Platform: AMD Accelerated Parallel Processing
  Device: Tahiti
    Driver version  : 1445.5 (VM) (Linux x64)
    Compute units   : 32
    Clock frequency : 1125 MHz

    Global memory bandwidth (GBPS)
      float   : 238.56
      float2  : 247.07
      float4  : 232.13
      float8  : 131.79
      float16 : 66.94

    Single-precision compute (GFLOPS)
      float   : 3817.28
      float2  : 3703.25
      float4  : 3654.96
      float8  : 3611.17
      float16 : 3605.42

    Double-precision compute (GFLOPS)
      double   : 907.47
      double2  : 895.83
      double4  : 885.86
      double8  : 884.52
      double16 : 880.63

    Integer compute (GIOPS)
      int   : 810.54
      int2  : 811.31
      int4  : 810.97
      int8  : 793.67
      int16 : 784.04

    Transfer bandwidth (GBPS)
      enqueueWriteBuffer         : 9.13
      enqueueReadBuffer          : 5.61
      enqueueMapBuffer(for read) : 18512.79
        memcpy from mapped ptr   : 4.22
      enqueueUnmap(after write)  : 40826.69
        memcpy to mapped ptr     : 4.44

    Kernel launch latency : 35.82 us