[错误]: bf16 MatMul/GEMM 返回损坏的输出 (cos 0.28-0.65, 5-135 倍的幅值) - f16 和 f32 控制器均正确
作者: paulalesius创建于 2026年9月18日更新于 2026年9月18日
标签bugsupport_request
| item | value |
|---|---|
| OpenVINO | 2026.4.0 (pip wheel 2026.4.0-22959-99c81491cc3-releases/2026/4), Python API (openvino, openvino.opset13) |
| OS | Ubuntu 24.04 LTS, x86_64, kernel 7.2.5 |
| CPU | AMD Ryzen 9 9950X (16 cores / 32 threads) |
| CPU ISA | AVX512-F, AVX512-BF16 (avx512_bf16), AVX512-VNNI, F16C; no Intel AMX (no amx_tile/amx_bf16/amx_int8), no AVX10.2 |
| Python deps | torch 2.14.0+cpu, numpy 2.4.6, ml_dtypes (real-model reference only) |
The CPU is AMD, so the bf16 GEMM exercises the AVX512-BF16 oneDNN path (the Intel AMX bf16 path is not present on this hardware). This may be relevant to where the bug lives.
内容来源: openvinotoolkit/openvino