# Copyright (c) 2026 LightSeek Foundation
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the "Software"), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.


import torch


def normalize_e4m3fn_to_e4m3fnuz(
    weight: torch.Tensor,
    weight_scale: torch.Tensor,
    input_scale: torch.Tensor | None = None,
) -> tuple[torch.Tensor, torch.Tensor, torch.Tensor | None]:
    assert weight.dtype == torch.float8_e4m3fn
    # The bits pattern 10000000(-128) represents zero in e4m3fn
    # but NaN in e4m3fnuz. So here we set it to 0.
    # https://onnx.ai/onnx/technical/float8.html
    weight_as_int8 = weight.view(torch.int8)
    ROCM_FP8_NAN_AS_INT = -128
    weight_as_int8[weight_as_int8 == ROCM_FP8_NAN_AS_INT] = 0
    weight = weight_as_int8.view(torch.float8_e4m3fnuz)

    # For the same bits representation, e4m3fnuz value is half of
    # the e4m3fn value, so we should double the scaling factor to
    # get the same dequantized value.
    # https://onnx.ai/onnx/technical/float8.html
    weight_scale = weight_scale * 2.0
    if input_scale is not None:
        input_scale = input_scale * 2.0
    return weight, weight_scale, input_scale