20231088/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

[CI][SpecDecode] Fix spec decode tests, use flash attention backend for spec decode CI tests. (#8975)

Browse Source
This commit is contained in:
Lily Liu 2024-09-30 17:51:40 -07:00 committed by GitHub
parent 1425a1bcf9
commit bce324487a
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 4 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
.buildkite/test-pipeline.yaml
View File

@ -207,8 +207,6 @@ steps:
- vllm/spec_decode - vllm/spec_decode
- tests/spec_decode - tests/spec_decode
commands: commands:
# See https://github.com/vllm-project/vllm/issues/5152
- export VLLM_ATTENTION_BACKEND=XFORMERS
- pytest -v -s spec_decode/e2e/test_multistep_correctness.py - pytest -v -s spec_decode/e2e/test_multistep_correctness.py
- pytest -v -s spec_decode --ignore=spec_decode/e2e/test_multistep_correctness.py - pytest -v -s spec_decode --ignore=spec_decode/e2e/test_multistep_correctness.py

5
tests/spec_decode/test_multi_step_worker.py
View File

@ -673,7 +673,10 @@ def test_use_draft_model_runner_advance_step():
worker.model_runner._gpu_advance_step.side_effect = ValueError( worker.model_runner._gpu_advance_step.side_effect = ValueError(
exception_secret) exception_secret)
seq_group_metadata_list, _, _ = create_batch(batch_size, k) seq_group_metadata_list, _, _ = create_batch(batch_size,
k,
block_size=block_size,
num_gpu_blocks=num_gpu_blocks)
# Fallback (should not call) when num_steps=1. # Fallback (should not call) when num_steps=1.
execute_model_req = ExecuteModelRequest( execute_model_req = ExecuteModelRequest(