51c8f60eb0
### What this PR does / why we need it?
Previously, the dummy run executed compute_logits only once, regardless
of num_speculative_tokens. This caused execute_model to hang on
compute_logits when lm head tensor parallelism exceeded 1. The fix
ensures compute_logits executes correctly during dummy run, matching
num_speculative_tokens.
I set the `non_blocking` argument to False when moving
`exceeds_max_model_len` to the CPU. From what I understand, using
`non_blocking=True` and immediately accessing the tensor on the CPU can
cause accuracy problems. However, this issue doesn't happen when
transferring data to a device. ref:
https://discuss.pytorch.org/t/should-we-set-non-blocking-to-true/38234/18
- vLLM version: v0.11.0
- vLLM main:
2918c1b49c
---------
Signed-off-by: Jade Zheng <zheng.shoujian@outlook.com>
74 lines
2.6 KiB
Python
74 lines
2.6 KiB
Python
import numpy as np
|
|
import torch
|
|
from vllm.config import CUDAGraphMode
|
|
from vllm.v1.spec_decode.ngram_proposer import \
|
|
NgramProposer as VllmNgramProposer
|
|
|
|
from vllm_ascend.spec_decode.interface import Proposer, SpecDcodeType
|
|
|
|
|
|
class NgramProposer(VllmNgramProposer, Proposer):
|
|
|
|
def __init__(self, vllm_config, device, runner):
|
|
super().__init__(vllm_config)
|
|
self.name = SpecDcodeType.NGRAM
|
|
self.device = device
|
|
self.runner = runner
|
|
|
|
def load_model(self, *args, **kwargs):
|
|
# No model to load.
|
|
pass
|
|
|
|
@torch.inference_mode()
|
|
def dummy_run(self,
|
|
num_tokens,
|
|
with_prefill=None,
|
|
skip_attn=None,
|
|
num_reqs=None,
|
|
num_tokens_across_dp=None,
|
|
aclgraph_runtime_mode: CUDAGraphMode = CUDAGraphMode.NONE,
|
|
batch_descriptor=None,
|
|
dummy_compute_logits=lambda hidden_states: None):
|
|
pass
|
|
|
|
def generate_token_ids(self,
|
|
valid_sampled_token_ids: list[np.ndarray],
|
|
sampling_metadata=None,
|
|
scheduler_output=None,
|
|
spec_decode_metadata=None,
|
|
positions=None,
|
|
num_scheduled_tokens=None,
|
|
hidden_states=None,
|
|
attn_metadata=None,
|
|
aux_hidden_states=None) -> list[list[int]]:
|
|
valid_ngram_requests = []
|
|
for i, sampled_ids in enumerate(valid_sampled_token_ids):
|
|
num_sampled_ids = sampled_ids.shape[0]
|
|
if not num_sampled_ids:
|
|
continue
|
|
|
|
req_id = self.runner.input_batch.req_ids[i]
|
|
if req_id in self.runner.input_batch.spec_decode_unsupported_reqs:
|
|
continue
|
|
|
|
num_tokens = self.runner.input_batch.num_tokens_no_spec[i]
|
|
if num_tokens >= self.runner.input_batch.max_model_len:
|
|
# Skip requests that have already reached the max model length.
|
|
continue
|
|
|
|
start_idx = self.runner.input_batch.num_tokens_no_spec[i]
|
|
end_idx = start_idx + num_sampled_ids
|
|
self.runner.input_batch.token_ids_cpu[
|
|
i, start_idx:end_idx] = sampled_ids
|
|
|
|
valid_ngram_requests.append(i)
|
|
|
|
draft_token_ids = self.batch_propose(
|
|
len(valid_sampled_token_ids),
|
|
valid_ngram_requests,
|
|
self.runner.input_batch.num_tokens_no_spec,
|
|
self.runner.input_batch.token_ids_cpu,
|
|
)
|
|
|
|
return draft_token_ids
|