mlzxy / caffe-fpga-opencl Goto Github PK

Dear BenBBear,
When I run the 'Step by Step of a MNIST Demo' with FPGA, I compiled the net.tcl with SDAccel.But i encountered an warnning:
WARN: READ THIS!!

• If you build you net.cl kernel with / without DEBUG flags, then you have to set -l debug / -l [no/net/layer].
• If you build you net.cl with / without pingpong buffer, then you have to run the program in (-v 12) / (-v 20) correspondingly.
  I chose to continue,and the error shows like this:
  *** Error in `network_fpga.exe': malloc(): memory corruption (fast): 0x000000000a7dc140 ***
  ERROR: Kernel execution failed. Kernel exited with the code 6
  Do you have a solution? Thank you very much.

Hello, I am trying to run the project with NVIDIA GPU and ubuntu16.04. I use CMake tools to create the Makefile, make successfully and create a opencl_learning.exe. But I come across some problems when using it. As the picture showing following, ERROR: From clGetDeviceIDs at platform: NVIDIA CUDA -> CL_DEVICE_NOT_FOUND occurs.

Any suggestions will be appreciated!
By the ways, I also try to make in the CPU-only computer and I have the following problems when make the Makefile, which is created by CMake tools.