[PR #100] [MERGED] Refactor some launch-related structures #247

New issue

Closed

opened 2026-10-26 21:15:17 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-26 21:15:17 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/DeepGEMM/pull/100
Author: @LyricZhao
Created: 5/15/2025
Status: ✅ Merged
Merged: 5/15/2025
Merged by: @LyricZhao

Base: main ← Head: remove-tuner

📝 Commits (5)

816b390 Refactor launch-related structures
4373af2 Add DG_PRINT_CONFIGS
350989e Unify ceil_divs
3b412f4 Unify kwargs usages
104a6ec Add __assertfail

📊 Changes

11 files changed (+259 additions, -447 deletions)

View changed files

📝 README.md (+1 -1)
📝 deep_gemm/include/deep_gemm/fp8_wgrad_gemm.cuh (+4 -4)
📝 deep_gemm/jit/compiler.py (+7 -7)
📝 deep_gemm/jit/runtime.py (+21 -9)
📝 deep_gemm/jit_kernels/gemm.py (+26 -44)
📝 deep_gemm/jit_kernels/m_grouped_gemm.py (+47 -63)
📝 deep_gemm/jit_kernels/runtime.py (+83 -140)
➖ deep_gemm/jit_kernels/tuner.py (+0 -82)
📝 deep_gemm/jit_kernels/wgrad_gemm.py (+53 -74)
📝 tests/test_core.py (+4 -4)
📝 tests/test_jit.py (+13 -19)

📄 Description

Remove tuner
Refactor launch arguments
Refactor heuristic config prints
Remove Layout
Some cleanup and typo fixes

_{🔄 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/100 **Author:** [@LyricZhao](https://github.com/LyricZhao) **Created:** 5/15/2025 **Status:** ✅ Merged **Merged:** 5/15/2025 **Merged by:** [@LyricZhao](https://github.com/LyricZhao) **Base:** `main` ← **Head:** `remove-tuner` --- ### 📝 Commits (5) - [`816b390`](https://github.com/deepseek-ai/DeepGEMM/commit/816b39053adb1921e6517c95847e14a4cc2eddbd) Refactor launch-related structures - [`4373af2`](https://github.com/deepseek-ai/DeepGEMM/commit/4373af2e82ab083f1f6f50bd8add978fee7f23e6) Add `DG_PRINT_CONFIGS` - [`350989e`](https://github.com/deepseek-ai/DeepGEMM/commit/350989eef34bc0e48a33a7fcae4656ebc53fe4c0) Unify `ceil_div`s - [`3b412f4`](https://github.com/deepseek-ai/DeepGEMM/commit/3b412f458a4e2cf816fc51d8d3a7e1a8549f915c) Unify `kwargs` usages - [`104a6ec`](https://github.com/deepseek-ai/DeepGEMM/commit/104a6ec109d8fe5e39141d19639c4f2067f506fb) Add `__assertfail` ### 📊 Changes **11 files changed** (+259 additions, -447 deletions) <details> <summary>View changed files</summary> 📝 `README.md` (+1 -1) 📝 `deep_gemm/include/deep_gemm/fp8_wgrad_gemm.cuh` (+4 -4) 📝 `deep_gemm/jit/compiler.py` (+7 -7) 📝 `deep_gemm/jit/runtime.py` (+21 -9) 📝 `deep_gemm/jit_kernels/gemm.py` (+26 -44) 📝 `deep_gemm/jit_kernels/m_grouped_gemm.py` (+47 -63) 📝 `deep_gemm/jit_kernels/runtime.py` (+83 -140) ➖ `deep_gemm/jit_kernels/tuner.py` (+0 -82) 📝 `deep_gemm/jit_kernels/wgrad_gemm.py` (+53 -74) 📝 `tests/test_core.py` (+4 -4) 📝 `tests/test_jit.py` (+13 -19) </details> ### 📄 Description - Remove tuner - Refactor launch arguments - Refactor heuristic config prints - Remove `Layout` - Some cleanup and typo fixes --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-10-26 21:15:17 +00:00