filter out unsupported cuda versions (#3990)

Author: lanluo-nvidia

.github/scripts/filter-matrix.py

@@ -15,6 +15,8 @@
 jetpack_cuda_versions: List[str] = ["cu126"]
 # rtx 1.2 currently only supports cu129 and cu130
 rtx_cuda_versions: List[str] = ["cu129", "cu130"]
+# trt 10.14.1 currently only supports cu129 and cu130
+trt_cuda_versions: List[str] = ["cu129", "cu130"]
 
 jetpack_container_image: str = "nvcr.io/nvidia/l4t-jetpack:r36.4.0"
 sbsa_container_image: str = "quay.io/pypa/manylinux_2_39_aarch64"
@@ -55,9 +57,11 @@ def filter_matrix_item(
         return False
     else:
         if use_rtx:
-            if item["desired_cuda"] in rtx_cuda_versions:
-                return True
-            return False
+            if item["desired_cuda"] not in rtx_cuda_versions:
+                return False
+        else:
+            if item["desired_cuda"] not in trt_cuda_versions:
+                return False
         if item["gpu_arch_type"] == "cuda-aarch64":
             # pytorch image:pytorch/manylinuxaarch64-builder:cuda12.8 comes with glibc2.28
             # however, TensorRT requires glibc2.31 on aarch64 platform

tests/py/dynamo/models/test_models.py

@@ -426,7 +426,7 @@ def test_resnet18_half(ir):
 
 @pytest.mark.unit
 @unittest.skipIf(
-    torch_trt.ENABLED_FEATURES.tensorrt_rtx,
+    torchtrt.ENABLED_FEATURES.tensorrt_rtx,
     "tensorrt_rtx does not support bfloat16",
 )
 def test_cosmos_true_div(ir):