/ 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/gpu/hardware/ 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/gpu/profiling/ 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/gpu/runtime/ 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/ 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-1-gpu/ 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-2-tirx/ 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-3-gemm/ 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/part-4-flash-attention/ 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/gpu/modern-gpu-programming-for-mlsys/reference/tirx-language/ 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/gpu-computing 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/cuda-programming 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/concept-note 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/gemm-optimization 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/gpu-programming 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/performance-profiling 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/triton-kernel 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/learning-roadmap 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/index-note 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/reference-note 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/gpu-architecture 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/fp4-quantization 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/nvfp4-quantization 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/architecture-note 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/tensor-core 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/rdma-networking 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/distributed-communication 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/blackwell-gpu 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/hopper-gpu 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/cutlass 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/concurrency-control 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/cublaslt 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/gpu-kernel 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/tutorial-note 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/interactive-visualization 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/source-note 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/blackwell 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/tirx 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/cuda 2026-07-22T08:52:27.504Z https://blog.zhenwei.site/tags/ 2026-07-22T08:52:27.504Z