[PR #280] [MERGED] Multiple updates and refactorings #340

New issue

Closed

opened 2026-10-26 21:16:27 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-26 21:16:27 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/DeepGEMM/pull/280
Author: @zheanxu
Created: 1/16/2026
Status: ✅ Merged
Merged: 1/16/2026
Merged by: @zheanxu

Base: main ← Head: public-release-2601

📝 Commits (1)

8aaed11 Multiple updates and refactorings

📊 Changes

55 files changed (+2705 additions, -890 deletions)

View changed files

📝 csrc/apis/attention.hpp (+12 -10)
📝 csrc/apis/einsum.hpp (+9 -7)
📝 csrc/apis/gemm.hpp (+199 -142)
➕ csrc/apis/hyperconnection.hpp (+70 -0)
📝 csrc/apis/layout.hpp (+51 -16)
📝 csrc/apis/runtime.hpp (+4 -1)
📝 csrc/indexing/main.cu (+4 -0)
📝 csrc/jit/compiler.hpp (+72 -10)
📝 csrc/jit/device_runtime.hpp (+1 -14)
📝 csrc/jit/handle.hpp (+0 -4)
📝 csrc/jit_kernels/heuristics/common.hpp (+46 -20)
📝 csrc/jit_kernels/heuristics/sm100.hpp (+11 -15)
📝 csrc/jit_kernels/heuristics/sm90.hpp (+5 -9)
📝 csrc/jit_kernels/impls/runtime_utils.hpp (+19 -7)
📝 csrc/jit_kernels/impls/sm100_bf16_gemm.hpp (+29 -17)
📝 csrc/jit_kernels/impls/sm100_fp8_gemm_1d1d.hpp (+101 -73)
➕ csrc/jit_kernels/impls/sm100_tf32_hc_prenorm_gemm.hpp (+149 -0)
📝 csrc/jit_kernels/impls/sm90_bf16_gemm.hpp (+13 -10)
📝 csrc/jit_kernels/impls/sm90_fp8_gemm_1d1d.hpp (+5 -3)
📝 csrc/jit_kernels/impls/sm90_fp8_gemm_1d2d.hpp (+76 -9)

...and 35 more files

📄 Description

No description provided

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/deepseek-ai/DeepGEMM/pull/280 **Author:** [@zheanxu](https://github.com/zheanxu) **Created:** 1/16/2026 **Status:** ✅ Merged **Merged:** 1/16/2026 **Merged by:** [@zheanxu](https://github.com/zheanxu) **Base:** `main` ← **Head:** `public-release-2601` --- ### 📝 Commits (1) - [`8aaed11`](https://github.com/deepseek-ai/DeepGEMM/commit/8aaed1168d111703e0d1f8ff0fc1769454acc7b0) Multiple updates and refactorings ### 📊 Changes **55 files changed** (+2705 additions, -890 deletions) <details> <summary>View changed files</summary> 📝 `csrc/apis/attention.hpp` (+12 -10) 📝 `csrc/apis/einsum.hpp` (+9 -7) 📝 `csrc/apis/gemm.hpp` (+199 -142) ➕ `csrc/apis/hyperconnection.hpp` (+70 -0) 📝 `csrc/apis/layout.hpp` (+51 -16) 📝 `csrc/apis/runtime.hpp` (+4 -1) 📝 `csrc/indexing/main.cu` (+4 -0) 📝 `csrc/jit/compiler.hpp` (+72 -10) 📝 `csrc/jit/device_runtime.hpp` (+1 -14) 📝 `csrc/jit/handle.hpp` (+0 -4) 📝 `csrc/jit_kernels/heuristics/common.hpp` (+46 -20) 📝 `csrc/jit_kernels/heuristics/sm100.hpp` (+11 -15) 📝 `csrc/jit_kernels/heuristics/sm90.hpp` (+5 -9) 📝 `csrc/jit_kernels/impls/runtime_utils.hpp` (+19 -7) 📝 `csrc/jit_kernels/impls/sm100_bf16_gemm.hpp` (+29 -17) 📝 `csrc/jit_kernels/impls/sm100_fp8_gemm_1d1d.hpp` (+101 -73) ➕ `csrc/jit_kernels/impls/sm100_tf32_hc_prenorm_gemm.hpp` (+149 -0) 📝 `csrc/jit_kernels/impls/sm90_bf16_gemm.hpp` (+13 -10) 📝 `csrc/jit_kernels/impls/sm90_fp8_gemm_1d1d.hpp` (+5 -3) 📝 `csrc/jit_kernels/impls/sm90_fp8_gemm_1d2d.hpp` (+76 -9) _...and 35 more files_ </details> ### 📄 Description _No description provided_ --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-10-26 21:16:27 +00:00