Platform: AMD Accelerated Parallel Processing
  Device: Loveland
    Driver version : 1113.2 (Linux x86)
    Compute units  : 2

    Global memory bandwidth (GBPS)
      float   : 6.92
      float2  : 7.16
      float4  : 7.03
      float8  : 3.62
      float16 : 1.91

    Single-precision compute (GFLOPS)
      float   : 15.74
      float2  : 31.40
      float4  : 62.36
      float8  : 62.60
      float16 : 63.09

    No double precision support! Skipped

    Integer compute (GIOPS)
      int   : 7.87
      int2  : 15.72
      int4  : 15.71
      int8  : 15.70
      int16 : 15.71

    Transfer bandwidth (GBPS)
      enqueueWriteBuffer         : 2.94
      enqueueReadBuffer          : 1.93
      enqueueMapBuffer(for read) : 1462.07
        memcpy from mapped ptr   : 1.91
      enqueueUnmap(after write)  : 263.46
        memcpy to mapped ptr     : 1.98

    Kernel launch latency : 460.72 us

  Device: AMD E-350 Processor
    Driver version : 1113.2 (sse2) (Linux x86)
    Compute units  : 2

    Global memory bandwidth (GBPS)
      float   : 2.41
      float2  : 1.67
      float4  : 2.40
      float8  : 2.22
      float16 : 2.19

    Single-precision compute (GFLOPS)
      float   : 1.28
      float2  : 1.28
      float4  : 5.04
      float8  : 5.03
      float16 : 1.54

    Double-precision compute (GFLOPS)
      double   : 0.91
      double2  : 1.82
      double4  : 1.82
      double8  : 1.82
      double16 : 0.58

    Integer compute (GIOPS)
      int   : 1.59
      int2  : 0.66
      int4  : 0.73
      int8  : 0.73
      int16 : 0.75

    Transfer bandwidth (GBPS)
      enqueueWriteBuffer         : 3.00
      enqueueReadBuffer          : 1.54
      enqueueMapBuffer(for read) : 9304.52
        memcpy from mapped ptr   : 1.55
      enqueueUnmap(after write)  : 6804.45
        memcpy to mapped ptr     : 1.48

    Kernel launch latency : 183.70 us