# Copyright (c) 2023-2025, Songlin Yang, Yu Zhang

import warnings

import torch

from fla.ops.simple_gla import fused_chunk_simple_gla


@torch.compiler.disable
def fused_chunk_retention(
    q: torch.Tensor,
    k: torch.Tensor,
    v: torch.Tensor,
    scale: float | None = None,
    initial_state: torch.Tensor | None = None,
    output_final_state: bool = False,
    cu_seqlens: torch.LongTensor | None = None,
    head_first: bool = False,
) -> tuple[torch.Tensor, torch.Tensor]:
    r"""
    Args:
        q (torch.Tensor):
            queries of shape `[B, T, H, K]`.
        k (torch.Tensor):
            keys of shape `[B, T, H, K]`.
        v (torch.Tensor):
            values of shape `[B, T, H, V]`.
        scale (Optional[float]):
            Scale factor for the attention scores.
            If not provided, it will default to `1 / sqrt(K)`. Default: `None`.
        initial_state (Optional[torch.Tensor]):
            Initial state of shape `[N, H, K, V]` for `N` input sequences.
            For equal-length input sequences, `N` equals the batch size `B`.
            Default: `None`.
        output_final_state (Optional[bool]):
            Whether to output the final state of shape `[N, H, K, V]`. Default: `False`.
        cu_seqlens (torch.LongTensor):
            Cumulative sequence lengths of shape `[N+1]` used for variable-length training,
            consistent with the FlashAttention API.
        head_first (Optional[bool]):
            Whether the inputs are in the head-first format.
            Default: `False`.
            This argument has been deprecated.

    Returns:
        o (torch.Tensor):
            Outputs of shape `[B, T, H, V]`.
        final_state (torch.Tensor):
            Final state of shape `[N, H, K, V]` if `output_final_state=True` else `None`.

    """
    if head_first:
        raise DeprecationWarning(
            "head_first is deprecated and will be removed in a future version. "
            "Please use head_first=False for now instead.",
        )
    if not head_first and q.shape[1] < q.shape[2]:
        warnings.warn(
            f"Input tensor shape suggests potential format mismatch: seq_len ({q.shape[1]}) < num_heads ({q.shape[2]}). "
            "This may indicate the inputs were passed in head-first format [B, H, T, ...] "
            "when head_first=False was specified. "
            "Please verify your input tensor format matches the expected shape [B, T, H, ...].",
        )
    g_gamma = (1 - q.new_tensor(2., dtype=torch.float).pow(-5. - q.new_tensor(range(q.shape[2]), dtype=torch.float))).log()
    o, final_state = fused_chunk_simple_gla(
        q=q,
        k=k,
        v=v,
        g_gamma=g_gamma,
        scale=scale,
        initial_state=initial_state,
        output_final_state=output_final_state,
        cu_seqlens=cu_seqlens,
    )
    return o, final_state