1080Ti 检查失败：error == cudaSuccess (8 vs. 0) invalid device function

Question

我在py-faster-Rcnn中得到了这句话

F0205 17:04:17.806836  4300 pooling_layer.cu:212] Check failed: error == cudaSuccess (8 vs. 0)  invalid device function
*** Check failure stack trace: ***

显卡：Geforce 1080Ti

CUDA：7.5

Cudnn : v5

我的Makefile.config

CUDA_ARCH := -gencode arch=compute_20,code=sm_20 \
        -gencode arch=compute_20,code=sm_21 \
        -gencode arch=compute_30,code=sm_30 \
        -gencode arch=compute_35,code=sm_35 \
        -gencode arch=compute_50,code=sm_50 \
        -gencode arch=compute_52,code=sm_52 \
        -gencode arch=compute_60,code=sm_60 \
        -gencode arch=compute_61,code=sm_61 \
        -gencode arch=compute_61,code=compute_61

我能做什么？

Answer 1

安装 CUDA 8.0 应该可以解决问题。

从this可以看出GeForce GTX 1080 Ti GPU属于Pascal GPU Micro架构，计算能力版本6.1，支持CUDA 8.0或更高版本。

CUDA SDK 8.0 support for compute capability 2.0 – 6.x (Fermi, Kepler, Maxwell, Pascal), last version with support for Compute Capability 2.x (Fermi)

CUDA SDK 7.5 support for compute capability 2.0 – 5.x (Fermi, Kepler, Maxwell)

1080Ti 检查失败：error == cudaSuccess (8 vs. 0) invalid device function

1080Ti Check failed: error == cudaSuccess (8 vs. 0) invalid device function

caffe

pycaffe

cudnn