biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Attention][1/n] Remove usage of deprecated seq_lens_cpu and num_computed_tokens_cpu CommonAttentionMetadata properties (#31773)

Browse Source 
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
This commit is contained in:
Lucas Wilkinson
2026-01-06 07:05:17 -05:00
committed by GitHub
parent 14df02b4e1
commit e0327c9db2
9 changed files with 23 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/v1/attention/test_mla_backends.py
View File

@@ -154,12 +154,12 @@ def create_and_prepopulate_kv_cache(
MLA KV cache tensor
"""
batch_size = len(kv_c_contexts)
seq_lens = common_attn_metadata.seq_lens_cpu
seq_lens = common_attn_metadata.seq_lens.cpu()
query_lens = (
common_attn_metadata.query_start_loc_cpu[1:]
- common_attn_metadata.query_start_loc_cpu[:-1]
)
context_lens = common_attn_metadata.num_computed_tokens_cpu
context_lens = seq_lens - query_lens
block_table = common_attn_metadata.block_table_tensor
slot_mapping = common_attn_metadata.slot_mapping