Platform: ARM Platform
  Device: Mali-T628
    Driver version  : 1.1 (Linux ARM)
    Compute units   : 4
    Clock frequency : 600 MHz

    Global memory bandwidth (GBPS)
      float   : 4.32
      float2  : 6.29
      float4  : 6.61
      float8  : 6.33
      float16 : 5.07

    Single-precision compute (GFLOPS)
      float   : 2.61
      float2  : 5.84
      float4  : 6.20
      float8  : 7.23
      float16 : 7.20

    Double-precision compute (GFLOPS)
      double   : 3.73
      double2  : 12.91
      double4  : 9.40
      double8  : 16.53
      double16 : 16.00

    Integer compute (GIOPS)
      int   : 2.10
      int2  : 5.89
      int4  : 5.87
      int8  : 7.35
      int16 : 34.14

    Transfer bandwidth (GBPS)
      enqueueWriteBuffer         : 4.72
      enqueueReadBuffer          : 2.73
      enqueueMapBuffer(for read) : 556.23
        memcpy from mapped ptr   : 2.15
      enqueueUnmap(after write)  : 724.13
        memcpy to mapped ptr     : 2.24

    Kernel launch latency : 179.17 us