ubuntu cuda 输出全0

[frspring]

平台(如果交叉编译请再附上交叉编译目标平台):

Platform(Include target platform as well if cross-compiling):

Ubuntu20.04

Github版本:

Github Version: 2.0.1

编译方式:

Compiling Method

本地编译

请在这里粘贴cmake参数或使用的cmake脚本路径以及完整输出
-- Use Threadpool, forbid openmp
-- >>>>>>>>>>>>>
-- MNN BUILD INFO:
--      System: Linux
--      Processor: x86_64
--      Version: 2.0.1
--      Metal: OFF
--      OpenCL: ON
--      OpenGL: OFF
--      Vulkan: OFF
--      ARM82: OFF
--      oneDNN: OFF
--      TensorRT: OFF
--      CoreML: OFF
--      CUDA: ON
--      OpenMP: OFF
--      BF16: OFF
--      ThreadPool: ON
--      Hidden: TRUE
--      Build Path: /home/SENSETIME/chuguanyi1/Software/MNN/build
-- WIN_USE_ASM: 
-- x86_64: Open SSE
-- MNN_AVX512:ON
-- Enabling CUDA support (version: 11.2, archs: )
-- message  -D_FORCE_INLINES -Wno-deprecated-gpu-targets -w -O3;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75 !!!!!!!!!!!
-- MNN Depends:MNN
-- Configuring done
-- Generating done```

在ubuntu20.04上使用cuda进行inference，输出全为0。
显卡是GTX1060，cuda版本是11.4。

[rlczddl]

同样是ubuntu,看了下配置选项和你一样,我的cuda后端却没法用

[Captain-1314]

我也是这个问题，不知道为什么输出全为0

[wangzhaode]

测试的什么程序？

[bitxsw93]

Please update lastest mnn code, retry again. If having any problem, please let me know.

[frspring]

Please update lastest mnn code, retry again. If having any problem, please let me know.

update to 2.3.0 and it worked

[wangzhaode]

Please update lastest mnn code, retry again. If having any problem, please let me know.

update to 2.3.0 and it worked

Done!