[bugfix] fix the complex and potentially problematic generate_kv_idx. (#5957)

### What this PR does / why we need it?
In long-sequence scenarios, the chunked-prefill component may encounter
dimension misalignment issues, which previously occurred during
precision testing on the code_generate_lite dataset. This PR removes
redundant computations and instead derives the value using existing
results and straightforward calculations.
- vLLM version: v0.13.0
- vLLM main:
2c24bc6996

Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
This commit is contained in:
Qiu
2026-01-21 14:21:02 +08:00
committed by GitHub
parent 12a668b1d9
commit 58ff465821
5 changed files with 4 additions and 57 deletions

View File

@@ -70,8 +70,6 @@ class AscendPrefillContextParallelMetadata:
pcp_allgather_restore_idx: torch.Tensor = None
cp_kv_recover_idx_for_chunk: torch.Tensor = None
num_actual_tokens_pcp_padded: int = 0
num_computed_tokens_of_pcp_dcp: list[list[list[int]]] | None = None