sgl-project/sglang · error · ValueError

kv-canary: has_swa_lut must be bool, got {type(has_swa_lut).

Error message

kv-canary: has_swa_lut must be bool, got {type(has_swa_lut).__name__}

What it means

has_swa_lut is a compile-time specialization flag for the JIT offsets kernel and must be a genuine Python bool, since the value is used to select which kernel variant is compiled and launched.

Source

Thrown at python/sglang/kernels/ops/kv_canary/plan/offsets_kernel.py:175

        raise ValueError(
            f"kv-canary: write_offsets_len must be positive, got {write_offsets_len}"
        )
    if write_req_capacity < 0:
        raise ValueError(
            f"kv-canary: write_req_capacity must be non-negative, got {write_req_capacity}"
        )
    if verify_capacity < 0:
        raise ValueError(
            f"kv-canary: verify_capacity must be non-negative, got {verify_capacity}"
        )
    if req_to_token_stride0 <= 0:
        raise ValueError(
            f"kv-canary: req_to_token_stride0 must be positive, got {req_to_token_stride0}"
        )
    if lut_len < 0:
        raise ValueError(f"kv-canary: lut_len must be non-negative, got {lut_len}")
    if not isinstance(has_swa_lut, bool):
        raise ValueError(
            f"kv-canary: has_swa_lut must be bool, got {type(has_swa_lut).__name__}"
        )
    if has_swa_lut and lut_len <= 0:
        raise ValueError("kv-canary: lut_len must be positive when has_swa_lut is True")
    if not has_swa_lut and lut_len != 0:
        raise ValueError("kv-canary: lut_len must be 0 when has_swa_lut is False")

    _require_len(req_pool_indices, "req_pool_indices", bs)
    _require_len(prefix_lens, "prefix_lens", bs)
    _require_len(extend_seq_lens, "extend_seq_lens", bs)
    _require_2d(req_to_token, "req_to_token")
    _require_min_len(lut_tensor, "lut_tensor", max(lut_len, 1))
    _require_min_len(
        out_verify_offsets_scratch,
        "out_verify_offsets_scratch",
        _PLAN_BS_BLOCK_SIZE + 1,
    )
    _require_len(out_write_offsets, "out_write_offsets", write_offsets_len)

View on GitHub (pinned to 0132848349)

Solutions

1. Pass has_swa_lut = full_to_swa_index_mapping is not None (a real bool)
2. Convert foreign boolean types with bool(...) before calling

Example fix

// before
launch_plan_offsets_kernel(..., has_swa_lut=int(lut is not None))
// after
launch_plan_offsets_kernel(..., has_swa_lut=lut is not None)

has_swa_lut = bool(full_to_swa_index_mapping is not None)

def as_kernel_flag(value) -> bool:
    if not isinstance(value, bool):
        raise TypeError(f"expected bool, got {type(value).__name__}")
    return value

When it happens

Trigger: Calling launch_plan_offsets_kernel with has_swa_lut as a non-bool (e.g. 0/1, numpy bool, torch bool tensor, or None).

Common situations: Passing full_to_swa_index_mapping is not None directly without converting to bool, or forwarding a numpy/torch boolean from data pipelines.

Related errors

• kv-canary: launch_canary_plan_kernels requires full_to_swa_i
• kv-canary: launch_canary_plan_kernels verify_capacity={verif
• kv-canary: launch_plan_entries_kernel requires req_to_verify
• kv-canary: write_offsets_len must be positive, got {write_of
• kv-canary: write_req_capacity must be non-negative, got {wri

AI-assisted analysis of sgl-project/sglang@0132848349 (2026-08-28). Data as JSON: /api/errors/f3b0c82c4cc15612. Report an issue: GitHub.