Added xfail for expected failures

thunder/tests/framework.py

@@ -77,6 +77,43 @@ def _bitsandbytes_available():
 BITSANDBYTES_AVAILABLE = _bitsandbytes_available()
 
 
+# TODO This change should be handled properly in the future
+def _pytorch_removed_args_tensor_mask() -> bool:
+    """Check if PyTorch removed args_tensor_mask from autograd_function_apply.
+
+    PyTorch removed args_tensor_mask in https://github.com/pytorch/pytorch/pull/166788
+    (commit 5cf15aef144fd03379a5796e37698eee5e4575b8, Dec 12, 2025).
+
+    Returns True if args_tensor_mask is NO LONGER accepted (new PyTorch).
+    """
+    import re
+
+    version = torch.__version__
+    # Nightly versions look like: 2.6.0.dev20251212+cpu
+    match = re.search(r"\.dev(\d{8})", version)
+    if match:
+        nightly_date = int(match.group(1))
+        # args_tensor_mask removed around Dec 12, 2025
+        return nightly_date >= 20251212
+
+    # Lightning AI nightly builds have version strings like: 2.10.0a0+git62c80e7
+    # We conservatively check against the base version (the portion before "+")
+    # because we do not know whether the problematic commit removing args_tensor_mask
+    # is before or after 2.10.0a0+git62c80e7. Therefore, we return True (masked as removed)
+    # if we are on version 2.10.0a0 or later.
+    base_version = packaging.version.parse(version.split("+")[0])
+    if base_version >= packaging.version.parse("2.10.0a0"):
+        return True
+
+    return False
+
+
+xfail_if_args_tensor_mask_removed = pytest.mark.xfail(
+    _pytorch_removed_args_tensor_mask(),
+    reason="PyTorch >= 2.10.0a0+git62c80e7 or nightly >= 20251212 removed args_tensor_mask from autograd_function_apply (PR #166788)",
+)
+
+
 def version_between(version: str, *, min_ver: str | None = None, max_ver: str | None = None):
     v = packaging.version.parse(version)
     if min_ver is not None and v < packaging.version.parse(min_ver):

thunder/tests/test_dynamo.py

@@ -32,6 +32,7 @@
     DynamoThunderExecutor,
     IS_WINDOWS,
     requiresCUDA,
+    xfail_if_args_tensor_mask_removed,
 )
 from thunder.tests.make_tensor import make_tensor
 from thunder.dynamo.report import (
@@ -369,6 +370,7 @@ def func(x):
             strict=True,
             reason="torch.compile Windows support is still WIP - https://github.com/pytorch/pytorch/issues/122094",
         ),
+        xfail_if_args_tensor_mask_removed,
     ),
 )
 def test_splitter_autograd_function(executor, device: str, dtype: dtypes.dtype, dynamic: bool | None):
@@ -1631,6 +1633,7 @@ def compile(self, fn, **kwargs):
 
 
 @requiresCUDA
+@xfail_if_args_tensor_mask_removed
 def test_autograd_function_fx_report(tmp_path):
     class Sin(torch.autograd.Function):
         @staticmethod

thunder/tests/test_jit_general.py

@@ -13,7 +13,7 @@
 
 import thunder
 
-from thunder.tests.framework import requiresCUDA, IS_WINDOWS
+from thunder.tests.framework import requiresCUDA, IS_WINDOWS, xfail_if_args_tensor_mask_removed
 from thunder.core.options import CACHE_OPTIONS
 import thunder.core.prims as prims
 from thunder import pytorch_executor, nvfuser_executor
@@ -1252,6 +1252,7 @@ def f(x):
 
 
 @pytest.mark.filterwarnings("ignore:Please use torch.vmap")
+@xfail_if_args_tensor_mask_removed
 def test_autograd_function_apply():
     # see https://github.com/Lightning-AI/lightning-thunder/issues/1248#issuecomment-2388655917
     # for why `torch.foo` instead of `torch.Tensor.foo`
@@ -1328,6 +1329,7 @@ def my_sin_with_wrong_backward(x):
         gradcheck(jitted, (x,))
 
 
+@xfail_if_args_tensor_mask_removed
 def test_autograd_function_apply_with_no_grad():
     # This case is using `torch` operations
     def forward(_, x):

thunder/tests/test_update_aliases.py

@@ -21,6 +21,7 @@
     TorchCompileExecutor,
     nvFuserExecutor,
     requiresCUDA,
+    xfail_if_args_tensor_mask_removed,
 )
 from thunder.torch import _torch_to_thunder_function_map, _inplace_to_out_of_place
 
@@ -477,6 +478,7 @@ def f(a):
 
 @instantiate(
     dtypes=(dtypes.float32,),
+    decorators=(xfail_if_args_tensor_mask_removed,),
 )
 def test_higher_order_inplace_alias_update(executor, device, dtype):
     torch_dtype = dtypes.to_torch_dtype(dtype)