A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 10, 20:15 UTC
[Bugfix] Replace BaseException with specific exceptions in FLA utils (#31590)
Signed-off-by: c0de128 <kevin.mckay@outlook.com>
Kevin McKay committed
Jan 1, 2026 at 21:27 UTC
5fff44064b38012d5675d54620dc3dc3c28eb522
1 file changed
+1
-1
vllm/model_executor/layers/fla/ops/utils.py
+1
-1
@@ -119,7 +119,7 @@ def input_guard(fn: Callable[..., torch.Tensor]) -> Callable[..., torch.Tensor]:
119
def get_available_device() -> str:
120
try:
121
return triton.runtime.driver.active.get_current_target().backend
122
- except BaseException:
122
+ except (RuntimeError, AttributeError):
123
return "cpu"
124
125