Skip to content

vllm.v1.worker.gpu.cudagraph_utils

CudaGraphManager

Source code in vllm/v1/worker/gpu/cudagraph_utils.py 
 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
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 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
class CudaGraphManager:
    def __init__(
        self,
        vllm_config: VllmConfig,
        device: torch.device,
    ):
        self.vllm_config = vllm_config
        self.scheduler_config = vllm_config.scheduler_config
        self.device = device

        self.max_model_len = vllm_config.model_config.max_model_len
        self.max_num_reqs = self.scheduler_config.max_num_seqs
        self.max_num_tokens = self.scheduler_config.max_num_batched_tokens
        self.dp_size = vllm_config.parallel_config.data_parallel_size
        self.compilation_config = vllm_config.compilation_config
        assert self.compilation_config is not None
        self.cudagraph_mode: CUDAGraphMode
        if self.compilation_config.cudagraph_mode is None:
            self.cudagraph_mode = CUDAGraphMode.NONE
        else:
            self.cudagraph_mode = self.compilation_config.cudagraph_mode
        self.cudagraph_sizes = get_cudagraph_sizes(
            self.compilation_config.cudagraph_capture_sizes,
            self.max_num_reqs,
            self.max_num_tokens,
            self.cudagraph_mode,
        )

        self.graphs: dict[int, torch.cuda.CUDAGraph] = {}
        self.pool = torch.cuda.graph_pool_handle()
        self.hidden_states: torch.Tensor | None = None

    def needs_capture(self) -> bool:
        return len(self.cudagraph_sizes) > 0

    def get_cudagraph_size(
        self,
        scheduler_output: SchedulerOutput,
        num_tokens_after_padding: int,
    ) -> int | None:
        return get_cudagraph_size(
            num_tokens_after_padding,
            scheduler_output.num_scheduled_tokens.values(),
            self.cudagraph_sizes,
            self.cudagraph_mode,
        )

    def capture_graph(
        self,
        num_tokens: int,
        model: nn.Module,
        input_buffers: InputBuffers,
        block_tables: BlockTables,
        attn_metadata_builders: list[AttentionMetadataBuilder],
        kv_cache_config: KVCacheConfig,
    ) -> None:
        num_reqs = min(num_tokens, self.max_num_reqs)
        input_ids = input_buffers.input_ids.gpu[:num_tokens]
        positions = input_buffers.positions[:num_tokens]
        attn_metadata = prepare_inputs_to_capture(
            num_reqs,
            num_tokens,
            input_buffers,
            block_tables,
            attn_metadata_builders,
            self.max_model_len,
            kv_cache_config,
        )
        num_tokens_across_dp = make_num_tokens_across_dp(self.dp_size, num_tokens)

        # Warm up.
        with set_forward_context(
            attn_metadata,
            self.vllm_config,
            num_tokens=num_tokens,
            cudagraph_runtime_mode=CUDAGraphMode.NONE,
            num_tokens_across_dp=num_tokens_across_dp,
        ):
            hidden_states = model(
                input_ids=input_ids,
                positions=positions,
            )
            if self.hidden_states is None:
                self.hidden_states = torch.empty_like(hidden_states)

        # Capture the graph.
        assert num_tokens not in self.graphs
        graph = torch.cuda.CUDAGraph()
        with (
            set_forward_context(
                attn_metadata,
                self.vllm_config,
                num_tokens=num_tokens,
                cudagraph_runtime_mode=CUDAGraphMode.NONE,
                num_tokens_across_dp=num_tokens_across_dp,
            ),
            torch.cuda.graph(graph, self.pool),
        ):
            hidden_states = model(
                input_ids=input_ids,
                positions=positions,
            )
            self.hidden_states[:num_tokens] = hidden_states
        self.graphs[num_tokens] = graph

    @torch.inference_mode()
    def capture(
        self,
        model: nn.Module,
        input_buffers: InputBuffers,
        block_tables: BlockTables,
        attn_metadata_builders: list[AttentionMetadataBuilder],
        kv_cache_config: KVCacheConfig,
    ) -> None:
        capture_graphs(
            self.cudagraph_sizes,
            self.device,
            self.capture_graph,
            model=model,
            input_buffers=input_buffers,
            block_tables=block_tables,
            attn_metadata_builders=attn_metadata_builders,
            kv_cache_config=kv_cache_config,
        )

    def run(self, num_tokens: int) -> torch.Tensor:
        assert num_tokens in self.graphs
        self.graphs[num_tokens].replay()
        assert self.hidden_states is not None
        return self.hidden_states[:num_tokens]

compilation_config instance-attribute 

compilation_config = compilation_config

cudagraph_mode instance-attribute 

cudagraph_mode: CUDAGraphMode

cudagraph_sizes instance-attribute 

cudagraph_sizes = get_cudagraph_sizes(
    cudagraph_capture_sizes,
    max_num_reqs,
    max_num_tokens,
    cudagraph_mode,
)

device instance-attribute 

device = device

dp_size instance-attribute 

dp_size = data_parallel_size

graphs instance-attribute 

graphs: dict[int, CUDAGraph] = {}

hidden_states instance-attribute 

hidden_states: Tensor | None = None

max_model_len instance-attribute 

max_model_len = max_model_len

max_num_reqs instance-attribute 

max_num_reqs = max_num_seqs

max_num_tokens instance-attribute 

max_num_tokens = max_num_batched_tokens

pool instance-attribute 

pool = graph_pool_handle()

scheduler_config instance-attribute 

scheduler_config = scheduler_config

vllm_config instance-attribute 

vllm_config = vllm_config

__init__ 

__init__(vllm_config: VllmConfig, device: device)
Source code in vllm/v1/worker/gpu/cudagraph_utils.py 
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
def __init__(
    self,
    vllm_config: VllmConfig,
    device: torch.device,
):
    self.vllm_config = vllm_config
    self.scheduler_config = vllm_config.scheduler_config
    self.device = device

    self.max_model_len = vllm_config.model_config.max_model_len
    self.max_num_reqs = self.scheduler_config.max_num_seqs
    self.max_num_tokens = self.scheduler_config.max_num_batched_tokens
    self.dp_size = vllm_config.parallel_config.data_parallel_size
    self.compilation_config = vllm_config.compilation_config
    assert self.compilation_config is not None
    self.cudagraph_mode: CUDAGraphMode
    if self.compilation_config.cudagraph_mode is None:
        self.cudagraph_mode = CUDAGraphMode.NONE
    else:
        self.cudagraph_mode = self.compilation_config.cudagraph_mode
    self.cudagraph_sizes = get_cudagraph_sizes(
        self.compilation_config.cudagraph_capture_sizes,
        self.max_num_reqs,
        self.max_num_tokens,
        self.cudagraph_mode,
    )

    self.graphs: dict[int, torch.cuda.CUDAGraph] = {}
    self.pool = torch.cuda.graph_pool_handle()
    self.hidden_states: torch.Tensor | None = None

capture 

capture(
    model: Module,
    input_buffers: InputBuffers,
    block_tables: BlockTables,
    attn_metadata_builders: list[AttentionMetadataBuilder],
    kv_cache_config: KVCacheConfig,
) -> None
Source code in vllm/v1/worker/gpu/cudagraph_utils.py 
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
@torch.inference_mode()
def capture(
    self,
    model: nn.Module,
    input_buffers: InputBuffers,
    block_tables: BlockTables,
    attn_metadata_builders: list[AttentionMetadataBuilder],
    kv_cache_config: KVCacheConfig,
) -> None:
    capture_graphs(
        self.cudagraph_sizes,
        self.device,
        self.capture_graph,
        model=model,
        input_buffers=input_buffers,
        block_tables=block_tables,
        attn_metadata_builders=attn_metadata_builders,
        kv_cache_config=kv_cache_config,
    )

capture_graph 

capture_graph(
    num_tokens: int,
    model: Module,
    input_buffers: InputBuffers,
    block_tables: BlockTables,
    attn_metadata_builders: list[AttentionMetadataBuilder],
    kv_cache_config: KVCacheConfig,
) -> None
Source code in vllm/v1/worker/gpu/cudagraph_utils.py 
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 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
def capture_graph(
    self,
    num_tokens: int,
    model: nn.Module,
    input_buffers: InputBuffers,
    block_tables: BlockTables,
    attn_metadata_builders: list[AttentionMetadataBuilder],
    kv_cache_config: KVCacheConfig,
) -> None:
    num_reqs = min(num_tokens, self.max_num_reqs)
    input_ids = input_buffers.input_ids.gpu[:num_tokens]
    positions = input_buffers.positions[:num_tokens]
    attn_metadata = prepare_inputs_to_capture(
        num_reqs,
        num_tokens,
        input_buffers,
        block_tables,
        attn_metadata_builders,
        self.max_model_len,
        kv_cache_config,
    )
    num_tokens_across_dp = make_num_tokens_across_dp(self.dp_size, num_tokens)

    # Warm up.
    with set_forward_context(
        attn_metadata,
        self.vllm_config,
        num_tokens=num_tokens,
        cudagraph_runtime_mode=CUDAGraphMode.NONE,
        num_tokens_across_dp=num_tokens_across_dp,
    ):
        hidden_states = model(
            input_ids=input_ids,
            positions=positions,
        )
        if self.hidden_states is None:
            self.hidden_states = torch.empty_like(hidden_states)

    # Capture the graph.
    assert num_tokens not in self.graphs
    graph = torch.cuda.CUDAGraph()
    with (
        set_forward_context(
            attn_metadata,
            self.vllm_config,
            num_tokens=num_tokens,
            cudagraph_runtime_mode=CUDAGraphMode.NONE,
            num_tokens_across_dp=num_tokens_across_dp,
        ),
        torch.cuda.graph(graph, self.pool),
    ):
        hidden_states = model(
            input_ids=input_ids,
            positions=positions,
        )
        self.hidden_states[:num_tokens] = hidden_states
    self.graphs[num_tokens] = graph

get_cudagraph_size 

get_cudagraph_size(
    scheduler_output: SchedulerOutput,
    num_tokens_after_padding: int,
) -> int | None
Source code in vllm/v1/worker/gpu/cudagraph_utils.py 
59
60
61
62
63
64
65
66
67
68
69
def get_cudagraph_size(
    self,
    scheduler_output: SchedulerOutput,
    num_tokens_after_padding: int,
) -> int | None:
    return get_cudagraph_size(
        num_tokens_after_padding,
        scheduler_output.num_scheduled_tokens.values(),
        self.cudagraph_sizes,
        self.cudagraph_mode,
    )

needs_capture 

needs_capture() -> bool
Source code in vllm/v1/worker/gpu/cudagraph_utils.py 
56
57
def needs_capture(self) -> bool:
    return len(self.cudagraph_sizes) > 0

run 

run(num_tokens: int) -> Tensor
Source code in vllm/v1/worker/gpu/cudagraph_utils.py 
149
150
151
152
153
def run(self, num_tokens: int) -> torch.Tensor:
    assert num_tokens in self.graphs
    self.graphs[num_tokens].replay()
    assert self.hidden_states is not None
    return self.hidden_states[:num_tokens]

capture_graphs 

capture_graphs(
    cudagraph_sizes: dict[int, int],
    device: device,
    capture_fn: Callable,
    **capture_kwargs,
) -> None
Source code in vllm/v1/worker/gpu/cudagraph_utils.py 
205
206
207
208
209
210
211
212
213
214
215
216
217
218
def capture_graphs(
    cudagraph_sizes: dict[int, int],
    device: torch.device,
    capture_fn: Callable,
    **capture_kwargs,
) -> None:
    # Capture larger graphs first.
    sizes_to_capture = sorted(set(cudagraph_sizes.values()), reverse=True)
    if is_global_first_rank():
        sizes_to_capture = tqdm(sizes_to_capture, desc="Capturing CUDA graphs")

    with graph_capture(device=device):
        for size in sizes_to_capture:
            capture_fn(size, **capture_kwargs)

get_cudagraph_size 

get_cudagraph_size(
    num_tokens_after_dp_padding: int,
    num_tokens_per_request: Iterable[int],
    cudagraph_sizes: dict[int, int],
    cudagraph_mode: CUDAGraphMode,
) -> int | None
Source code in vllm/v1/worker/gpu/cudagraph_utils.py 
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
def get_cudagraph_size(
    num_tokens_after_dp_padding: int,
    num_tokens_per_request: Iterable[int],
    cudagraph_sizes: dict[int, int],
    cudagraph_mode: CUDAGraphMode,
) -> int | None:
    size = cudagraph_sizes.get(num_tokens_after_dp_padding)
    if size is None:
        # No CUDA graph for this size.
        return None
    if cudagraph_mode == CUDAGraphMode.FULL_DECODE_ONLY:
        all_decode = all(x == 1 for x in num_tokens_per_request)
        if not all_decode:
            # Prefill is included.
            return None
    return size

get_cudagraph_sizes 

get_cudagraph_sizes(
    capture_sizes: list[int] | None,
    max_num_reqs: int,
    max_num_tokens: int,
    cudagraph_mode: CUDAGraphMode,
) -> dict[int, int]
Source code in vllm/v1/worker/gpu/cudagraph_utils.py 
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
def get_cudagraph_sizes(
    capture_sizes: list[int] | None,
    max_num_reqs: int,
    max_num_tokens: int,
    cudagraph_mode: CUDAGraphMode,
) -> dict[int, int]:
    if not cudagraph_mode.has_full_cudagraphs():
        return {}
    if not capture_sizes:
        return {}

    capture_sizes = sorted(capture_sizes)
    # Limit the capture sizes to the max number of requests or tokens.
    upper_bound = (
        max_num_reqs
        if cudagraph_mode == CUDAGraphMode.FULL_DECODE_ONLY
        else max_num_tokens
    )
    capture_sizes = [x for x in capture_sizes if x <= upper_bound]
    if not capture_sizes:
        return {}

    cudagraph_sizes: dict[int, int] = {}
    for i in range(1, capture_sizes[-1] + 1):
        for x in capture_sizes:
            if i <= x:
                cudagraph_sizes[i] = x
                break
    return cudagraph_sizes

prepare_inputs_to_capture 

prepare_inputs_to_capture(
    num_reqs: int,
    num_tokens: int,
    input_buffers: InputBuffers,
    block_tables: BlockTables,
    attn_metadata_builders: list[AttentionMetadataBuilder],
    max_model_len: int,
    kv_cache_config: KVCacheConfig,
) -> dict[str, Any]
Source code in vllm/v1/worker/gpu/cudagraph_utils.py 
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
def prepare_inputs_to_capture(
    num_reqs: int,
    num_tokens: int,
    input_buffers: InputBuffers,
    block_tables: BlockTables,
    attn_metadata_builders: list[AttentionMetadataBuilder],
    max_model_len: int,
    kv_cache_config: KVCacheConfig,
) -> dict[str, Any]:
    num_tokens_per_req = num_tokens // num_reqs
    query_start_loc = input_buffers.query_start_loc
    query_start_loc.np[: num_reqs + 1] = np.arange(num_reqs + 1) * num_tokens_per_req
    query_start_loc.np[num_reqs:] = num_tokens
    query_start_loc.copy_to_gpu()
    seq_lens_np = np.full(num_reqs, max_model_len, dtype=np.int32)
    # HACK(woosuk): For faster warmup, we set seq_lens (GPU) to num_tokens
    # rather than max_model_len. This introduces a discrepancy between
    # seq_lens (on GPU) and seq_lens_np (on CPU), which may cause issues for
    # certain attention backends.
    input_buffers.seq_lens[:num_reqs] = num_tokens
    input_buffers.seq_lens[num_reqs:] = 0

    input_block_tables = [x[:num_reqs] for x in block_tables.input_block_tables]
    slot_mappings = block_tables.slot_mappings[:, :num_tokens]

    attn_metadata = build_attn_metadata(
        attn_metadata_builders=attn_metadata_builders,
        num_reqs=num_reqs,
        num_tokens=num_tokens,
        query_start_loc_gpu=query_start_loc.gpu[: num_reqs + 1],
        query_start_loc_cpu=query_start_loc.cpu[: num_reqs + 1],
        seq_lens=input_buffers.seq_lens,
        seq_lens_np=seq_lens_np,
        num_computed_tokens_cpu=None,  # FIXME
        block_tables=input_block_tables,
        slot_mappings=slot_mappings,
        kv_cache_config=kv_cache_config,
    )
    return attn_metadata