from typing import Optional

import torch


def repeat_kv(hidden_states: torch.Tensor, n_rep: int) -> torch.Tensor:
    """
    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
    """
    batch, num_key_value_heads, slen, head_dim = hidden_states.shape
    if n_rep == 1:
        return hidden_states
    hidden_states = hidden_states[:, :, None, :, :].expand(batch, num_key_value_heads, n_rep, slen, head_dim)
    return hidden_states.reshape(batch, num_key_value_heads * n_rep, slen, head_dim)


def sdpa_attention_paged_forward(
    module: torch.nn.Module,
    query: torch.Tensor,
    key: torch.Tensor,
    value: torch.Tensor,
    attention_mask: Optional[torch.Tensor],
    dropout: float = 0.0,
    scaling: Optional[float] = None,
    is_causal: Optional[bool] = None,
    **kwargs,
) -> tuple[torch.Tensor, None]:
    cache = kwargs.pop("cache", None)
    if cache is not None:
        key, value = cache.update(key, value, module.layer_idx, **kwargs)
    if hasattr(module, "num_key_value_groups"):
        key = repeat_kv(key, module.num_key_value_groups)
        value = repeat_kv(value, module.num_key_value_groups)

    causal_mask = attention_mask
    query = query.contiguous()
    key = key.contiguous()
    value = value.contiguous()
    attn_output = torch.nn.functional.scaled_dot_product_attention(
        query,
        key,
        value,
        attn_mask=causal_mask,
        dropout_p=dropout,
        scale=scaling,
        is_causal=False,
    )
    attn_output = attn_output.transpose(1, 2).contiguous()

    return attn_output, None