debug0

Author: zihugithub

.github/workflows/qa-l0-pytorch-wheel.yml

@@ -87,10 +87,6 @@ jobs:
           source .github/workflows/scripts/gpu_check.sh
           wait_for_gpu
 
-          # Additional GPU information print
-          nvidia-smi
-          python3 -c "import torch; print('CUDA Available:', torch.cuda.is_available()); print('GPU Count:', torch.cuda.device_count())"
-
       - name: Run L0 C++ Unit Tests
         # timeout-minutes: 60
         env:

.github/workflows/qa-l0-te-cpp-unittest-pytorch-lint.yml

@@ -112,10 +112,6 @@ jobs:
           source .github/workflows/scripts/gpu_check.sh
           wait_for_gpu
 
-          # Additional GPU information print
-          nvidia-smi
-          python3 -c "import torch; print('CUDA Available:', torch.cuda.is_available()); print('GPU Count:', torch.cuda.device_count())"
-
       - name: Run L1 PyTorch Thunder Integration Tests
         env:
           XML_LOG_DIR: "/logs/pytorch/thunder"

.github/workflows/qa-l1-te-cpp-pytorch-tests.yml

@@ -103,10 +103,6 @@ jobs:
           source .github/workflows/scripts/gpu_check.sh
           wait_for_gpu
 
-          # Additional GPU information print
-          nvidia-smi
-          python3 -c "import torch; print('CUDA Available:', torch.cuda.is_available()); print('GPU Count:', torch.cuda.device_count())"
-
       - name: Run QA L3 PyTorch FlashAttention Versions Test
         # timeout-minutes: 30
         env:

.github/workflows/qa-l3-te-pytorch-fa-versions-test.yml

@@ -64,4 +64,4 @@ wait_for_gpu() {
         echo "Note: If you are using a new chip type, please add GPU idle detection method for your chip"
         exit 1
     fi
-}
+}

.github/workflows/scripts/gpu_check.sh

@@ -94,10 +94,6 @@ jobs:
           source .github/workflows/scripts/gpu_check.sh
           wait_for_gpu
 
-          # Additional GPU information print
-          nvidia-smi
-          python3 -c "import torch; print('CUDA Available:', torch.cuda.is_available()); print('GPU Count:', torch.cuda.device_count())"
-
       - name: Plugin Test
         timeout-minutes: 30
         run: |

.github/workflows/te-plugin-tests.yml

@@ -47,16 +47,14 @@
 def generate_build_config(skip_cuda_build):
     """Generate build-time configuration file."""
     config_template_path = (
-        current_file_path / "transformer_engine" / "plugin" /
-        "core" / "_build_config.py.template"
+        current_file_path / "transformer_engine" / "plugin" / "core" / "_build_config.py.template"
     )
     config_output_path = (
-        current_file_path / "transformer_engine" / "plugin" /
-        "core" / "_build_config.py"
+        current_file_path / "transformer_engine" / "plugin" / "core" / "_build_config.py"
     )
 
     if config_template_path.exists():
-        with open(config_template_path, 'r') as f:
+        with open(config_template_path, "r") as f:
             template = f.read()
 
         config_content = template.format(
@@ -65,7 +63,7 @@ def generate_build_config(skip_cuda_build):
             platform=platform.platform(),
         )
 
-        with open(config_output_path, 'w') as f:
+        with open(config_output_path, "w") as f:
             f.write(config_content)
 
         print(f"Generated build config: {config_output_path}")
@@ -77,7 +75,7 @@ def generate_build_config(skip_cuda_build):
 BUILD_TIME = "{datetime.now().isoformat()}"
 BUILD_PLATFORM = "{platform.platform()}"
 """
-        with open(config_output_path, 'w') as f:
+        with open(config_output_path, "w") as f:
             f.write(config_content)
         print(f"Generated minimal build config: {config_output_path}")
 
@@ -86,7 +84,7 @@ class CustomInstall(InstallCommand):
     """Custom install command to generate build config."""
 
     user_options = InstallCommand.user_options + [
-        ('skip-cuda-build', None, 'Skip CUDA build'),
+        ("skip-cuda-build", None, "Skip CUDA build"),
     ]
 
     def initialize_options(self):

setup.py

@@ -31,18 +31,21 @@ def skip_cuda_build() -> bool:
     # Fall back to build-time configuration
     try:
         from transformer_engine.plugin.core._build_config import SKIP_CUDA_BUILD
+
         return SKIP_CUDA_BUILD
     except ImportError:
         # If build config doesn't exist, default to False
         return False
 
+
 # Load plugin system - this handles module registration and backend initialization
 # The _module_setup inside core will:
 # 1. Register modules under both full and short names for relative imports
 # 2. Load all available backends (flagos, reference, vendor/cuda, etc.)
 # 3. Register transformer_engine_torch module from the selected backend
 import transformer_engine.plugin.core  # noqa: F401
 
+
 @functools.lru_cache(maxsize=None)
 def _is_package_installed(package) -> bool:
     """Check if the given package is installed via pip."""

transformer_engine/common/__init__.py

@@ -9,11 +9,13 @@
     get_registry,
 )
 
+
 def __getattr__(name):
     if name == "tefl":
         return _get_tefl_module()
     raise AttributeError(f"module {__name__!r} has no attribute {name!r}")
 
+
 __all__ = [
     "TEFLBackendBase",
     "TEFLModule",