Platform: ARM Platform
  Device: Mali-T628
    Driver version  : 1.2 (Linux ARM)
    Compute units   : 2
    Clock frequency : 600 MHz

    Global memory bandwidth (GBPS)
      float   : 2.25
      float2  : 2.87
      float4  : 6.41
      float8  : 6.15
      float16 : 4.49

    Single-precision compute (GFLOPS)
      float   : 8.23
      float2  : 2.94
      float4  : 18.27
      float8  : 17.38
      float16 : 3.61

    Double-precision compute (GFLOPS)
      double   : 1.78
      double2  : 0.84
      double4  : 8.59
      double8  : 8.60
      double16 : 8.57

    Integer compute (GIOPS)
      int   : 1.41
      int2  : 2.96
      int4  : 2.98
      int8  : 3.68
      int16 : 17.28

    Transfer bandwidth (GBPS)
      enqueueWriteBuffer         : 4.70
      enqueueReadBuffer          : 2.74
      enqueueMapBuffer(for read) : 475.44
        memcpy from mapped ptr   : 2.17
      enqueueUnmap(after write)  : 654.24
        memcpy to mapped ptr     : 2.21

    Kernel launch latency : 206.34 us