[Kernel][Test] Extend lightning_attn and awq_triton kernel tests to XPU (#43307)

Signed-off-by: Dobrzyniewicz, Agata <agata.dobrzyniewicz@intel.com>
This commit is contained in:
Agata Dobrzyniewicz
2026-06-04 14:25:59 -04:00
committed by GitHub
parent 439203d32c
commit a947f7a420
3 changed files with 37 additions and 19 deletions
@@ -13,9 +13,15 @@ from vllm.model_executor.layers.quantization.awq_triton import (
awq_dequantize_triton,
awq_gemm_triton,
)
from vllm.platforms import current_platform
from vllm.utils.torch_utils import set_random_seed
device = "cuda"
pytestmark = pytest.mark.skipif(
not (current_platform.is_cuda_alike() or current_platform.is_xpu()),
reason="AWQ Triton kernels require CUDA/ROCm or XPU.",
)
device = current_platform.device_type
def reverse_awq_order(t: torch.Tensor):