求助：gromacs运行mdrun时检测到GPU但无法使用

Tttgb

MD日志中关于GPU提示架构问题
  GPU info:
    Number of GPUs detected: 1
    #0: NVIDIA Quadro K2200, compute cap.: 5.0, ECC:  no, stat: not in set of targeted devices

WARNING: The gmx mdrun binary does not include support for the CUDA architecture
of the GPU ID #0 (compute capability 5.0) detected during detection. By default,
GROMACS supports all architectures of compute capability >= 3.5, so your GPU
might be rare, or some architectures were disabled in the build. Consult the
install guide for how to use the GMX_CUDA_TARGET_SM and GMX_CUDA_TARGET_COMPUTE
CMake variables to add this architecture.


显卡信息：


安装时的代码：
cmake .. -DCMAKE_INSTALL_PREFIX=/usr/local/gmx2022.2 -DGMX_USE_RDTSCP=ON -DGMX_GPU=CUDA -DCUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda-11.7

看了下gromacs关于CUDA的描述，也没弄明白应该怎么解决。
By default, code will be generated for the most common CUDA architectures. However, to reduce build time and binary size we do not generate code for every single possible architecture, which in rare cases (say, Tegra systems) can result in the default build not being able to use some GPUs. If this happens, or if you want to remove some architectures to reduce binary size and build time, you can alter the target CUDA architectures. This can be done either with the GMX_CUDA_TARGET_SM or GMX_CUDA_TARGET_COMPUTE CMake variables, which take a semicolon delimited string with the two digit suffixes of CUDA (virtual) architectures names, for instance “35;50;51;52;53;60”. For details, see the “Options for steering GPU code generation” section of the nvcc man / help or Chapter 6. of the nvcc manual.

sobereva

这么低端的GPU拿来加速也没多大意义