Skip to content

vllm.attention.backends.cpu_mla

CPUMLABackend

Bases: AttentionBackend

Source code in vllm/attention/backends/cpu_mla.py 
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
class CPUMLABackend(AttentionBackend):

    @staticmethod
    def get_name() -> str:
        return "CPU_MLA"

    @staticmethod
    def get_metadata_cls() -> Type["CPUMLAMetadata"]:
        return CPUMLAMetadata

    @staticmethod
    def get_builder_cls() -> Type["CPUMLAMetadataBuilder"]:
        return CPUMLAMetadataBuilder

    @staticmethod
    def get_state_cls() -> Type["MLACommonState"]:
        return MLACommonState

    @staticmethod
    def get_impl_cls() -> Type["CPUMLAImpl"]:
        return CPUMLAImpl

    @staticmethod
    def get_kv_cache_shape(
        num_blocks: int,
        block_size: int,
        num_kv_heads: int,  # assumed to be 1 for MLA
        head_size: int,
    ) -> Tuple[int, ...]:
        return (num_blocks, block_size, head_size)

    @staticmethod
    def swap_blocks(
        src_kv_cache: torch.Tensor,
        dst_kv_cache: torch.Tensor,
        src_to_dst: torch.Tensor,
    ) -> None:
        ops.swap_blocks(src_kv_cache, dst_kv_cache, src_to_dst)

    @staticmethod
    def copy_blocks(
        kv_caches: List[torch.Tensor],
        src_to_dists: torch.Tensor,
    ) -> None:
        ops.copy_blocks_mla(kv_caches, src_to_dists)

    @staticmethod
    def get_supported_head_sizes() -> List[int]:
        return [576]

copy_blocks staticmethod 

copy_blocks(
    kv_caches: List[Tensor], src_to_dists: Tensor
) -> None
Source code in vllm/attention/backends/cpu_mla.py 
60
61
62
63
64
65
@staticmethod
def copy_blocks(
    kv_caches: List[torch.Tensor],
    src_to_dists: torch.Tensor,
) -> None:
    ops.copy_blocks_mla(kv_caches, src_to_dists)

get_builder_cls staticmethod 

get_builder_cls() -> Type[CPUMLAMetadataBuilder]
Source code in vllm/attention/backends/cpu_mla.py 
31
32
33
@staticmethod
def get_builder_cls() -> Type["CPUMLAMetadataBuilder"]:
    return CPUMLAMetadataBuilder

get_impl_cls staticmethod 

get_impl_cls() -> Type[CPUMLAImpl]
Source code in vllm/attention/backends/cpu_mla.py 
39
40
41
@staticmethod
def get_impl_cls() -> Type["CPUMLAImpl"]:
    return CPUMLAImpl

get_kv_cache_shape staticmethod 

get_kv_cache_shape(
    num_blocks: int,
    block_size: int,
    num_kv_heads: int,
    head_size: int,
) -> Tuple[int, ...]
Source code in vllm/attention/backends/cpu_mla.py 
43
44
45
46
47
48
49
50
@staticmethod
def get_kv_cache_shape(
    num_blocks: int,
    block_size: int,
    num_kv_heads: int,  # assumed to be 1 for MLA
    head_size: int,
) -> Tuple[int, ...]:
    return (num_blocks, block_size, head_size)

get_metadata_cls staticmethod 

get_metadata_cls() -> Type[CPUMLAMetadata]
Source code in vllm/attention/backends/cpu_mla.py 
27
28
29
@staticmethod
def get_metadata_cls() -> Type["CPUMLAMetadata"]:
    return CPUMLAMetadata

get_name staticmethod 

get_name() -> str
Source code in vllm/attention/backends/cpu_mla.py 
23
24
25
@staticmethod
def get_name() -> str:
    return "CPU_MLA"

get_state_cls staticmethod 

get_state_cls() -> Type[MLACommonState]
Source code in vllm/attention/backends/cpu_mla.py 
35
36
37
@staticmethod
def get_state_cls() -> Type["MLACommonState"]:
    return MLACommonState

get_supported_head_sizes staticmethod 

get_supported_head_sizes() -> List[int]
Source code in vllm/attention/backends/cpu_mla.py 
67
68
69
@staticmethod
def get_supported_head_sizes() -> List[int]:
    return [576]

swap_blocks staticmethod 

swap_blocks(
    src_kv_cache: Tensor,
    dst_kv_cache: Tensor,
    src_to_dst: Tensor,
) -> None
Source code in vllm/attention/backends/cpu_mla.py 
52
53
54
55
56
57
58
@staticmethod
def swap_blocks(
    src_kv_cache: torch.Tensor,
    dst_kv_cache: torch.Tensor,
    src_to_dst: torch.Tensor,
) -> None:
    ops.swap_blocks(src_kv_cache, dst_kv_cache, src_to_dst)

CPUMLAImpl

Bases: MLACommonImpl[CPUMLAMetadata]

Source code in vllm/attention/backends/cpu_mla.py 
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
class CPUMLAImpl(MLACommonImpl[CPUMLAMetadata]):

    def __init__(
            self,
            num_heads: int,
            head_size: int,
            scale: float,
            num_kv_heads: int,
            alibi_slopes: Optional[List[float]],
            sliding_window: Optional[int],
            kv_cache_dtype: str,
            blocksparse_params: Optional[Dict[str, Any]],
            logits_soft_cap: Optional[float],
            attn_type: str,
            kv_sharing_target_layer_name: Optional[str],
            # MLA Specific Arguments
            **mla_args) -> None:
        super().__init__(num_heads, head_size, scale, num_kv_heads,
                         alibi_slopes, sliding_window, kv_cache_dtype,
                         blocksparse_params, logits_soft_cap, attn_type,
                         kv_sharing_target_layer_name, **mla_args)

        unsupported_features = [
            alibi_slopes, sliding_window, blocksparse_params, logits_soft_cap
        ]
        if any(unsupported_features):
            raise NotImplementedError(
                "CPUMLAImpl does not support one of the following: "
                "alibi_slopes, sliding_window, blocksparse_params, "
                "logits_soft_cap")

        if attn_type != AttentionType.DECODER:
            raise NotImplementedError("Encoder self-attention and "
                                      "encoder/decoder cross-attention "
                                      "are not implemented for "
                                      "CPUMLAImpl")

        # states is implemented.
        if is_quantized_kv_cache(self.kv_cache_dtype):
            raise NotImplementedError(
                "CPUMLAImpl with FP8 KV cache not yet supported")

    def _forward_prefill(
            self,
            q: torch.Tensor,
            kv_c_normed: torch.Tensor,
            k_pe: torch.Tensor,
            kv_c_and_k_pe_cache: torch.Tensor,
            attn_metadata: CPUMLAMetadata,  # type: ignore[override]
    ) -> torch.Tensor:

        prefill_metadata = attn_metadata.prefill_metadata
        assert prefill_metadata is not None

        kv_nope = self.kv_b_proj(kv_c_normed)[0].view(\
            -1, self.num_heads, self.qk_nope_head_dim + self.v_head_dim)
        k_nope, v = kv_nope\
            .split([self.qk_nope_head_dim, self.v_head_dim], dim=-1)

        k = torch.cat((k_nope, k_pe.expand((*k_nope.shape[:-1], -1))), dim=-1)

        # For MLA the v head dim is smaller than qk head dim so we pad out
        # v with 0s to match the qk head dim
        v_padded = torch.nn.functional.pad(v, [0, q.shape[-1] - v.shape[-1]],
                                           value=0)

        output = torch.empty_like(q)
        ipex_ops.varlen_attention(
            query=q,
            key=k,
            value=v_padded,
            out=output,
            seqlen_q=prefill_metadata.prefill_query_start_loc,
            seqlen_k=prefill_metadata.prefill_query_start_loc,
            max_seqlen_q=prefill_metadata.max_query_len,
            max_seqlen_k=prefill_metadata.max_query_len,
            pdropout=0.0,
            softmax_scale=self.scale,
            zero_tensors=False,
            is_causal=True,
            return_softmax=False,
            gen_=None,
            logits_soft_cap=0.0,
            window_size_left=-1,
            window_size_right=-1,
            alibi_slopes=None,
        )

        # remove padding
        output = output.view(-1, self.num_heads,
                             q.shape[-1])[..., :v.shape[-1]]
        return output.reshape(-1, self.num_heads * v.shape[-1])

    def _forward_decode(
            self,
            q_nope: torch.Tensor,
            q_pe: torch.Tensor,
            kv_c_and_k_pe_cache: torch.Tensor,
            attn_metadata: CPUMLAMetadata,  # type: ignore[override]
    ) -> torch.Tensor:
        assert kv_c_and_k_pe_cache.numel() > 0

        decode_meta = attn_metadata.decode_metadata
        assert decode_meta is not None

        q = torch.cat([q_nope, q_pe], dim=-1)
        o = q.new_empty(q.shape[0], self.num_heads, self.kv_lora_rank)

        # Run MQA
        ops.mla_decode_kvcache_cpu(o, q, kv_c_and_k_pe_cache, self.scale,
                                   decode_meta.block_tables,
                                   decode_meta.seq_lens_tensor)
        return self._v_up_proj(o)

__init__ 

__init__(
    num_heads: int,
    head_size: int,
    scale: float,
    num_kv_heads: int,
    alibi_slopes: Optional[List[float]],
    sliding_window: Optional[int],
    kv_cache_dtype: str,
    blocksparse_params: Optional[Dict[str, Any]],
    logits_soft_cap: Optional[float],
    attn_type: str,
    kv_sharing_target_layer_name: Optional[str],
    **mla_args,
) -> None
Source code in vllm/attention/backends/cpu_mla.py 
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
def __init__(
        self,
        num_heads: int,
        head_size: int,
        scale: float,
        num_kv_heads: int,
        alibi_slopes: Optional[List[float]],
        sliding_window: Optional[int],
        kv_cache_dtype: str,
        blocksparse_params: Optional[Dict[str, Any]],
        logits_soft_cap: Optional[float],
        attn_type: str,
        kv_sharing_target_layer_name: Optional[str],
        # MLA Specific Arguments
        **mla_args) -> None:
    super().__init__(num_heads, head_size, scale, num_kv_heads,
                     alibi_slopes, sliding_window, kv_cache_dtype,
                     blocksparse_params, logits_soft_cap, attn_type,
                     kv_sharing_target_layer_name, **mla_args)

    unsupported_features = [
        alibi_slopes, sliding_window, blocksparse_params, logits_soft_cap
    ]
    if any(unsupported_features):
        raise NotImplementedError(
            "CPUMLAImpl does not support one of the following: "
            "alibi_slopes, sliding_window, blocksparse_params, "
            "logits_soft_cap")

    if attn_type != AttentionType.DECODER:
        raise NotImplementedError("Encoder self-attention and "
                                  "encoder/decoder cross-attention "
                                  "are not implemented for "
                                  "CPUMLAImpl")

    # states is implemented.
    if is_quantized_kv_cache(self.kv_cache_dtype):
        raise NotImplementedError(
            "CPUMLAImpl with FP8 KV cache not yet supported")

_forward_decode 

_forward_decode(
    q_nope: Tensor,
    q_pe: Tensor,
    kv_c_and_k_pe_cache: Tensor,
    attn_metadata: CPUMLAMetadata,
) -> Tensor
Source code in vllm/attention/backends/cpu_mla.py 
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
def _forward_decode(
        self,
        q_nope: torch.Tensor,
        q_pe: torch.Tensor,
        kv_c_and_k_pe_cache: torch.Tensor,
        attn_metadata: CPUMLAMetadata,  # type: ignore[override]
) -> torch.Tensor:
    assert kv_c_and_k_pe_cache.numel() > 0

    decode_meta = attn_metadata.decode_metadata
    assert decode_meta is not None

    q = torch.cat([q_nope, q_pe], dim=-1)
    o = q.new_empty(q.shape[0], self.num_heads, self.kv_lora_rank)

    # Run MQA
    ops.mla_decode_kvcache_cpu(o, q, kv_c_and_k_pe_cache, self.scale,
                               decode_meta.block_tables,
                               decode_meta.seq_lens_tensor)
    return self._v_up_proj(o)

_forward_prefill 

_forward_prefill(
    q: Tensor,
    kv_c_normed: Tensor,
    k_pe: Tensor,
    kv_c_and_k_pe_cache: Tensor,
    attn_metadata: CPUMLAMetadata,
) -> Tensor
Source code in vllm/attention/backends/cpu_mla.py 
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
def _forward_prefill(
        self,
        q: torch.Tensor,
        kv_c_normed: torch.Tensor,
        k_pe: torch.Tensor,
        kv_c_and_k_pe_cache: torch.Tensor,
        attn_metadata: CPUMLAMetadata,  # type: ignore[override]
) -> torch.Tensor:

    prefill_metadata = attn_metadata.prefill_metadata
    assert prefill_metadata is not None

    kv_nope = self.kv_b_proj(kv_c_normed)[0].view(\
        -1, self.num_heads, self.qk_nope_head_dim + self.v_head_dim)
    k_nope, v = kv_nope\
        .split([self.qk_nope_head_dim, self.v_head_dim], dim=-1)

    k = torch.cat((k_nope, k_pe.expand((*k_nope.shape[:-1], -1))), dim=-1)

    # For MLA the v head dim is smaller than qk head dim so we pad out
    # v with 0s to match the qk head dim
    v_padded = torch.nn.functional.pad(v, [0, q.shape[-1] - v.shape[-1]],
                                       value=0)

    output = torch.empty_like(q)
    ipex_ops.varlen_attention(
        query=q,
        key=k,
        value=v_padded,
        out=output,
        seqlen_q=prefill_metadata.prefill_query_start_loc,
        seqlen_k=prefill_metadata.prefill_query_start_loc,
        max_seqlen_q=prefill_metadata.max_query_len,
        max_seqlen_k=prefill_metadata.max_query_len,
        pdropout=0.0,
        softmax_scale=self.scale,
        zero_tensors=False,
        is_causal=True,
        return_softmax=False,
        gen_=None,
        logits_soft_cap=0.0,
        window_size_left=-1,
        window_size_right=-1,
        alibi_slopes=None,
    )

    # remove padding
    output = output.view(-1, self.num_heads,
                         q.shape[-1])[..., :v.shape[-1]]
    return output.reshape(-1, self.num_heads * v.shape[-1])

CPUMLAMetadata dataclass

Bases: TorchSDPAMetadata

Source code in vllm/attention/backends/cpu_mla.py 
72
73
74
75
76
77
78
79
80
@dataclass
class CPUMLAMetadata(TorchSDPAMetadata):
    # New for MLA
    # Input positions for rotrary embeddings since for MLA the rotary
    # position embeddings are applied inside the attention backend
    input_positions: torch.Tensor = None

    # required by MLACommonImpl
    is_profile_run: bool = False

input_positions class-attribute instance-attribute 

input_positions: Tensor = None

is_profile_run class-attribute instance-attribute 

is_profile_run: bool = False

__init__ 

__init__(
    seq_lens_tensor: Optional[Tensor],
    max_decode_seq_len: int,
    block_tables: Optional[Tensor],
    num_prefills: int,
    num_prefill_tokens: int,
    num_decode_tokens: int,
    slot_mapping: Tensor,
    multi_modal_placeholder_index_maps: Optional[
        Dict[str, IndexMap]
    ],
    enable_kv_scales_calculation: bool,
    chunked_prefill: bool,
    seq_lens: Optional[List[int]] = None,
    max_query_len: Optional[int] = None,
    max_kv_len: Optional[int] = None,
    prefill_query_start_loc: Optional[Tensor] = None,
    kv_start_loc: Optional[Tensor] = None,
    prefill_block_tables: Optional[Tensor] = None,
    query_start_loc: Optional[Tensor] = None,
    encoder_seq_lens: Optional[List[int]] = None,
    encoder_seq_lens_tensor: Optional[Tensor] = None,
    max_encoder_seq_len: Optional[int] = None,
    num_encoder_tokens: Optional[int] = None,
    cross_slot_mapping: Optional[Tensor] = None,
    cross_block_tables: Optional[Tensor] = None,
    input_positions: Tensor = None,
    is_profile_run: bool = False,
) -> None

CPUMLAMetadataBuilder

Bases: AttentionMetadataBuilder[CPUMLAMetadata]

Source code in vllm/attention/backends/cpu_mla.py 
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
class CPUMLAMetadataBuilder(AttentionMetadataBuilder[CPUMLAMetadata]):

    def __init__(self, input_builder: ModelInputForCPUBuilder) -> None:
        self.chunked_prefill = input_builder.chunked_prefill
        self.input_builder = input_builder
        assert not self.chunked_prefill, \
            "chunked prefill is currently not supported"

    def prepare(self):
        self.input_data = self.input_builder.input_data

    def build(self, seq_lens, query_lens, cuda_graph_pad_size, batch_size):
        input_data = self.input_data
        prefill_seq_lens = seq_lens[0:input_data.num_prefills]
        prefill_query_lens = query_lens[0:input_data.num_prefills]
        slot_mapping = torch.tensor(input_data.slot_mapping,
                                    dtype=torch.long,
                                    device="cpu")

        # metadata for prefill
        if input_data.num_prefills > 0:
            query_lens_tensor = torch.tensor(prefill_query_lens,
                                             dtype=torch.int32,
                                             device="cpu")
            kv_lens_tensor = torch.tensor(prefill_seq_lens,
                                          dtype=torch.int32,
                                          device="cpu")
            query_start_loc = torch.zeros(input_data.num_prefills + 1,
                                          dtype=torch.int32,
                                          device="cpu")
            kv_start_loc = torch.zeros(input_data.num_prefills + 1,
                                       dtype=torch.int32,
                                       device="cpu")
            torch.cumsum(query_lens_tensor,
                         dim=0,
                         dtype=torch.int32,
                         out=query_start_loc[1:])
            torch.cumsum(kv_lens_tensor,
                         dim=0,
                         dtype=torch.int32,
                         out=kv_start_loc[1:])
            max_query_len = max(prefill_query_lens)
            max_kv_len = max(prefill_seq_lens)

            # for chunked-prefill
            if self.chunked_prefill:
                prefill_block_tables = make_tensor_with_pad(
                    self.input_data.prefill_block_tables,
                    pad=0,
                    dtype=torch.int32,
                    device="cpu",
                )
            else:
                prefill_block_tables = None

        else:
            query_start_loc = None
            kv_start_loc = None
            max_query_len = None
            max_kv_len = None
            prefill_block_tables = None

        # metadata for decode
        if input_data.num_decode_tokens != 0:
            seq_lens_tensor = torch.tensor(
                input_data.seq_lens[input_data.num_prefills:],
                dtype=torch.int32,
                device="cpu",
            )
            block_tables = make_tensor_with_pad(
                self.input_data.decode_block_tables,
                pad=0,
                dtype=torch.int32,
                device="cpu",
            )
        else:
            block_tables = torch.tensor([])
            seq_lens_tensor = torch.tensor(
                input_data.seq_lens[:input_data.num_prefills],
                dtype=torch.int32,
                device="cpu",
            )

        # For multi-modal models
        placeholder_index_maps = None
        if len(input_data.multi_modal_inputs_list) != 0:
            placeholder_index_maps = {
                modality: placeholder_map.index_map()
                for modality, placeholder_map in
                input_data.multi_modal_placeholder_maps.items()
            }

        return CPUMLAMetadata(
            chunked_prefill=self.chunked_prefill,
            seq_lens=prefill_seq_lens,
            seq_lens_tensor=seq_lens_tensor,
            max_query_len=max_query_len,
            max_kv_len=max_kv_len,
            prefill_query_start_loc=query_start_loc,
            kv_start_loc=kv_start_loc,
            max_decode_seq_len=input_data.max_decode_seq_len,
            num_prefills=input_data.num_prefills,
            num_prefill_tokens=input_data.num_prefill_tokens,
            num_decode_tokens=input_data.num_decode_tokens,
            block_tables=block_tables,
            prefill_block_tables=prefill_block_tables,
            slot_mapping=slot_mapping,
            multi_modal_placeholder_index_maps=placeholder_index_maps,
            enable_kv_scales_calculation=False,
            input_positions=torch.tensor([self.input_data.input_positions]))

chunked_prefill instance-attribute 

chunked_prefill = chunked_prefill

input_builder instance-attribute 

input_builder = input_builder

__init__ 

__init__(input_builder: ModelInputForCPUBuilder) -> None
Source code in vllm/attention/backends/cpu_mla.py 
85
86
87
88
89
def __init__(self, input_builder: ModelInputForCPUBuilder) -> None:
    self.chunked_prefill = input_builder.chunked_prefill
    self.input_builder = input_builder
    assert not self.chunked_prefill, \
        "chunked prefill is currently not supported"

build 

build(
    seq_lens, query_lens, cuda_graph_pad_size, batch_size
)
Source code in vllm/attention/backends/cpu_mla.py 
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
def build(self, seq_lens, query_lens, cuda_graph_pad_size, batch_size):
    input_data = self.input_data
    prefill_seq_lens = seq_lens[0:input_data.num_prefills]
    prefill_query_lens = query_lens[0:input_data.num_prefills]
    slot_mapping = torch.tensor(input_data.slot_mapping,
                                dtype=torch.long,
                                device="cpu")

    # metadata for prefill
    if input_data.num_prefills > 0:
        query_lens_tensor = torch.tensor(prefill_query_lens,
                                         dtype=torch.int32,
                                         device="cpu")
        kv_lens_tensor = torch.tensor(prefill_seq_lens,
                                      dtype=torch.int32,
                                      device="cpu")
        query_start_loc = torch.zeros(input_data.num_prefills + 1,
                                      dtype=torch.int32,
                                      device="cpu")
        kv_start_loc = torch.zeros(input_data.num_prefills + 1,
                                   dtype=torch.int32,
                                   device="cpu")
        torch.cumsum(query_lens_tensor,
                     dim=0,
                     dtype=torch.int32,
                     out=query_start_loc[1:])
        torch.cumsum(kv_lens_tensor,
                     dim=0,
                     dtype=torch.int32,
                     out=kv_start_loc[1:])
        max_query_len = max(prefill_query_lens)
        max_kv_len = max(prefill_seq_lens)

        # for chunked-prefill
        if self.chunked_prefill:
            prefill_block_tables = make_tensor_with_pad(
                self.input_data.prefill_block_tables,
                pad=0,
                dtype=torch.int32,
                device="cpu",
            )
        else:
            prefill_block_tables = None

    else:
        query_start_loc = None
        kv_start_loc = None
        max_query_len = None
        max_kv_len = None
        prefill_block_tables = None

    # metadata for decode
    if input_data.num_decode_tokens != 0:
        seq_lens_tensor = torch.tensor(
            input_data.seq_lens[input_data.num_prefills:],
            dtype=torch.int32,
            device="cpu",
        )
        block_tables = make_tensor_with_pad(
            self.input_data.decode_block_tables,
            pad=0,
            dtype=torch.int32,
            device="cpu",
        )
    else:
        block_tables = torch.tensor([])
        seq_lens_tensor = torch.tensor(
            input_data.seq_lens[:input_data.num_prefills],
            dtype=torch.int32,
            device="cpu",
        )

    # For multi-modal models
    placeholder_index_maps = None
    if len(input_data.multi_modal_inputs_list) != 0:
        placeholder_index_maps = {
            modality: placeholder_map.index_map()
            for modality, placeholder_map in
            input_data.multi_modal_placeholder_maps.items()
        }

    return CPUMLAMetadata(
        chunked_prefill=self.chunked_prefill,
        seq_lens=prefill_seq_lens,
        seq_lens_tensor=seq_lens_tensor,
        max_query_len=max_query_len,
        max_kv_len=max_kv_len,
        prefill_query_start_loc=query_start_loc,
        kv_start_loc=kv_start_loc,
        max_decode_seq_len=input_data.max_decode_seq_len,
        num_prefills=input_data.num_prefills,
        num_prefill_tokens=input_data.num_prefill_tokens,
        num_decode_tokens=input_data.num_decode_tokens,
        block_tables=block_tables,
        prefill_block_tables=prefill_block_tables,
        slot_mapping=slot_mapping,
        multi_modal_placeholder_index_maps=placeholder_index_maps,
        enable_kv_scales_calculation=False,
        input_positions=torch.tensor([self.input_data.input_positions]))

prepare 

prepare()
Source code in vllm/attention/backends/cpu_mla.py 
91
92
def prepare(self):
    self.input_data = self.input_builder.input_data