# SPDX-License-Identifier: Apache-2.0 from contextlib import contextmanager, nullcontext from typing import Any, ContextManager, Optional import numpy as np import torch import torch.nn as nn import torch.nn.functional as F from vllm.config import (CompilationMode, CUDAGraphMode, VllmConfig, get_layers_from_vllm_config) from vllm.distributed.parallel_state import (get_pp_group, get_tp_group, get_world_group, init_model_parallel_group, patch_tensor_parallel_group) from vllm.forward_context import get_forward_context from vllm.logger import logger from vllm.model_executor.layers.attention_layer_base import AttentionLayerBase from vllm.model_executor.model_loader import get_model from vllm.model_executor.models import supports_multimodal from vllm.model_executor.models.deepseek_v2 import DeepseekV32IndexerCache from vllm.model_executor.models.llama_eagle3 import Eagle3LlamaForCausalLM from vllm.triton_utils import HAS_TRITON, triton from vllm.utils.math_utils import cdiv from vllm.utils.platform_utils import is_pin_memory_available from vllm.v1.attention.backends.utils import CommonAttentionMetadata from vllm.v1.core.sched.output import SchedulerOutput from vllm.v1.sample.metadata import SamplingMetadata from vllm.v1.spec_decode.eagle import PADDING_SLOT_ID from vllm.v1.spec_decode.eagle import EagleProposer as VllmEagleProposer from vllm.v1.spec_decode.metadata import SpecDecodeMetadata from vllm.v1.worker.gpu_input_batch import CachedRequestState, InputBatch from vllm_ascend.ascend_forward_context import set_ascend_forward_context from vllm_ascend.attention.attention_mask import AttentionMaskBuilder from vllm_ascend.attention.attention_v1 import AscendAttentionState from vllm_ascend.attention.utils import AscendCommonAttentionMetadata from vllm_ascend.compilation.acl_graph import (ACLGraphWrapper, update_attn_dcp_pcp_params, update_attn_params, update_mla_attn_dcp_pcp_params, update_mla_attn_params) from vllm_ascend.ops.rotary_embedding import update_cos_sin from vllm_ascend.ops.triton.spec_decode.utils import \ prepare_inputs_padded_kernel from vllm_ascend.ops.triton.triton_utils import get_vectorcore_num from vllm_ascend.utils import enable_sp, shared_expert_dp_enabled # Currently we will fix block size to a small one since `num_reqs` can't be too large _PREPARE_INPUTS_BLOCK_SIZE = 4 # TODO: Remove it when the bug of fx-graph is solved # patch vllm_config to be in CompilationMode.NONE temporarily @contextmanager def _maybe_eager_context(vllm_config): raw_compilation_config_mode = vllm_config.compilation_config.mode vllm_config.compilation_config.mode = CompilationMode.NONE try: yield finally: vllm_config.compilation_config.mode = raw_compilation_config_mode # split hidden states along dimension of sequence def split_inputs_tp_to_sp(hidden_states, out): # tp and sp share the same group group = get_tp_group() world_size = group.world_size rank = group.rank num_tokens = hidden_states.shape[0] # the size per rank after padded padded_num_tokens_per_rank = (num_tokens + world_size - 1) // world_size # compute the start and end of slice start = padded_num_tokens_per_rank * rank end = padded_num_tokens_per_rank * (rank + 1) # copy only hidden_states in current rank hidden_states_curr_rank = hidden_states[start:end] out[:hidden_states_curr_rank.shape[0]] = hidden_states_curr_rank return out[:padded_num_tokens_per_rank] class EagleProposer(VllmEagleProposer): def __init__(self, vllm_config: VllmConfig, device: torch.device, runner=None): super().__init__(vllm_config, device, runner) self.use_async_scheduling = self.vllm_config.scheduler_config.async_scheduling # there is synchronization between mtp steps when enabling aclgraph, # disable aclgraph when use async scheduling to avoid the # synchronization overhead. # NOTE: we need to set aclgraph_runtime_mode to None in both dummy_run # and _propose. self.use_cuda_graph = ( self.vllm_config.compilation_config.mode == CompilationMode.VLLM_COMPILE and not self.vllm_config.model_config.enforce_eager and not self.use_async_scheduling and not self.vllm_config.speculative_config.enforce_eager) self.cudagraph_batch_sizes = list( sorted( self.vllm_config.compilation_config.cudagraph_capture_sizes)) self.pcp_size = self.runner.pcp_size self.decode_threshold = 1 + self.num_speculative_tokens self.query_start_loc = self.runner._make_buffer( self.runner.max_num_reqs + 1, dtype=torch.int32) self.arange_cpu = torch.arange(self.arange.shape[0], device="cpu", dtype=torch.int32) self.attn_mask_builder = AttentionMaskBuilder(self.device) self.enable_shared_expert_dp = shared_expert_dp_enabled() self.dcp_size = self.runner.dcp_size self.pcp_rank = self.runner.pcp_rank self.dcp_rank = self.runner.dcp_rank self.use_aclgraph = self.runner._use_aclgraph() self.full_indices = range( self.runner.max_num_tokens * self.pcp_size * self.dcp_size + self.pcp_size * self.dcp_size * self.runner.max_num_reqs) self.use_sparse = hasattr(vllm_config.model_config.hf_text_config, "index_topk") # TODO: Remove it when the bug of fx-graph is solved self.maybe_eager_context: ContextManager[Any] = nullcontext() if not self.use_cuda_graph and enable_sp(vllm_config): self.maybe_eager_context = _maybe_eager_context(vllm_config) def load_model(self, model: nn.Module) -> None: target_attn_layer_names = set( get_layers_from_vllm_config(self.vllm_config, AttentionLayerBase).keys()) target_indexer_layer_names = set( get_layers_from_vllm_config(self.vllm_config, DeepseekV32IndexerCache).keys()) with self.maybe_eager_context: self.model = get_model(vllm_config=self.vllm_config, model_config=self.vllm_config. speculative_config.draft_model_config) indexer_layers = get_layers_from_vllm_config( self.vllm_config, DeepseekV32IndexerCache).keys() draft_attn_layer = get_layers_from_vllm_config( self.vllm_config, AttentionLayerBase).keys() draft_attn_layer_names = draft_attn_layer - target_attn_layer_names draft_indexer_layer_names = indexer_layers - target_indexer_layer_names draft_attn_layer_names = draft_attn_layer_names - draft_indexer_layer_names assert len(draft_attn_layer_names) == 1 self.attn_layer_name = list(draft_attn_layer_names) self.attn_layer_names = self.attn_layer_name # share embed_tokens with the target model if needed if get_pp_group().world_size == 1: if self.method == "mtp": if self.vllm_config.model_config.is_deepseek_mla and \ torch.equal(self.model.model.embed_tokens.weight, model.model.embed_tokens.weight): # If pp>1, the weights of mtp and the main model's embedding are not on the same device. # check if mtp model use main model's embedding and LMhead logger.info( "The MTP head shares the same vocab embedding" \ " with the target model." ) self.model.model.embed_tokens = model.model.embed_tokens else: logger.info( " The MTP head loaded its own vocab embedding" \ " weights instead of sharing them with the target model." ) else: logger.info( "The EAGLE head shares the same vocab embedding" \ " with the target model." ) self.model.model.embed_tokens = model.model.embed_tokens else: logger.info( "Since PP > 1 or other reasons the model head loaded its own vocab embedding" \ " weights instead of sharing them with the target model." ) # share lm_head with the target model if needed # some model definition do not define lm_head explicitly # and reuse embed_tokens for lm_head, e.g., CohereForCausalLM if self.method == "eagle" and hasattr(model, "lm_head"): logger.info("Loading EAGLE LM head weights from the target model.") if supports_multimodal(model): self.model.lm_head = model.get_language_model().lm_head else: self.model.lm_head = model.lm_head if self.method == "mtp" and \ self.vllm_config.model_config.is_deepseek_mla: for _, layer_module in self.model.model.layers.items(): if torch.equal(layer_module.shared_head.head.weight, model.lm_head.weight): layer_module.shared_head.head = model.lm_head if self.vllm_config.compilation_config.cudagraph_mode.has_full_cudagraphs( ) and self.use_cuda_graph: self.update_stream = torch.npu.Stream() self.model = ACLGraphWrapper(self.model, self.vllm_config, runtime_mode=CUDAGraphMode.FULL) def get_model(self) -> nn.Module: # get raw model out of the aclgraph wrapper. if isinstance(self.model, ACLGraphWrapper): return self.model.unwrap() return self.model @torch.inference_mode() def dummy_run(self, num_tokens: int, with_prefill: bool = False, in_graph_capturing: bool = False, num_reqs: int = 0, num_tokens_across_dp: Optional[torch.Tensor] = None, aclgraph_runtime_mode: CUDAGraphMode = CUDAGraphMode.NONE, batch_descriptor=None, dummy_compute_logits=lambda hidden_states: None, is_profile=False): # update global cos, sin update_cos_sin(self.positions[:num_tokens]) attn_metadata = None if not self.use_cuda_graph: aclgraph_runtime_mode = CUDAGraphMode.NONE if aclgraph_runtime_mode == CUDAGraphMode.FULL and len( self.runner.attn_groups) > 0: num_computed_tokens_cpu = ( self.runner.input_batch. num_computed_tokens_cpu_tensor[:num_reqs]) self.query_start_loc.cpu[:num_reqs + 1] = torch.tensor( [0] + self.runner.actual_seq_lengths_q[:num_reqs], device="cpu", dtype=torch.int32) self.query_start_loc.copy_to_gpu() common_attn_metadata = AscendCommonAttentionMetadata( query_start_loc=self.query_start_loc.gpu[:num_reqs + 1], query_start_loc_cpu=self.query_start_loc.cpu[:num_reqs + 1], seq_lens_cpu=self.runner.seq_lens.cpu, seq_lens=self.runner.seq_lens.gpu[:num_reqs], num_reqs=num_reqs, num_actual_tokens=num_tokens, num_input_tokens=num_tokens, max_query_len=self.num_speculative_tokens + 1, num_computed_tokens_cpu=num_computed_tokens_cpu, actual_seq_lengths_q=self.runner.actual_seq_lengths_q, block_table_tensor=self.runner.input_batch.block_table[0]. get_device_tensor()[:num_reqs], slot_mapping=self.runner.input_batch.block_table[0]. slot_mapping.gpu, positions=self.runner.positions.gpu, attn_state=self.runner.attn_state, decode_token_per_req=self.runner.decode_token_per_req, max_seq_len=0, ) builder = self.runner.attn_groups[0][0].get_metadata_builder() attn_metadata_eagle = builder.build_for_graph_capture( common_attn_metadata, AscendAttentionState.ChunkedPrefill) attn_metadata = {} for layer_name in self.attn_layer_name: attn_metadata[layer_name] = attn_metadata_eagle model_input_ids = self.input_ids[:num_tokens] model_positions = self.positions[:num_tokens] model_previous_hidden_states = self.hidden_states[:num_tokens] for i in range(self.num_speculative_tokens): if i > 0 and in_graph_capturing and aclgraph_runtime_mode == CUDAGraphMode.FULL: aclgraph_runtime_mode = CUDAGraphMode.NONE with set_ascend_forward_context( attn_metadata, self.vllm_config, num_tokens=num_tokens, num_actual_tokens=0, in_profile_run=is_profile, batch_descriptor=batch_descriptor, aclgraph_runtime_mode=aclgraph_runtime_mode, is_draft_model=True): forward_context = get_forward_context() if forward_context.sp_enabled: model_previous_hidden_states = split_inputs_tp_to_sp( model_previous_hidden_states, model_previous_hidden_states) self.model( input_ids=model_input_ids, positions=model_positions, hidden_states=model_previous_hidden_states, ) if (forward_context.cudagraph_runtime_mode == CUDAGraphMode.FULL and not forward_context.capturing): update_attn_params( self.update_stream, forward_context, num_tokens, self.vllm_config, ) if forward_context.sp_enabled: model_previous_hidden_states = torch.ops.vllm.maybe_all_gather_and_maybe_unpad( model_previous_hidden_states, True) dummy_compute_logits(self.hidden_states) def _propose( self, # [num_tokens] target_token_ids: torch.Tensor, # [num_tokens] or [3, num_tokens] when M-RoPE is enabled target_positions: torch.Tensor, # [num_tokens, hidden_size] target_hidden_states: torch.Tensor, # [batch_size] next_token_ids: torch.Tensor, last_token_indices: Optional[torch.Tensor], common_attn_metadata: CommonAttentionMetadata, sampling_metadata: SamplingMetadata, mm_embed_inputs: Optional[tuple[list[torch.Tensor], torch.Tensor]] = None, req_scheduled_tokens=None, long_seq_metadata=None, num_prefill_reqs=0, num_decode_reqs=0, scheduler_output: SchedulerOutput = None, num_scheduled_tokens: int = 0, ) -> torch.Tensor: num_tokens = target_token_ids.shape[0] batch_size = next_token_ids.shape[0] if last_token_indices is None: last_token_indices = common_attn_metadata.query_start_loc[1:] - 1 if self.method == "eagle3": assert isinstance(self.get_model(), Eagle3LlamaForCausalLM) target_hidden_states = self.model.combine_hidden_states( target_hidden_states) assert target_hidden_states.shape[-1] == self.hidden_size # Shift the input ids by one token. # E.g., [a1, b1, b2, c1, c2, c3] -> [b1, b2, c1, c2, c3, c3] self.input_ids[:num_tokens - 1] = target_token_ids[1:] # Replace the last token with the next token. # E.g., [b1, b2, c1, c2, c3, c3] -> [a2, b2, b3, c2, c3, c4] self.input_ids[last_token_indices] = next_token_ids if self.use_cuda_graph and \ num_tokens <= self.cudagraph_batch_sizes[-1]: num_input_tokens = self.vllm_config.pad_for_cudagraph(num_tokens) else: num_input_tokens = num_tokens has_lora = len(self.runner.input_batch.lora_id_to_lora_request) > 0 if self.use_cuda_graph: aclgraph_runtime_mode, batch_descriptor = \ self.runner.cudagraph_dispatcher.dispatch(num_tokens=num_input_tokens, uniform_decode=True, has_lora=has_lora) else: aclgraph_runtime_mode = CUDAGraphMode.NONE batch_descriptor = None # copy inputs to buffer for cudagraph self.positions[:num_tokens] = target_positions self.hidden_states[:num_tokens] = target_hidden_states # FIXME(woosuk): The below two ops cause synchronization. Optimize. builder = self.runner.attn_groups[0][0].get_metadata_builder() attn_metadata = builder.build(0, common_attn_metadata, self.runner.get_model()) # update global cos, sin update_cos_sin(self.positions[:num_input_tokens]) per_layer_attn_metadata = {} for layer_name in self.attn_layer_name: per_layer_attn_metadata[layer_name] = attn_metadata with set_ascend_forward_context( per_layer_attn_metadata, self.vllm_config, num_tokens=num_input_tokens, num_actual_tokens=num_tokens, batch_descriptor=batch_descriptor, aclgraph_runtime_mode=aclgraph_runtime_mode, is_draft_model=True): # The lifecycle of `input_ids`, `positions`, `hidden_states` runs through all speculative tokens' proposings. # `model_input_ids`, `model_positions` and `model_hidden_states` are used to represent the inputs of speculative model. model_input_ids = self.input_ids[:num_input_tokens] model_positions = self.positions[:num_input_tokens] model_hidden_states = self.hidden_states[:num_input_tokens] forward_context = get_forward_context() if forward_context.sp_enabled: # split hidden states along sequence dimension # positions should not be split? model_hidden_states = split_inputs_tp_to_sp( model_hidden_states, model_hidden_states) last_hidden_states, hidden_states = self.model( input_ids=model_input_ids, positions=model_positions, hidden_states=model_hidden_states, ) if forward_context.cudagraph_runtime_mode == CUDAGraphMode.FULL: # TODO: support mla in future. update_attn_params( self.update_stream, forward_context, num_input_tokens, self.vllm_config, ) if forward_context.sp_enabled: # merge hidden states along sequence dimension last_hidden_states = torch.ops.vllm.maybe_all_gather_and_maybe_unpad( last_hidden_states.contiguous(), True) hidden_states = torch.ops.vllm.maybe_all_gather_and_maybe_unpad( hidden_states.contiguous(), True) sample_hidden_states = last_hidden_states[last_token_indices] logits = self.model.compute_logits(sample_hidden_states) draft_token_ids = logits.argmax(dim=-1) # Early exit if there is only one draft token to be generated. if self.num_speculative_tokens == 1: # [batch_size, 1] return draft_token_ids.view(-1, 1) # Generate the remaining draft tokens. draft_token_ids_tensor = torch.zeros( (self.num_speculative_tokens, *draft_token_ids.shape), dtype=draft_token_ids.dtype, device=self.device) draft_token_ids_tensor[0] = draft_token_ids positions = target_positions[last_token_indices] hidden_states = hidden_states[last_token_indices] last_token_indices = self.arange[:batch_size] if self.use_cuda_graph and \ batch_size <= self.cudagraph_batch_sizes[-1]: input_batch_size = self.vllm_config.pad_for_cudagraph(batch_size) else: input_batch_size = batch_size attn_metadata.num_actual_tokens = batch_size attn_metadata.max_query_len = 1 attn_metadata.query_start_loc = self.arange_cpu[:batch_size + 1] attn_metadata.num_decodes, attn_metadata.num_prefills, attn_metadata.num_decode_tokens, attn_metadata.num_prefill_tokens = 0, batch_size, 0, batch_size attn_metadata.num_actual_tokens_pcp_padded = attn_metadata.num_decode_tokens + attn_metadata.num_prefill_tokens attn_metadata.actual_seq_lengths_q = attn_metadata.query_start_loc[ 1:].tolist() attn_metadata.seq_lens_list = attn_metadata.seq_lens.tolist() attn_metadata.attn_state = AscendAttentionState.ChunkedPrefill if self.use_cuda_graph: aclgraph_runtime_mode, batch_descriptor = \ self.runner.cudagraph_dispatcher.dispatch(num_tokens=input_batch_size, uniform_decode=True, has_lora=has_lora) else: aclgraph_runtime_mode = CUDAGraphMode.NONE batch_descriptor = None for now_speculative in range(self.num_speculative_tokens - 1): # Update the inputs. # cast to int32 is crucial when eagle model is compiled. # tensor.argmax() returns int64 by default. input_ids = draft_token_ids_tensor[now_speculative] positions += 1 # NOTE(woosuk): We should handle the case where the draft model # generates tokens beyond the max model length. Since it is complex # to remove such requests from the batch, we keep them in the batch # but adjust the position ids and slot mappings to avoid the # out-of-range access during the model execution. The draft tokens # generated with this adjustment should be ignored. exceeds_max_model_len = positions >= self.vllm_config.model_config.max_model_len # Mask out the position ids that exceed the max model length. # Otherwise, we may get out-of-range error in RoPE. clamped_positions = torch.where(exceeds_max_model_len, 0, positions) # TODO: Increment the sequence lengths. attn_metadata.seq_lens = attn_metadata.seq_lens + 1 attn_metadata.seq_lens_list = [ _ + 1 for _ in attn_metadata.seq_lens_list ] # TODO: Consider max model length. # attn_metadata.max_seq_len = min(attn_metadata.max_seq_len, # self.max_model_len) # For the requests that exceed the max model length, we set the # TODO: sequence length to 1 to minimize their overheads in attention. if self.attn_metadata_builder is None: attn_metadata_builder = self._get_attention_metadata_builder() else: attn_metadata_builder = self.attn_metadata_builder block_size = attn_metadata_builder.kv_cache_spec.block_size # Compute the slot mapping. block_numbers = (clamped_positions // block_size) block_ids = attn_metadata.block_tables.gather( dim=1, index=block_numbers.view(-1, 1)) block_ids = block_ids.view(-1) slot_mapping_tmp = (block_ids * block_size + clamped_positions % block_size) # Mask out the slot mappings that exceed the max model length. # Otherwise, the KV cache will be inadvertently updated with the # padding tokens. slot_mapping_tmp.masked_fill_(exceeds_max_model_len, PADDING_SLOT_ID) # NOTE: ASCEND slot_mapping must on cpu attn_metadata.slot_mapping[:slot_mapping_tmp.shape[0]].copy_( slot_mapping_tmp.to(torch.int32)) attn_metadata.slot_mapping[slot_mapping_tmp.shape[0]:].fill_( PADDING_SLOT_ID) # copy inputs to buffer for cudagraph self.input_ids[:batch_size] = input_ids self.positions[:batch_size] = clamped_positions self.hidden_states[:batch_size] = hidden_states attn_mask = self.attn_mask_builder.get_splitfuse_attn_mask() attn_metadata.attn_mask = attn_mask # update global cos, sin update_cos_sin(self.positions[:input_batch_size]) # Run the model. with set_ascend_forward_context( per_layer_attn_metadata, self.vllm_config, num_tokens=input_batch_size, num_actual_tokens=batch_size, batch_descriptor=batch_descriptor, aclgraph_runtime_mode=aclgraph_runtime_mode, is_draft_model=True): # The lifecycle of `input_ids`, `positions`, `hidden_states` runs through all speculative tokens' proposings. # `model_input_ids`, `model_positions` and `model_hidden_states` are used to represent the inputs of speculative model. model_input_ids = self.input_ids[:input_batch_size] model_positions = self.positions[:input_batch_size] model_hidden_states = self.hidden_states[:input_batch_size] forward_context = get_forward_context() if forward_context.sp_enabled: # split hidden states along sequence dimension # positions should not be split？ model_hidden_states = split_inputs_tp_to_sp( model_hidden_states, model_hidden_states) last_hidden_states, hidden_states = self.model( input_ids=model_input_ids, positions=model_positions, hidden_states=model_hidden_states, ) if forward_context.cudagraph_runtime_mode == CUDAGraphMode.FULL: update_attn_params( self.update_stream, forward_context, input_batch_size, self.vllm_config, ) if forward_context.sp_enabled: # merge hidden states along sequence dimension last_hidden_states = torch.ops.vllm.maybe_all_gather_and_maybe_unpad( last_hidden_states.contiguous(), True) hidden_states = torch.ops.vllm.maybe_all_gather_and_maybe_unpad( hidden_states.contiguous(), True) hidden_states = hidden_states[:batch_size] logits = self.model.compute_logits(last_hidden_states[:batch_size]) # TODO(wenlong): get more than one token for tree attention draft_token_ids = logits.argmax(dim=-1) draft_token_ids_tensor[now_speculative + 1] = draft_token_ids # [batch_size, num_speculative_tokens] draft_token_ids = draft_token_ids_tensor.swapaxes(0, 1) return draft_token_ids def prepare_next_token_ids_padded( self, common_attn_metadata: CommonAttentionMetadata, sampled_token_ids: torch.Tensor, requests: dict[str, CachedRequestState], gpu_input_batch: InputBatch, discard_request_indices: torch.Tensor, num_discarded_requests: int, ) -> tuple[torch.Tensor, torch.Tensor]: """ This function is used to prepare the inputs for speculative decoding. It calculates the next token ids and the number of valid sampled tokens for each request, considering the "discarded" requests whose next token is not sampled and comes from `request.get_token_id()` instead. It also accounts for the rejected tokens in `sampled_token_ids`. This function must use device functions to operate on the inputs, and should not introduce any blocking CPU-GPU synchronization. """ # TODO(Ben): Combine this into a custom fused kernel # Precompute get_token_id for when there is no valid next token num_reqs = gpu_input_batch.num_reqs self.backup_next_token_ids.np[:num_reqs] = np.array([ requests[gpu_input_batch.req_ids[i]].get_token_id( common_attn_metadata.seq_lens_cpu[i].item()) for i in range(num_reqs) ]) self.backup_next_token_ids.copy_to_gpu(num_reqs) # Mask out the sampled tokens indices that should not be sampled. discard_sampled_tokens_req_indices = discard_request_indices[: num_discarded_requests] valid_sampled_token_ids_gpu = sampled_token_ids.clone() valid_sampled_token_ids_gpu.index_fill_( 0, discard_sampled_tokens_req_indices, -1) # Generate a mask for all valid tokens within those requests valid_mask = (valid_sampled_token_ids_gpu != -1) & ( valid_sampled_token_ids_gpu < gpu_input_batch.vocab_size) # Count the number of valid tokens in each request valid_sampled_tokens_count = valid_mask.sum(dim=1) # Get the rightmost valid index per row last_valid_indices = valid_sampled_tokens_count - 1 last_valid_indices_safe = torch.clamp(last_valid_indices, min=0) # Get last valid token from each row # (assume undefined state where there is no valid token) selected_tokens = torch.gather( valid_sampled_token_ids_gpu, 1, last_valid_indices_safe.unsqueeze(1)).squeeze(1) # Use last token if valid, pre-computed backup if not batch_size = valid_sampled_token_ids_gpu.shape[0] next_token_ids = torch.where( last_valid_indices != -1, selected_tokens, self.backup_next_token_ids.gpu[:batch_size], ) return next_token_ids, valid_sampled_tokens_count def prepare_inputs( self, common_attn_metadata: CommonAttentionMetadata, sampled_token_ids: list[list[int]], num_draft_tokens: list[int], ) -> tuple[CommonAttentionMetadata, torch.Tensor]: """ This function is used to prepare the inputs for speculative decoding. It updates to the common_attn_metadata to account for the rejected tokens (and newly sampled tokens). It also returns the token indices of the tokens that should be fed to the speculator. """ # E.g. # common_attn_metadata.query_start_loc{_cpu}: # [0, q1, q1 + q2, q1 + q2 + q3] # common_attn_metadata.seq_lens{_cpu}: [s1, s2, s3] # num_rejected_tokens: [n1, n2, n3] # This function computes the intermediate values: # num_tokens_per_req: [q1 - n1, q2 - n2, q3 - n3] # And returns: # common_attn_metadata.query_start_loc{_cpu}: # [0, q1 - n1, q1 + q2 - n1 - n2, q1 + q2 + q3 - n1 - n2 - n3] # common_attn_metadata.seq_lens{_cpu}: # [s1 - n1 + 1, s2 - n2 + 1, s3 - n3 + 1] # token_indices: [0, 1, ..., q1 - n1 - 1, # q1, q1 + 1, ..., q1 + q2 - n2 - 1, # q1 + q2, q1 + q2 + 1, ..., q1 + q2 + q3 - n3 - 1] num_actual_reqs = len(num_draft_tokens) num_rejected_tokens = [ n + 1 - len(sampled_token_ids[i]) if n > 0 else 0 for i, n in enumerate(num_draft_tokens) ] num_rejected_tokens = torch.tensor(num_rejected_tokens, dtype=torch.int32) device = common_attn_metadata.query_start_loc.device query_start_loc_cpu = common_attn_metadata.query_start_loc_cpu[: num_actual_reqs + 1] seq_lens_cpu = common_attn_metadata.seq_lens_cpu[:num_actual_reqs] new_seq_lens_cpu = seq_lens_cpu - num_rejected_tokens # [0, q1, q1 + q2, q1 + q2 + q3] -> [q1, q2, q3] new_query_len_per_req = query_start_loc_cpu[ 1:] - query_start_loc_cpu[:-1] # [q1, q2, q3] -> [q1 - n1, q2 - n2, q3 - n3] new_num_tokens_per_req = new_query_len_per_req - num_rejected_tokens new_num_tokens_per_req_np = new_num_tokens_per_req.numpy() # [q1 - n1, q2 - n2, q3 - n3] -> # [0, q1 - n1, q1 + q2 - n1 - n2, q1 + q2 + q3 - n1 - n2 - n3] new_query_start_loc_cpu = torch.zeros( query_start_loc_cpu.shape, dtype=torch.int32, pin_memory=is_pin_memory_available(), ) new_query_start_loc_np = new_query_start_loc_cpu.numpy() np.cumsum(new_num_tokens_per_req_np, out=new_query_start_loc_np[1:]) total_num_tokens = new_query_start_loc_np[-1] # Example assuming num_tokens_per_req_np = [2, 4, 3] # this implies that `new_query_start_locs` is: # [0, 2, 6, 9] -> # [0, 0, 2, 2, 2, 2, 6, 6, 6] # _r1_ ____r2____ ___r3__ new_query_start_locs_expanded = np.repeat(new_query_start_loc_np[:-1], new_num_tokens_per_req_np) # [0, 1, 2, 3, 4, 5, 6, 7, 8] -> # [0, 1, 0, 1, 2, 3, 0, 1, 2] # _r1_ ____r2____ ___r3__ token_offests = (self.token_arange_np[:total_num_tokens] - new_query_start_locs_expanded) # Expand starting positions to match token pattern # [0, q1, q1 + q2] -> # [0, 0, q1, q1, q1, q1, q1 + q2, q1 + q2, q1 + q2] # _r1_ _____r2_______ ___________r3____________ old_query_start_locs_expanded = np.repeat( query_start_loc_cpu[:-1].numpy(), new_num_tokens_per_req_np) # Final token indices are: # [0, 1, // req 1 # q1 + 0, q1 + 1, q1 + 2, q1 + 3, // req 2 # q1 + q2 + 0, q1 + q2 + 1, q1 + q2 + 2] // req 3 token_indices_np = token_offests + old_query_start_locs_expanded token_indices = torch.from_numpy(token_indices_np).to( device, non_blocking=True) common_attn_metadata.slot_mapping[:token_indices.shape[0]].copy_( common_attn_metadata.slot_mapping[token_indices]) common_attn_metadata.slot_mapping[token_indices.shape[0]:].fill_(-1) # NOTE: Currently positions and seq_lens are not used in attn forward # so we do not need to fixed them. But if they are used in the future, # we should fixed them. spec_common_attn_metadata = AscendCommonAttentionMetadata( query_start_loc=new_query_start_loc_cpu.to(device, non_blocking=True), query_start_loc_cpu=new_query_start_loc_cpu, seq_lens=new_seq_lens_cpu.to(device, non_blocking=True), seq_lens_cpu=new_seq_lens_cpu, num_computed_tokens_cpu=common_attn_metadata. num_computed_tokens_cpu, num_reqs=common_attn_metadata.num_reqs, num_actual_tokens=total_num_tokens, num_input_tokens=common_attn_metadata.num_input_tokens, max_query_len=new_query_len_per_req.max().item(), block_table_tensor=common_attn_metadata.block_table_tensor, slot_mapping=common_attn_metadata.slot_mapping, actual_seq_lengths_q=self.runner.actual_seq_lengths_q, positions=common_attn_metadata.positions[token_indices], attn_state=self.runner.attn_state, decode_token_per_req=self.runner.decode_token_per_req, max_seq_len=0) return spec_common_attn_metadata, token_indices def prepare_inputs_padded( self, common_attn_metadata: CommonAttentionMetadata, spec_decode_metadata: SpecDecodeMetadata, valid_sampled_tokens_count: torch.Tensor, ) -> tuple[CommonAttentionMetadata, torch.Tensor, torch.Tensor]: """ This function is used to prepare the inputs for speculative decoding It updates the common_attn_metadata for speculative decoding, but does not consider the rejected tokens. Instead, all tokens are included as inputs to the speculator, with the rejected tokens used as padding and filtered out later by `token_indices_to_sample`. No blocking CPU operations should be introduced in this function. """ if HAS_TRITON: num_reqs = common_attn_metadata.num_reqs device = valid_sampled_tokens_count.device token_indices_to_sample = torch.empty((num_reqs, ), dtype=torch.int32, device=device) num_blocks_needed = triton.cdiv(num_reqs, _PREPARE_INPUTS_BLOCK_SIZE) num_vector_core = get_vectorcore_num() grid_size = min(num_blocks_needed, num_vector_core) grid = (grid_size, ) prepare_inputs_padded_kernel[grid]( spec_decode_metadata.cu_num_draft_tokens, valid_sampled_tokens_count, common_attn_metadata.query_start_loc, token_indices_to_sample, num_reqs, BLOCK_SIZE=_PREPARE_INPUTS_BLOCK_SIZE, ) else: num_draft_tokens_gpu = torch.cat([ spec_decode_metadata.cu_num_draft_tokens[0:1], spec_decode_metadata.cu_num_draft_tokens[1:] - spec_decode_metadata.cu_num_draft_tokens[:-1], ]) num_rejected_tokens_gpu = torch.where( num_draft_tokens_gpu > 0, num_draft_tokens_gpu + 1 - valid_sampled_tokens_count, torch.zeros_like(num_draft_tokens_gpu), ) token_indices_to_sample = ( common_attn_metadata.query_start_loc[1:] - 1 - num_rejected_tokens_gpu) query_start_loc_cpu = common_attn_metadata.query_start_loc_cpu new_query_len_per_req = query_start_loc_cpu[ 1:] - query_start_loc_cpu[:-1] total_num_tokens = query_start_loc_cpu[-1].item() token_indices = self.arange[:total_num_tokens] # NOTE: Currently positions and seq_lens are not used in attn forward # so we do not need to fixed them. But if they are used in the future, # we should fixed them. spec_common_attn_metadata = AscendCommonAttentionMetadata( query_start_loc=common_attn_metadata.query_start_loc, query_start_loc_cpu=query_start_loc_cpu, seq_lens_cpu=common_attn_metadata.seq_lens_cpu, num_reqs=common_attn_metadata.num_reqs, num_actual_tokens=common_attn_metadata.num_actual_tokens if self.pcp_size > 1 else total_num_tokens, num_input_tokens=common_attn_metadata.num_input_tokens, max_query_len=new_query_len_per_req.max().item(), actual_seq_lengths_q=self.runner.actual_seq_lengths_q, block_table_tensor=common_attn_metadata.block_table_tensor, slot_mapping=common_attn_metadata.slot_mapping, positions=common_attn_metadata.positions, attn_state=self.runner.attn_state, decode_token_per_req=self.runner.decode_token_per_req, num_computed_tokens_cpu=common_attn_metadata. num_computed_tokens_cpu, seq_lens=common_attn_metadata.seq_lens, max_seq_len=0) return spec_common_attn_metadata, token_indices, token_indices_to_sample def _split_pcp_input(self, req_scheduled_tokens, input_ids, target_hidden_states): """ Split prefill input_ids and target_hidden_states in pcp group. 1. input_ids padding: [t0, t1, t2, t3, t4, t5] -> [t0, t1, t2, t3, t4, t5, pad, pad] 2. split input_ids: pcp0 [t0, t1, pad, pad], pcp1 [t2, t3, t4, t5] 3. split target_hidden_states (already include pcp padding): [h0, h1, h2, h3, h4, h5, pad, pad] -> pcp0 [h0, h1, pad, pad], pcp1 [h2, h3, h4, h5] 4. also update max_query_len, seq_lens, cu_num_tokens according to pcp split. """ if len(req_scheduled_tokens) == 0: # no prefill inputs to split, return empty result return ( 0, torch.zeros([0], device='npu'), torch.zeros([0, target_hidden_states.size(1)], device='npu'), 0, torch.zeros([0]), torch.tensor([0], dtype=torch.int32), ) def _pcp_pad_and_split(num_tokens): num_pcp_padded_scheduled_tokens = cdiv( num_tokens, 2 * self.pcp_size) * 2 * self.pcp_size pcp_pad = num_pcp_padded_scheduled_tokens - num_tokens chunk_size = num_pcp_padded_scheduled_tokens // (2 * self.pcp_size) # split position_ids (and use split position_ids to split input_ids afterwards) req_position_cp: list[int] = [] req_position_cp.extend( self.full_indices[self.pcp_rank * chunk_size:(self.pcp_rank + 1) * chunk_size]) req_position_cp.extend( self.full_indices[num_pcp_padded_scheduled_tokens - (self.pcp_rank + 1) * chunk_size:num_pcp_padded_scheduled_tokens - self.pcp_rank * chunk_size]) return req_position_cp, num_pcp_padded_scheduled_tokens, pcp_pad num_pcp_scheduled_tokens = [] ori_start_index = 0 pad_start_index = 0 pcp_split_input_ids_list = [] pcp_split_hidden_states_list = [] for ori_num_tokens in req_scheduled_tokens.values(): req_position_pcp, num_pcp_padded_scheduled_tokens, num_pcp_pad = \ _pcp_pad_and_split(ori_num_tokens) actual_num_tokens = len(req_position_pcp) num_pcp_scheduled_tokens.append(actual_num_tokens) pad_input_ids = F.pad( input_ids[ori_start_index:ori_start_index + ori_num_tokens], (0, num_pcp_pad)) ori_start_index += ori_num_tokens pcp_chunk_indices = [ pad_start_index + pos for pos in req_position_pcp ] pcp_split_input_ids = pad_input_ids[req_position_pcp] pcp_split_hidden_states = target_hidden_states[pcp_chunk_indices] pcp_split_input_ids_list.append(pcp_split_input_ids) pcp_split_hidden_states_list.append(pcp_split_hidden_states) pad_start_index += num_pcp_padded_scheduled_tokens num_tokens = sum(num_pcp_scheduled_tokens) input_ids = torch.cat(pcp_split_input_ids_list) target_hidden_states = torch.cat(pcp_split_hidden_states_list, dim=0) max_query_len = max(num_pcp_scheduled_tokens) seq_lens = torch.tensor(num_pcp_scheduled_tokens, dtype=torch.int32) cu_num_tokens = torch.tensor( np.insert(np.cumsum(np.array(num_pcp_scheduled_tokens)), 0, 0)) return num_tokens, input_ids, target_hidden_states, max_query_len, seq_lens, cu_num_tokens # update full-graph params for one spec token def _update_full_graph_params(self, forward_context, num_tokens): if self.vllm_config.model_config.use_mla: if self.pcp_size * self.dcp_size > 1: update_mla_attn_dcp_pcp_params(self.update_stream, forward_context, num_tokens) else: update_mla_attn_params(self.update_stream, forward_context, num_tokens, self.vllm_config.speculative_config) else: if self.pcp_size * self.dcp_size > 1: update_attn_dcp_pcp_params(self.update_stream, forward_context, num_tokens) else: update_attn_params(self.update_stream, forward_context, num_tokens, self.vllm_config)