various fixes

Author: drisspg

test/prototype/mx_formats/test_mx_linear.py

@@ -381,15 +381,16 @@ def test_inference_print_str():
 )
 @pytest.mark.skipif(not is_sm_at_least_100, reason="Reqs sm100")
 @pytest.mark.parametrize("elem_dtype", [torch.float8_e4m3fn])
-def test_inference_subclass(elem_dtype):
+@pytest.mark.parametrize("bias", [True, False])
+def test_inference_subclass(elem_dtype, bias: bool):
     """
     Smoke test for inference compile
     """
     if elem_dtype in (torch.float8_e4m3fn, torch.float8_e5m2):
         if not is_sm_at_least_89():
             pytest.skip("CUDA capability >= 8.9 required for float8 in triton")
 
-    m = nn.Sequential(nn.Linear(32, 128, bias=False, dtype=torch.bfloat16))
+    m = nn.Sequential(nn.Linear(32, 128, bias=bias, dtype=torch.bfloat16))
     m = m.cuda()
     m_mx = copy.deepcopy(m)
     config = MXFPConfig()
@@ -401,6 +402,8 @@ def test_inference_subclass(elem_dtype):
     y_mx = m_mx(x)
     sqnr = compute_error(y_ref, y_mx)
     if elem_dtype is torch.float8_e4m3fn:
-        assert sqnr >= 20.0
+        assert (
+            sqnr >= 20.0 if not bias else 14
+        )  # TODO see if this is working as expected
     else:
         assert sqnr >= 11.5

torchao/core/config.py

@@ -175,6 +175,7 @@ def config_to_dict(config: AOBaseConfig) -> Dict[str, Any]:
     "torchao.quantization",
     "torchao.sparsity.sparse_api",
     "torchao.prototype.quantization",
+    "torchao.prototype.mx_formats",
 }
 
 

torchao/prototype/mx_formats/__init__.py

@@ -4,6 +4,7 @@
     MXLinearConfig,
     MXLinearRecipeName,
 )
+from torchao.prototype.mx_formats.mx_subclass import MXFPConfig
 
 # import mx_linear here to register the quantize_ transform logic
 # ruff: noqa: I001
@@ -14,4 +15,5 @@
     "MXInferenceLinearConfig",
     "MXLinearConfig",
     "MXLinearRecipeName",
+    "MXFPConfig",
 ]

torchao/prototype/mx_formats/mx_ops.py

@@ -20,6 +20,9 @@
 from typing import Any, Dict, Optional
 
 import torch
+from torch.utils._python_dispatch import (
+    return_and_correct_aliasing,
+)
 from torch.utils._pytree import tree_map
 
 import torchao.ops
@@ -52,21 +55,28 @@ def decorator(func):
     return decorator
 
 
-@implements([aten.detach.default])
-def mx_desugar_op(aten_op, args, kwargs=None):
-    old = args[0]
-    new_data = aten_op(old._data, *args[1:], **kwargs)
-    new = MXTensor(
-        old._scale_e8m0,
-        new_data,
-        old._elem_dtype,
-        old._block_size,
-        old._orig_dtype,
-        old._use_fp4_custom_triton_dequant_kernel,
-        old._gemm_kernel_choice,
-        old._pack_fp6,
+# @implements([aten.detach.default])
+# def mx_desugar_op(aten_op, args, kwargs=None):
+#     old = args[0]
+#     new_data = aten_op(old._data, *args[1:], **kwargs)
+#     new = MXTensor(
+#         old._scale_e8m0,
+#         new_data,
+#         old._elem_dtype,
+#         old._block_size,
+#         old._orig_dtype,
+#         old._use_fp4_custom_triton_dequant_kernel,
+#         old._gemm_kernel_choice,
+#         old._pack_fp6,
+#     )
+#     return new
+
+
+@implements([aten.detach.default, aten.alias.default])
+def _(func, types, args, kwargs):
+    return return_and_correct_aliasing(
+        func, args, kwargs, args[0]._apply_fn_to_data(func)
     )
-    return new
 
 
 def _get_gemm_choice(
@@ -85,12 +95,15 @@ def _get_gemm_choice(
     return choice_a if choice_a is not None else choice_b
 
 
-@implements([aten.mm.default, aten.matmul.default])
-def mx_mm(aten_op, args, kwargs=None):
-    a = args[0]
-    b = args[1]
-    assert isinstance(a, MXTensor) and isinstance(b, MXTensor)
+def _addmm_mx_dispatch(
+    a: MXTensor, b: MXTensor, aten_op, bias: Optional[torch.Tensor] = None
+) -> torch.Tensor:
+    """
+    Core implementation shared between mx_mm and mx_addmm.
+    The only difference is whether bias is None or not.
+    """
     gemm_choice = _get_gemm_choice(a._gemm_kernel_choice, b._gemm_kernel_choice)
+
     if gemm_choice in (MXGemmKernelChoice.CUBLAS, MXGemmKernelChoice.CUTLASS):
         # real MX gemm backed by torchao's CUTLASS kernels
         M, K, N = a.shape[0], a.shape[1], b.shape[1]
@@ -112,28 +125,63 @@ def mx_mm(aten_op, args, kwargs=None):
                 b._data,
                 a_scale_block.view(torch.float8_e8m0fnu),
                 b_scale_block.view(torch.float8_e8m0fnu),
+                bias=bias,
                 out_dtype=torch.bfloat16,
             )
         else:
             assert a._elem_dtype == DTYPE_FP4
             assert b._elem_dtype == DTYPE_FP4
             assert gemm_choice is MXGemmKernelChoice.CUTLASS, "unsupported"
+            # FP4 operations
             res = torchao.ops.mx_fp4_bf16(
                 a._data, b._data, a_scale_block, b_scale_block
             )
+
+        # TODO update kernel to accept optional
+        if bias is not None:
+            res = res + bias
     else:
         # emulated MX gemm
         a_hp = a.to_dtype(a._orig_dtype)
         b_hp = b.to_dtype(b._orig_dtype)
         # assert memory layout we expect to be required in hardware
         assert a_hp.is_contiguous()
         assert b_hp.t().is_contiguous()
-        res = aten_op(a_hp, b_hp)
+
+        # Call appropriate aten_op based on whether bias is provided
+        if bias is not None:
+            res = aten_op(bias, a_hp, b_hp)  # addmm
+        else:
+            res = aten_op(a_hp, b_hp)  # mm
+
     return res
 
 
+@implements([aten.mm.default, aten.matmul.default])
+def mx_mm(func, types, args, kwargs):
+    a = args[0]
+    b = args[1]
+    assert isinstance(a, MXTensor) and isinstance(b, MXTensor)
+
+    return _addmm_mx_dispatch(a, b, func)
+
+
+@implements([aten.addmm.default])
+def mx_addmm(func, types, args, kwargs):
+    assert (
+        isinstance(args[0], torch.Tensor)
+        and isinstance(args[1], MXTensor)
+        and isinstance(args[2], MXTensor)
+    )
+    bias = args[0]
+    a = args[1]
+    b = args[2]
+
+    return _addmm_mx_dispatch(a, b, func, bias=bias)
+
+
 @implements([aten.t.default])
-def mx_t(aten_op, args, kwargs=None):
+def mx_t(func, types, args, kwargs):
     # For now, only transpose(input, 0, 1) is supported.
     old = args[0]
     new = MXTensor(
@@ -150,7 +198,7 @@ def mx_t(aten_op, args, kwargs=None):
 
 
 @implements([aten.sum.dim_IntList])
-def mx_cast_up_op(aten_op, args, kwargs=None):
+def mx_cast_up_op(func, types, args, kwargs):
     """Be careful with this function, this is a "fallback" op that
     casts the output of the op to the original precision. And performs the op.
 
@@ -166,11 +214,11 @@ def unwrap(x):
 
     new_args = tree_map(unwrap, args)
     new_kwargs = tree_map(unwrap, kwargs)
-    return aten_op(*new_args, **new_kwargs)
+    return func(*new_args, **new_kwargs)
 
 
 @implements([aten.view.default])
-def mx_view_op(aten_op, args, kwargs=None):
+def mx_view_op(func, types, args, kwargs):
     data = args[0]._data
     new_size = args[1]
     if args[0]._elem_dtype == DTYPE_FP4:
@@ -179,7 +227,7 @@ def mx_view_op(aten_op, args, kwargs=None):
     elif args[0]._elem_dtype in [DTYPE_FP6_E3M2, DTYPE_FP6_E2M3] and args[0]._pack_fp6:
         # special case fp6 as we pack 4 elements in 3 bytes
         new_size = tensor_size_hpx3_to_fp6x4(new_size, data.is_contiguous())
-    new_data = aten_op(data, new_size, *args[2:], **kwargs)
+    new_data = func(data, new_size, *args[2:], **kwargs)
     return MXTensor(
         args[0]._scale_e8m0,
         new_data,
@@ -193,7 +241,7 @@ def mx_view_op(aten_op, args, kwargs=None):
 
 
 @implements([aten._to_copy.default])
-def autocast_to_copy(aten_op, args, kwargs=None):
+def autocast_to_copy(func, types, args, kwargs):
     """This gets called when running matmul under autocast
     when the input is a MXTensor, presenting as a fp32
     tensor.

torchao/prototype/mx_formats/mx_subclass.py

@@ -24,7 +24,7 @@
 from torchao.quantization.transform_module import (
     register_quantize_module_handler,
 )
-from torchao.utils import is_sm_at_least_100
+from torchao.utils import TORCH_VERSION_AT_LEAST_2_5, is_sm_at_least_100
 
 
 @dataclass
@@ -115,3 +115,9 @@ def _mx_inference_linear_transform(module: torch.nn.Module, config: MXFPConfig):
     module.weight = torch.nn.Parameter(quantized_weight, requires_grad=False)
     module.extra_repr = types.MethodType(_linear_extra_repr, module)
     return module
+
+
+if TORCH_VERSION_AT_LEAST_2_5:
+    torch.serialization.add_safe_globals(
+        [MXTensor, MXGemmKernelChoice, _input_activation_quant_func_mxfp]
+    )

torchao/prototype/mx_formats/mx_tensor.py

@@ -18,7 +18,7 @@
 """
 
 from enum import Enum, auto
-from typing import Dict, Union
+from typing import Callable, Dict, Union
 
 import torch
 
@@ -562,7 +562,7 @@ def __torch_dispatch__(cls, func, types, args, kwargs=None):
         from torchao.prototype.mx_formats.mx_ops import MX_OPS_TABLE
 
         if func in MX_OPS_TABLE:
-            return MX_OPS_TABLE[func](func, args, kwargs)
+            return MX_OPS_TABLE[func](func, types, args, kwargs)
 
         raise NotImplementedError(f"{func} not implemented")
 
@@ -631,5 +631,21 @@ def __tensor_unflatten__(
             metadata["_pack_fp6"],
         )
 
+    def _apply_fn_to_data(self, fn: Callable):
+        """Applies a fn to all tensor components stored on this class"""
+        tensor_names, ctx = self.__tensor_flatten__()
+
+        # Apply the function to each tensor component
+        new_tensors = {}
+        for name in tensor_names:
+            new_tensors[name] = fn(getattr(self, name))
+
+        return self.__class__.__tensor_unflatten__(
+            new_tensors,
+            ctx,
+            None,  # outer_size parameter
+            None,  # outer_stride parameter
+        )
+
     # Do not force the MXTensor type on the returned tensor
     __torch_function__ = torch._C._disabled_torch_function_impl