DeepSeek/DeepGEMM
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepGEMM synced 2025-06-26 23:15:49 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Fix 12.9 compatibility

Browse Source
This commit is contained in:
Chenggang Zhao
2025-05-07 13:23:40 +08:00
parent 085b4a1532
commit 8702f910e3
4 changed files with 24 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
deep_gemm/jit/runtime.py
View File

@@ -48,7 +48,8 @@ class Runtime:
command = [f'{CUDA_HOME}/bin/cuobjdump', '-symbols', path]
result = subprocess.run(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
assert result.returncode == 0
kernel_names = [line.split()[-1] for line in result.stdout.splitlines() if line.startswith('STT_FUNC')]
kernel_names = [line.split()[-1] for line in result.stdout.splitlines()
if line.startswith('STT_FUNC') and '__instantiate_kernel' not in line]
assert len(kernel_names) == 1, f'Too many kernels in the library: {kernel_names}'
# Load kernel from the library