SDAccel Profile Summary


Generated on: 2019-02-02 00:51:07

Profiled application: Xilinx_fpgacnn_best.exe

Target platform: Xilinx

Target devices: xilinx:adm-pcie-7v3:1ddr:3.0-0

Flow mode: System Run

Tool version: 2016.3



OpenCL API Calls

API Name Number Of Calls Total Time (ms) Minimum Time (ms) Average Time (ms) Maximum Time (ms)
clCreateProgramWithBinary12288.472288.472288.472288.47
clCreateContext11701.631701.631701.631701.63
clFinish1292.491292.491292.491292.491
clEnqueueWriteBuffer21.445480.3618330.7227391.08365
clEnqueueReadBuffer10.8313350.8313350.8313350.831335
clGetPlatformIDs30.1193240.0088270.03977470.066607
clSetKernelArg60.080210.0096690.01336830.027591
clCreateBuffer30.0701620.0146430.02338730.038768
clEnqueueNDRangeKernel10.0411380.0411380.0411380.041138
clReleaseMemObject30.0401840.009770.01339470.02058
clCreateKernel10.029470.029470.029470.02947
clGetEventProfilingInfo20.0219220.0091610.0109610.012761
clGetDeviceInfo20.0204140.0084890.0102070.011925
clGetPlatformInfo20.0195940.0084930.0097970.011101
clBuildProgram10.01810.01810.01810.0181
clCreateCommandQueue10.0147360.0147360.0147360.014736
clReleaseProgram10.0110840.0110840.0110840.011084
clReleaseKernel10.0110030.0110030.0110030.011003
clReleaseCommandQueue10.0109150.0109150.0109150.010915
clReleaseContext10.0106870.0106870.0106870.010687
clGetDeviceIDs10.0103350.0103350.0103350.010335

Kernel Execution

Kernel Number Of Enqueues Total Time (ms) Minimum Time (ms) Average Time (ms) Maximum Time (ms)
good_cnn1291.977291.977291.977291.977

Compute Unit Utilization

Device Compute Unit Kernel Global Work Size Local Work Size Number Of Calls Total Time (ms) Minimum Time (ms) Average Time (ms) Maximum Time (ms)
xilinx:adm-pcie-7v3:1ddr:3.0-0good_cnn_1good_cnn4:1:11:1:11290.751290.751290.751290.751
xilinx:adm-pcie-7v3:1ddr:3.0-0good_cnn_2good_cnn4:1:11:1:11290.859290.859290.859290.859
xilinx:adm-pcie-7v3:1ddr:3.0-0good_cnn_3good_cnn4:1:11:1:11291.721291.721291.721291.721
xilinx:adm-pcie-7v3:1ddr:3.0-0good_cnn_4good_cnn4:1:11:1:11291.684291.684291.684291.684

Data Transfer: Host and Global Memory

Context:Number of Devices Transfer Type Number Of Transfers Transfer Rate (MB/s) Average Bandwidth Utilization (%) Average Size (KB) Total Time (ms) Average Time (ms)
context0:1READ11152.86999012.009062746.4960.6475110.647511
context0:1WRITE21376.16350314.335036684.3840.9946260.497313

Data Transfer: Kernels and Global Memory

Device Transfer Type Number Of Transfers Transfer Rate (MB/s) Average Bandwidth Utilization (%) Average Size (KB) Average Time (ns)
xilinx:adm-pcie-7v3:1ddr:3.0-0READ279486.126070.05317770.064686.174
xilinx:adm-pcie-7v3:1ddr:3.0-0WRITE118632.556690.02219350.062926459.989

Top Data Transfer: Kernels and Global Memory

Device Kernel Name Number of Transfers Average Bytes per Transfer Transfer Efficiency (%) Total Data Transfer (MB) Total Write (MB) Total Read (MB) Transfer Rate (MB/s) Average Bandwidth Utilization (%)
xilinx:adm-pcie-7v3:1ddr:3.0-0ALL3981163.68011.554692.535170.7464961.788674.341380.0376856