[diffusion] CI: skip the LTX-2 RMSNorm-modulation guard test off the CUDA platform (#43574)

Co-authored-by: Mick <mickqian@radixark.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
This commit is contained in:
Mick
2026-10-11 08:54:37 +08:00
committed by GitHub
co-authored by Mick Claude Opus 5.5
parent 65386cb95a
commit 3022989bad
@@ -21,9 +21,11 @@ import pytest
import torch
from sglang.kernels.ops.diffusion import can_use_fused_rmsnorm_modulation
from sglang.kernels.ops.diffusion.common.platform import is_cuda
@unittest.skipUnless(torch.cuda.is_available(), "needs CUDA")
# Not torch.cuda.is_available(): ROCm answers it too, and has no such kernel.
@unittest.skipUnless(is_cuda(), "the fused kernel is CUDA-only")
class TestFusedRmsNormModulationGuard(unittest.TestCase):
"""The guard answers exactly the question the launcher would raise on."""