xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-04-17 01:37:04 +08:00
Code Issues Packages Projects Releases Wiki Activity

Add compatibility notes and docs links to MTP/PCP error messages

Browse Source 
- Add documentation links to MTP and PCP error messages for consistency
  with DCP error message
- Add notes indicating no backends currently support these features
- Remove suggestion to use --attention-backend for PCP since no
  backends support it yet

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
Signed-off-by: yurekami <yurekami@users.noreply.github.com>
This commit is contained in:
yurekami 2025-12-25 04:08:54 +09:00
parent 79e0db60ee
commit 4b7df5710a
1 changed files with 10 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
vllm/v1/worker/cp_utils.py
View File

@ -34,10 +34,13 @@ def check_attention_cp_compatibility(vllm_config: VllmConfig) -> None:
f"current attention backend "
f"'{layer_impl.__class__.__name__}'.\n\n"
f"To resolve this issue, try one of the following:\n"
f" 1. Use a different attention backend by specifying:\n"
f" --attention-backend <backend>\n"
f" 2. Set cp_kv_cache_interleave_size to 1\n"
f" 3. Disable speculative decoding"
f" 1. Set cp_kv_cache_interleave_size to 1\n"
f" 2. Disable speculative decoding\n\n"
f"Note: No backends currently support MTP with "
f"cp_kv_cache_interleave_size > 1.\n\n"
f"For more information, see:\n"
f" https://docs.vllm.ai/en/latest/serving/"
f"distributed_serving.html"
)
if dcp_size > 1 and not layer_impl.need_to_return_lse_for_decode:
@ -65,11 +68,10 @@ def check_attention_cp_compatibility(vllm_config: VllmConfig) -> None:
f"backend that supports PCP. The current backend "
f"'{layer_impl.__class__.__name__}' does not support this "
f"feature.\n\n"
f"To resolve this issue, try one of the following:\n"
f" 1. Use a compatible attention backend by specifying:\n"
f" --attention-backend <backend>\n"
f" 2. Disable PCP by removing the "
f"To resolve this issue:\n"
f" Disable PCP by removing the "
f"--prefill-context-parallel-size flag\n\n"
f"Note: No backends currently support PCP.\n\n"
f"For more information, see:\n"
f" https://docs.vllm.ai/en/latest/serving/"
f"distributed_serving.html"