Revert "[Disagg][Perf] Use NPU event sync instead of blocking tolist#3194
Revert "[Disagg][Perf] Use NPU event sync instead of blocking tolist#3194Yikun merged 1 commit intovllm-project:mainfrom
Conversation
…to avoid unintentional copy ops blocking across different NPU streams, improving disagg TTIT/TTFT (vllm-project#2788)" This reverts commit 6995a7b.
|
👋 Hi! Thank you for contributing to the vLLM Ascend project. The following points will speed up your PR merge:
If CI fails, you can run linting and testing checks locally according Contributing and Testing. |
There was a problem hiding this comment.
Code Review
This pull request reverts a previous performance optimization that used NPU event synchronization to avoid blocking tolist() calls. The changes correctly remove the custom _to_list implementation and its associated test, reverting to the standard tolist() method. The revert is clean and complete, with all related code and tests being removed or reverted as expected. While this reintroduces a performance bottleneck for disaggregated setups, it is presumably done to fix a correctness or stability issue introduced by the optimization. No issues were found in the revert itself.
…llm-project#3194) …to avoid unintentional copy ops blocking across different NPU streams, improving disagg TTIT/TTFT (vllm-project#2788)" ### What this PR does / why we need it? This reverts commit 6995a7b. We'll add it back once the issue is fixed. related issue: vllm-project#3195 ### How was this patch tested? - vLLM version: v0.10.2 - vLLM main: vllm-project/vllm@52d0cb8 Signed-off-by: huangdong2022 <huangdong51@huawei.com>
…llm-project#3194) …to avoid unintentional copy ops blocking across different NPU streams, improving disagg TTIT/TTFT (vllm-project#2788)" ### What this PR does / why we need it? This reverts commit 6995a7b. We'll add it back once the issue is fixed. related issue: vllm-project#3195 ### How was this patch tested? - vLLM version: v0.10.2 - vLLM main: vllm-project/vllm@52d0cb8
…llm-project#3194) …to avoid unintentional copy ops blocking across different NPU streams, improving disagg TTIT/TTFT (vllm-project#2788)" ### What this PR does / why we need it? This reverts commit 6995a7b. We'll add it back once the issue is fixed. related issue: vllm-project#3195 ### How was this patch tested? - vLLM version: v0.10.2 - vLLM main: vllm-project/vllm@52d0cb8 Signed-off-by: luolun <luolun1995@cmbchina.com>
…llm-project#3194) …to avoid unintentional copy ops blocking across different NPU streams, improving disagg TTIT/TTFT (vllm-project#2788)" ### What this PR does / why we need it? This reverts commit 6995a7b. We'll add it back once the issue is fixed. related issue: vllm-project#3195 ### How was this patch tested? - vLLM version: v0.10.2 - vLLM main: vllm-project/vllm@52d0cb8 Signed-off-by: luolun <luolun1995@cmbchina.com>
…llm-project#3194) …to avoid unintentional copy ops blocking across different NPU streams, improving disagg TTIT/TTFT (vllm-project#2788)" ### What this PR does / why we need it? This reverts commit 6995a7b. We'll add it back once the issue is fixed. related issue: vllm-project#3195 ### How was this patch tested? - vLLM version: v0.10.2 - vLLM main: vllm-project/vllm@52d0cb8 Signed-off-by: hwhaokun <haokun0405@163.com>
…llm-project#3194) …to avoid unintentional copy ops blocking across different NPU streams, improving disagg TTIT/TTFT (vllm-project#2788)" ### What this PR does / why we need it? This reverts commit 6995a7b. We'll add it back once the issue is fixed. related issue: vllm-project#3195 ### How was this patch tested? - vLLM version: v0.10.2 - vLLM main: vllm-project/vllm@52d0cb8 Signed-off-by: nsdie <yeyifan@huawei.com>
…llm-project#3194) …to avoid unintentional copy ops blocking across different NPU streams, improving disagg TTIT/TTFT (vllm-project#2788)" ### What this PR does / why we need it? This reverts commit 6995a7b. We'll add it back once the issue is fixed. related issue: vllm-project#3195 ### How was this patch tested? - vLLM version: v0.10.2 - vLLM main: vllm-project/vllm@52d0cb8
…llm-project#3194) …to avoid unintentional copy ops blocking across different NPU streams, improving disagg TTIT/TTFT (vllm-project#2788)" ### What this PR does / why we need it? This reverts commit 6995a7b. We'll add it back once the issue is fixed. related issue: vllm-project#3195 ### How was this patch tested? - vLLM version: v0.10.2 - vLLM main: vllm-project/vllm@52d0cb8
…to avoid unintentional copy ops blocking across different NPU streams, improving disagg TTIT/TTFT (#2788)"
What this PR does / why we need it?
This reverts commit 6995a7b. We'll add it back once the issue is fixed.
related issue: #3195
How was this patch tested?