Merge pull request #174 from SmallDoges:auto-workflow

Remove CUDA architecture '120' for compatibility

Author: algo-home

.github/workflows/publish.yml

@@ -52,7 +52,7 @@ jobs:
         # Without this we get import error (undefined symbol: _ZN3c105ErrorC2ENS_14SourceLocationESs)
         # when building without C++11 ABI and using it on nvcr images.
         cxx11_abi: ["FALSE", "TRUE"]
-        arch: ["80", "90", "100", "120"]
+        arch: ["80", "90", "100"]
         include:
             - torch-version: "2.9.0.dev20250904"
               cuda-version: "13.0"

setup.py

@@ -79,7 +79,7 @@ def should_skip_cuda_build():
 
 @functools.lru_cache(maxsize=None)
 def cuda_archs():
-    return os.getenv("FLASH_DMATTN_CUDA_ARCHS", "80;90;100;120").split(";")
+    return os.getenv("FLASH_DMATTN_CUDA_ARCHS", "80;90;100").split(";")
 
 
 def detect_preferred_sm_arch() -> Optional[str]: