{"record":{"id":"bff7140e177ef8a4","repo":"sgl-project/sglang","slug":"seqused-k-tensor-must-be-int32","errorCode":null,"errorMessage":"seqused_k tensor must be Int32","messagePattern":"seqused_k tensor must be Int32","errorType":"validation","errorClass":"TypeError","httpStatus":null,"severity":"error","filePath":"python/sglang/kernels/ops/attention/flash_attn/cute/flash_fwd.py","lineNumber":223,"sourceCode":"            # SplitKV writes float32 partial outputs; Q/K/V still fp16/bf16.\n            if const_expr(not (mQ_type == mK_type == mV_type)):\n                raise TypeError(\"Q/K/V must have the same data type\")\n            if const_expr(mO_type != Float32):\n                raise TypeError(\"SplitKV partial output (mO) must be Float32\")\n        elif const_expr(not (mQ_type == mK_type == mV_type == mO_type)):\n            raise TypeError(\"All tensors must have the same data type\")\n        if const_expr(mQ_type not in [cutlass.Float16, cutlass.BFloat16]):\n            raise TypeError(\"Only Float16 or BFloat16 is supported\")\n        if const_expr(mLSE_type not in [None, Float32]):\n            raise TypeError(\"LSE tensor must be Float32\")\n        if const_expr(mCuSeqlensQ_type not in [None, Int32]):\n            raise TypeError(\"cu_seqlens_q tensor must be Int32\")\n        if const_expr(mCuSeqlensK_type not in [None, Int32]):\n            raise TypeError(\"cu_seqlens_k tensor must be Int32\")\n        if const_expr(mSeqUsedQ_type not in [None, Int32]):\n            raise TypeError(\"seqused_q tensor must be Int32\")\n        if const_expr(mSeqUsedK_type not in [None, Int32]):\n            raise TypeError(\"seqused_k tensor must be Int32\")\n        assert mQ_type == self.dtype\n\n    def _setup_attributes(self):\n        # ///////////////////////////////////////////////////////////////////////////////\n        # Shared memory layout: Q/K/V\n        # ///////////////////////////////////////////////////////////////////////////////\n        (\n            sQ_layout_atom,\n            sK_layout_atom,\n            sV_layout_atom,\n            sO_layout_atom,\n            sP_layout_atom,\n        ) = self._get_smem_layout_atom()\n        self.sQ_layout = cute.tile_to_shape(\n            sQ_layout_atom,\n            (self.tile_m, self.tile_hdim),\n            (0, 1),\n        )","sourceCodeStart":205,"sourceCodeEnd":241,"githubUrl":"https://github.com/sgl-project/sglang/blob/0132848349585cfe6aae51c4941cbae872505f8a/python/sglang/kernels/ops/attention/flash_attn/cute/flash_fwd.py#L205-L241","documentation":"The optional seqused_k tensor (per-batch actual key/value sequence lengths for masking) must be Int32, mirroring the seqused_q check. Passing None is allowed.","triggerScenarios":"Providing seqused_k with a dtype other than Int32 (typically Int64) to FlashAttentionForward.","commonSituations":"Reusing an int64 sequence-length table for both seqused_q and seqused_k; masking KV padding with un-cast metadata tensors.","solutions":["Cast: seqused_k = seqused_k.to(torch.int32)","Keep all length metadata tensors int32 consistently in your attention wrapper"],"exampleFix":"// before\nseqused_k = kv_lens  # int64\n// after\nseqused_k = kv_lens.to(torch.int32)","handlingStrategy":"type-guard","validationCode":"if seqused_k is not None:\n    assert seqused_k.dtype == torch.int32","typeGuard":"def int32_or_none(t) -> bool:\n    return t is None or t.dtype == torch.int32","tryCatchPattern":null,"preventionTips":["Validate both seqused tensors together before calling the kernel","Standardize on int32 metadata buffers across the attention backend"],"tags":["cuda","dtype","flash-attention","masking","int32"],"backgroundTag":null,"analyzedSha":"0132848349585cfe6aae51c4941cbae872505f8a","analyzedAt":"2026-08-28T05:10:05.995Z","schemaVersion":2},"datasetVersion":"2026-08-28T06:17:29.519Z"}