使用的harbor.sourcefind.cn:5443/dcu/admin/base/vllm:0.9.2-ubuntu22.04-dtk25.04.2-rocblas1121-das1.7-py3.10-20251203-bw官方镜像。 rocBLAS error: No Tensile solution found for { a: 0x145346c00000, a_type: "bf16_r", b: 0x145381400000, b_type: "bf16_r", c: 0x145366e00000, c_type: "bf16_r", d: 0x145366e00000, d_type: "bf16_r", compute_type: "f32_r", transA: 'N', transB: 'N', M: 4096, N: 10240, K: 1024, alpha: 1, row_stride_a: 1, col_stride_a: 4128, row_stride_b: 1, col_stride_b: 1024, row_stride_c: 1, col_stride_c: 4096, row_stride_d: 1, col_stride_d: 4096, beta: 0, batch_count: 1, strided_batch: true, stride_a: 1, stride_b: 1, stride_c: 1, stride_d: 1, offset_a: 0, offset_b: 0, offset_c: 0, offset_d: 0, atomics_mode: atomics_allowed } This message will be only be displayed once, unless the ROCBLAS_VERBOSE_TENSILE_ERROR environment variable is set. "/root/.cache/vllm/torch_compile_cache/0a6687b52b/rank_0_0/inductor_cache/gl/cgluglpy2beapc7vq7eo7bp7zhr4qzybou7huz5kvwt23ua2nqxp.py", line 289, in call ERROR 12-25 10:06:35 [core.py:603] extern_kernels.mm(buf2, arg4_1, out=buf3) ERROR 12-25 10:06:35 [core.py:603] RuntimeError: CUDA error: HIPBLAS_STATUS_NOT_SUPPORTED when calling hipblasGemmEx( handle, opa, opb, m, n, k, &falpha, a, HIP_R_16BF, lda, b, HIP_R_16BF, ldb, &fbeta, c, HIP_R_16BF, ldc, compute_type, HIPBLAS_GEMM_DEFAULT)