ValueError: The buffer size in the given workspace is insufficient for the given problem size. Buffer: 1179648 bytes, Required: 3932160 bytes.

用户在使用 vLLM 0.23.1rc1 或更新 nightly 版本(2026-07-01 之后)部署单节点 8×H200 GPU,运行 zai-org/GLM-5.2-FP8(753B MoE)模型,设置 --tensor-parallel-size 8 时触发。报错出现在 CUDA graph

![[Bug]: meta-llama/Llama-3.2-1B-Instruct Fails With ROCM_ATTN Due To Seg Fault](https://www.chat-gpts.plus/wp-content/uploads/2026/07/36180-1b1e4943-768x403.jpg)






