fix(musa): support float64 in-place log

README.md

@@ -66,6 +66,7 @@ That's it! Supported `torch.cuda.*` APIs are automatically redirected to `torch.
 | C++ nvJPEG porting | nvJPEG source and build settings → MTJPEG |
 | ctypes Libraries | `ctypes.CDLL` with CUDA function names → MUSA equivalents |
 | Unified Accelerator API | `torch.accelerator.empty_cache()`, `memory_stats()`, `Stream`, `Event`, ... |
+| MUSA float64 in-place log | On `torch_musa < 2.11.0.post2`, `Tensor.log_()` reuses the supported out-of-place operation while preserving the in-place contract |
 | Triton CUDA Extra | `tl.extra.cuda` → `tl.extra.musa` compatibility on MUSA |
 | Triton Fused MoE | Triton 3.2.0 MTT S5000 tuning configs for vLLM and SGLang |
 
@@ -392,7 +393,7 @@ See `src/torchada/_mappings/` for 400+ mapping rules grouped by API domain.
 
 ```
 # pyproject.toml or requirements.txt
-torchada>=0.1.86
+torchada>=0.1.87
 ```
 
 ### Step 2: Conditional Import

README_CN.md

@@ -66,6 +66,7 @@ torch.cuda.synchronize()
 | C++ nvJPEG 移植 | nvJPEG 源码及构建配置 → MTJPEG |
 | ctypes 库加载 | `ctypes.CDLL` 使用 CUDA 函数名 → 自动转换为 MUSA |
 | 统一加速器 API | `torch.accelerator.empty_cache()`、`memory_stats()`、`Stream`、`Event` 等 |
+| MUSA float64 原地对数 | `torch_musa < 2.11.0.post2` 时，`Tensor.log_()` 复用受支持的非原地操作，同时保持原地操作契约 |
 | Triton CUDA Extra | MUSA 上的 `tl.extra.cuda` → `tl.extra.musa` 兼容 |
 | Triton 融合 MoE | 面向 vLLM 和 SGLang 的 Triton 3.2.0 MTT S5000 调优配置 |
 
@@ -376,7 +377,7 @@ if torchada.is_gpu_device(device):  # 在 CUDA 和 MUSA 上都能工作
 
 ```
 # pyproject.toml 或 requirements.txt
-torchada>=0.1.86
+torchada>=0.1.87
 ```
 
 ### 步骤 2：条件导入

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "torchada"
-version = "0.1.86"
+version = "0.1.87"
 description = "Adapter package for torch_musa to act exactly like PyTorch CUDA"
 readme = "README.md"
 license = {text = "MIT"}

src/torchada/__init__.py

@@ -24,7 +24,7 @@
     from torch.utils.cpp_extension import CUDAExtension, BuildExtension, CUDA_HOME
 """
 
-__version__ = "0.1.86"
+__version__ = "0.1.87"
 
 from . import cuda, utils
 

src/torchada/_patch.py

@@ -40,6 +40,7 @@
 
 _patched = False
 _original_init_process_group = None
+_original_tensor_log_ = None
 
 # Registry for patch functions
 _patch_registry: List[Callable[[], None]] = []
@@ -160,6 +161,28 @@ def _patch_inductor_template_heuristics():
             heuristic_cache.clear()
 
 
+@patch_function
+@requires_import("torch_musa")
+def _patch_tensor_log_():
+    """Backport MUSA float64 ``Tensor.log_`` for torch_musa < 2.11.0.post2."""
+    global _original_tensor_log_
+
+    if not is_musa_platform() or _original_tensor_log_ is not None:
+        return
+    if not _is_pre_torch_musa_2_11_0_post2(torch.musa.__version__):
+        return
+
+    _original_tensor_log_ = torch.Tensor.log_
+
+    @functools.wraps(_original_tensor_log_)
+    def patched_log_(self):
+        if self.device.type == "musa" and self.dtype == torch.float64:
+            return self.copy_(torch.log(self))
+        return _original_tensor_log_(self)
+
+    torch.Tensor.log_ = patched_log_
+
+
 # Cache for translated device strings - avoids repeated string operations
 _device_str_cache = {}
 
@@ -1780,8 +1803,9 @@ def __getitem__(self, name: str):
 def _is_pre_torch_musa_2_11_0_post2(version) -> bool:
     """Return whether the torch_musa version predates 2.11.0.post2.
 
-    torch_musa 2.11.0.post2 fixes the unified accelerator memory APIs. Older
-    releases still need torchada to force those calls through torch.musa.
+    torch_musa 2.11.0.post2 fixes the unified accelerator memory APIs and the
+    float64 in-place ``Tensor.log_``. Older releases still need torchada to
+    force those memory calls through torch.musa and to backport the log path.
     Ignore the local version suffix (for example ``+musa5.2.0``), because it
     identifies the MUSA stack build rather than the torch_musa fix level.
 

tests/test_cuda_patching.py

@@ -3034,6 +3034,102 @@ def invalid_lse_result(
             flash_attn._flash_attn_forward("q", "k", "v")
 
 
+class TestTensorLogPatch:
+    """CPU coverage for the MUSA float64 Tensor.log_ compatibility patch."""
+
+    def test_float64_log_patch_respects_torch_musa_version(self, monkeypatch):
+        import sys
+        from types import ModuleType, SimpleNamespace
+
+        import torch
+
+        from torchada import _patch
+
+        original_log = torch.Tensor.log_
+        monkeypatch.setattr(torch.Tensor, "log_", original_log)
+        monkeypatch.setitem(sys.modules, "torch_musa", ModuleType("torch_musa"))
+        monkeypatch.setattr(_patch, "is_musa_platform", lambda: True)
+        monkeypatch.setattr(_patch, "_original_tensor_log_", None)
+        monkeypatch.setattr(
+            torch,
+            "musa",
+            SimpleNamespace(__version__="2.11.0.post2"),
+            raising=False,
+        )
+
+        _patch._patch_tensor_log_()
+        assert torch.Tensor.log_ is original_log
+
+        torch.musa.__version__ = "2.11.0.post1+musa5.2.0"
+        _patch._patch_tensor_log_()
+        assert torch.Tensor.log_ is not original_log
+
+    def test_float64_log_patch_uses_out_of_place_copy_on_musa(self, monkeypatch):
+        import sys
+        from types import ModuleType, SimpleNamespace
+
+        import torch
+
+        from torchada import _patch
+
+        original_log = torch.Tensor.log_
+        monkeypatch.setattr(torch.Tensor, "log_", original_log)
+        monkeypatch.setitem(sys.modules, "torch_musa", ModuleType("torch_musa"))
+        monkeypatch.setattr(_patch, "is_musa_platform", lambda: True)
+        monkeypatch.setattr(_patch, "_original_tensor_log_", None)
+        monkeypatch.setattr(
+            torch,
+            "musa",
+            SimpleNamespace(__version__="2.11.0.post1+musa5.2.0"),
+            raising=False,
+        )
+
+        _patch._patch_tensor_log_()
+        patched_log = torch.Tensor.log_
+
+        log_args = []
+        logged = object()
+
+        def fake_torch_log(tensor):
+            log_args.append(tensor)
+            return logged
+
+        original_calls = []
+
+        def fake_original_log_(tensor):
+            original_calls.append(tensor)
+            return tensor
+
+        monkeypatch.setattr(torch, "log", fake_torch_log)
+        monkeypatch.setattr(_patch, "_original_tensor_log_", fake_original_log_)
+
+        class FakeTensor:
+            def __init__(self, device_type, dtype):
+                self.device = SimpleNamespace(type=device_type)
+                self.dtype = dtype
+                self.copied = None
+
+            def copy_(self, other):
+                self.copied = other
+                return self
+
+        musa_f64 = FakeTensor("musa", torch.float64)
+        result = patched_log(musa_f64)
+        assert result is musa_f64
+        assert musa_f64.copied is logged
+        assert log_args == [musa_f64]
+        assert original_calls == []
+
+        cpu_f64 = FakeTensor("cpu", torch.float64)
+        assert patched_log(cpu_f64) is cpu_f64
+        assert cpu_f64.copied is None
+
+        musa_f32 = FakeTensor("musa", torch.float32)
+        assert patched_log(musa_f32) is musa_f32
+        assert musa_f32.copied is None
+        assert original_calls == [cpu_f64, musa_f32]
+
+
 class TestAcceleratorModuleWrapper:
     """Test the _AcceleratorModuleWrapper priority / fallback logic in isolation.
 

tests/test_platform.py

@@ -59,7 +59,7 @@ def test_get_version(self):
 
         version = torchada.get_version()
         assert version == torchada.__version__
-        assert version == "0.1.86"
+        assert version == "0.1.87"
         assert isinstance(version, str)
 
     def test_project_version_matches_runtime_version(self):