EngineX/xc-llm-ascend
3
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

mfix bug when max_seqs=14 in mtp=2 scenario and raise error when cudagraph_capture_sizes can't be an integer multiple of uniform_decode_query_lentp (#3910)

Browse Source 
### What this PR does / why we need it?
1. Revert [bugfix for mtp in
fullgraph](0948483642)
and support it when vllm supports
2. raise error when cudagraph_capture_sizes can't be an integer multiple
of uniform_decode_query_len
3. bugfix when max_num_seqs=14 in mtp=2 scenario

### Does this PR introduce _any_ user-facing change?
no

### How was this patch tested?

- vLLM version: v0.11.0
- vLLM main:
83f478bb19

---------

Signed-off-by: zouyida2052 <zouyida2002@gmail.com>
This commit is contained in:
zouyida2052
2025-10-31 09:24:50 +08:00
committed by GitHub
parent 35a913cf1e
commit 1966885be2
4 changed files with 63 additions and 61 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
vllm_ascend/worker/model_runner_v1.py
View File

@@ -4035,7 +4035,23 @@ class NPUModelRunner(LoRAModelRunnerMixin):
if aclgraph_mode.decode_mode() == CUDAGraphMode.FULL and \
aclgraph_mode.separate_routine():
compilation_cases_decode = sorted(self.aclgraph_batch_sizes)
max_num_tokens = self.scheduler_config.max_num_seqs * \
self.uniform_decode_query_len
decode_cudagraph_batch_sizes = [
x for x in self.aclgraph_batch_sizes if x <= max_num_tokens
and x >= self.uniform_decode_query_len
]
compilation_cases_decode = sorted(decode_cudagraph_batch_sizes)
# TODO: refactor this when vLLM supports mtp>1
if not all(x % self.uniform_decode_query_len == 0
for x in decode_cudagraph_batch_sizes):
raise ValueError(
"In the MTP fullgraph scenario, each graph size must be an integer multiple of "
f"(num_speculative_tokens + 1): {self.uniform_decode_query_len}. "
f"Please modify the cudagraph_capture_sizes variable to be integer multiple of {self.uniform_decode_query_len}, "
f"while ensuring the maximum cudagraph_capture_sizes does not exceed max_num_seqs * (num_speculative_tokens + 1): {max_num_tokens}. "
"For example, with MTP=2 and max_num_seqs=16, we recommend setting cudagraph_capture_sizes to [48]."
)
self._capture_aclgraphs(
compilation_cases=compilation_cases_decode,
aclgraph_runtime_mode=CUDAGraphMode.FULL,