Fix histc(out=) to reject mismatched out dtype instead of silent cast

src/ATen/native/xpu/SummaryOps.cpp

@@ -55,6 +55,13 @@ Tensor& _histc_out_xpu(
     const Scalar& min,
     const Scalar& max,
     Tensor& result) {
+  TORCH_CHECK(
+      self.dtype() == result.dtype(),
+      "torch.histogram: input tensor and hist tensor should",
+      " have the same dtype, but got input ",
+      self.dtype(),
+      " and hist ",
+      result.dtype());
   auto ret = _histc_xpu(self, bins, min, max);
   at::native::resize_output(result, ret.sizes());
   result.copy_(ret);

test/regressions/test_histc.py

@@ -0,0 +1,35 @@
+# Copyright 2020-2026 Intel Corporation
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+
+# Owner(s): ["module: intel"]
+
+import torch
+from torch.testing._internal.common_utils import run_tests, TestCase
+
+
+class TestHistc(TestCase):
+    def test_histc_out_rejects_mismatched_dtype(self):
+        # Regression for intel/torch-xpu-ops#5237: _histc_out_xpu finished
+        # with result.copy_(ret), so a float histogram written into an
+        # integral out= was silently truncated. CPU rejects this call.
+        x = torch.linspace(1, 8, 8, device="xpu", dtype=torch.float32)
+        out = torch.empty(4, device="xpu", dtype=torch.int64)
+
+        with self.assertRaisesRegex(RuntimeError, "should have the same dtype"):
+            torch.histc(x, bins=4, min=0, max=8, out=out)
+
+    def test_histc_out_matching_dtype_still_works(self):
+        x = torch.linspace(1, 8, 8, device="xpu", dtype=torch.float32)
+        out = torch.empty(0, device="xpu", dtype=torch.float32)
+
+        torch.histc(x, bins=4, min=0, max=8, out=out)
+        self.assertEqual(out, torch.histc(x, bins=4, min=0, max=8))
+
+
+if __name__ == "__main__":
+    run_tests()

test/xpu/skip_list_common.py

@@ -183,7 +183,6 @@
         # Exception: The supported dtypes for linalg.multi_dot on device type xpu are incorrect!
         "test_dtypes_linalg_multi_dot_xpu",
         # For CUDA it's skipped explicitly in common_methods_invocations.py in upstream. We can skip it here
-        "test_out_histc_xpu_float32",
         "test_out_mean_xpu_float32",
         # FakeTensor mismatch in outputs_alias_inputs for aten.view.default
         # Known upstream issue: https://github.com/pytorch/pytorch/issues/159150

test/xpu/xpu_test_utils.py

@@ -1070,6 +1070,23 @@ def gen_xpu_wrappers(op_name, wrappers):
                         else:
                             wrapper.device_type = "xpu"
                             replaced = True
+                    elif (
+                        isinstance(wrapper.device_type, (list, tuple))
+                        and "xpu" in wrapper.device_type
+                        and unittest.expectedFailure in wrapper.decorators
+                        and (op_name, wrapper.test_name) in _cuda_xfail_xpu_pass
+                    ):
+                        # Upstream may scope one xfail to several devices at
+                        # once (device_type=("cuda", "xpu")). Drop XPU from
+                        # the scope so a test that now passes on XPU does
+                        # not report an unexpected success.
+                        replaced = True
+                        new_wrapper = copy.copy(wrapper)
+                        new_wrapper.device_type = tuple(
+                            d for d in wrapper.device_type if d != "xpu"
+                        )
+                        wrapper_xpu.append(new_wrapper)
+                        continue
                     elif (
                         wrapper.device_type is None
                         and unittest.expectedFailure in wrapper.decorators