[PR #119] [CLOSED] Add initial support for Nvidia Blackwell (SM120) #255

New issue

Closed

opened 2026-10-26 21:15:24 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-26 21:15:24 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/deepseek-ai/DeepGEMM/pull/119
Author: @celsowm
Created: 6/24/2025
Status: ❌ Closed

Base: main ← Head: feat/blackwell-sm120-support

📝 Commits (1)

93ea479 Add initial support for Nvidia Blackwell (SM120)

📊 Changes

2 files changed (+7 additions, -3 deletions)

View changed files

📝 CMakeLists.txt (+5 -2)
📝 deep_gemm/jit/compiler.py (+2 -1)

📄 Description

This change introduces the necessary compiler flags and CMake configurations to enable support for the Nvidia Blackwell SM120 architecture.

Modified deep_gemm/jit/compiler.py to include sm_120 and compute_120 flags for NVCC and NVRTC.
Updated CMakeLists.txt to add the new architecture flags for the build process.

Further testing on Blackwell hardware is required to validate MMA instruction compatibility and overall performance.

_{🔄 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/119 **Author:** [@celsowm](https://github.com/celsowm) **Created:** 6/24/2025 **Status:** ❌ Closed **Base:** `main` ← **Head:** `feat/blackwell-sm120-support` --- ### 📝 Commits (1) - [`93ea479`](https://github.com/deepseek-ai/DeepGEMM/commit/93ea4797c071aa61a42053b78d554e8031dd5928) Add initial support for Nvidia Blackwell (SM120) ### 📊 Changes **2 files changed** (+7 additions, -3 deletions) <details> <summary>View changed files</summary> 📝 `CMakeLists.txt` (+5 -2) 📝 `deep_gemm/jit/compiler.py` (+2 -1) </details> ### 📄 Description This change introduces the necessary compiler flags and CMake configurations to enable support for the Nvidia Blackwell SM120 architecture. - Modified deep_gemm/jit/compiler.py to include sm_120 and compute_120 flags for NVCC and NVRTC. - Updated CMakeLists.txt to add the new architecture flags for the build process. Further testing on Blackwell hardware is required to validate MMA instruction compatibility and overall performance. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-10-26 21:15:24 +00:00