xai-org/x-algorithm · error · ValueError

mask_block_cnt and mask_block_idx must be provided for block

Error message

mask_block_cnt and mask_block_idx must be provided for block sparsity.

What it means

infer_block_sparse_expected_shapes requires the mask_block_idx tensor to be present when block sparsity is enabled; None means the sparsity pattern itself is missing, so expected shapes cannot be inferred. This instance fires early, before num_m_blocks is read from mask_block_idx.shape[2].

Source

Thrown at phoenix/xrex/cutedsl/ranker_fa4/block_sparsity.py:313

    batch_size: int,
    num_head: int,
    seqlen_q: int,
    seqlen_k: int,
    m_block_size: int,
    n_block_size: int,
    q_stage: int,
    context: str,
    sparse_block_size_q: int | None = None,
    sparse_block_size_kv: int | None = None,
) -> Tuple[Tuple[int, int, int], Tuple[int, int, int, int], int]:
    base_m_block = q_stage * m_block_size
    base_n_block = n_block_size
    if sparse_block_size_kv is None:
        sparse_block_size_kv = base_n_block
    if sparse_block_size_kv != base_n_block:
        raise ValueError(f"Block sparse tensors{context} require BLOCK_SIZE_KV={base_n_block}.")
    if tensors.mask_block_idx is None:
        raise ValueError("mask_block_cnt and mask_block_idx must be provided for block sparsity.")
    num_m_blocks = tensors.mask_block_idx.shape[2]

    if sparse_block_size_q is None:
        sparse_block_size_q = get_sparse_q_block_size(tensors, seqlen_q)
        if sparse_block_size_q is None and base_m_block != 1:
            raise ValueError(
                f"Block sparse tensors{context} require explicit sparse_block_size[0] "
                f"to disambiguate block size for seqlen_q={seqlen_q} and num_m_blocks={num_m_blocks}."
            )
        if sparse_block_size_q is None:
            sparse_block_size_q = ceildiv(seqlen_q, num_m_blocks)

    if sparse_block_size_q % base_m_block != 0:
        raise ValueError(
            f"Block sparse tensors{context} have block size {sparse_block_size_q}, "
            f"which must be a multiple of {base_m_block}."
        )

View on GitHub (pinned to 24c60942c5)

Solutions

1. Provide both mask_block_cnt and mask_block_idx tensors
2. Check for None fields on the container before calling the API
3. Disable block sparsity entirely if it was enabled by mistake

Example fix

# before
tensors = BlockSparseTensors(mask_block_cnt=cnt, mask_block_idx=None)

# after
tensors = BlockSparseTensors(mask_block_cnt=cnt, mask_block_idx=idx)

assert tensors.mask_block_cnt is not None and tensors.mask_block_idx is not None

def has_block_sparsity_pair(t) -> bool:
    return t.mask_block_cnt is not None and t.mask_block_idx is not None

When it happens

Trigger: Enabling block sparsity (passing mask_block_cnt or a sparse config) but leaving tensors.mask_block_idx as None; constructing the sparse-tensor container with only one of the two mask tensors.

Common situations: Partially initialized dataclasses/containers where the idx field was never populated; conditionally loading only counts; forgetting to attach the precomputed index tensor when wiring up the ranker inputs.

Related errors

• {name}_block_cnt and {name}_block_idx must be on the same de
• {name} must have dtype torch.int32
• valid_block_upper and valid_block_lower must both be provide
• type checking expression %s failed: invalid argument type: %
• Non-optional parameter %s must be declared before optional p

AI-assisted analysis of xai-org/x-algorithm@24c60942c5 (2026-08-28). Data as JSON: /api/errors/5c2beadccc996f35. Report an issue: GitHub.