doonny / PipeCNN

An OpenCL-based FPGA Accelerator for Convolutional Neural Networks
Apache License 2.0
1.26k stars 370 forks source link

Software emulation OpenCL runtime error,can not return a result in executing Layer 1, Xilinx/Vitis #175

Open zhangxian314 opened 1 year ago

zhangxian314 commented 1 year ago

/home/zhangxian/mywork/CNN-A/CNN-A/Emulation-SW/CNN-A /home/zhangxian/mywork/CNN-A/CNN-A_system/Emulation-SW/binary_container_1.xclbin


PipeCNN: An OpenCL-Based FPGA Accelerator for CNNs


***Xilinx

Platform: Xilinx Totally 1 device(s) are found Using Device 0: xilinx_u50_gen3x16_xdma_201920_3 Device OpenCL Version: OpenCL 1.0 Device Max Compute Units: 0 Device Max WorkGroup Size: 4294967295 Device Max WorkItem Size: 4294967295 Device Global Memory Size: 8192 MBytes Device Local Memory Size: 16 KBytes Device Max Clock Freq: 300 Mhz

138455872 total weights read 1024 total output reference read

==========================11111111============================1 ==========================22222222============================ ===============un use opencv================================ ==========load image path===============/home/zhangxian/mywork/CNN-A/CNN-A/src/data/data_vgg/image.dat

150528 bytes image data read from binary files

Executing Layer 1:

Launching single work-item kernel winbuffer

Launching single work-item kernel Conv

Launching single work-item kernel MemWr ====================333333333============================ ====================333333333=======pool_on===================== ====================44444444============333333================

It can not return a result in executing Layer 1, so the next layer can not be executed.