Platform: Portable Computing Language
  Device: pthread-AMD EPYC 7A53 64-Core Processor
    Driver version  : 1.8 (Linux x64)
    Compute units   : 128
    Clock frequency : 2000 MHz
64 warnings generated.

    Global memory bandwidth (GBPS)
      float   : 30.74
      float2  : 31.35
      float4  : 31.11
      float8  : 33.88
      float16 : 27.41

    Single-precision compute (GFLOPS)
      float   : 81.79
      float2  : 160.46
      float4  : 308.39
      float8  : 589.43
      float16 : 107.23

    No half precision support! Skipped

    Double-precision compute (GFLOPS)
      double   : 81.46
      double2  : 159.08
      double4  : 293.62
      double8  : 55.83
      double16 : 78.10

    Integer compute (GIOPS)
      int   : 198.51
      int2  : 378.71
      int4  : 750.36
      int8  : 1407.42
      int16 : 2162.47

    Integer compute Fast 24bit (GIOPS)
      int   : 122.67
      int2  : 176.17
      int4  : 344.37
      int8  : 617.54
      int16 : 113.10

    Transfer bandwidth (GBPS)
      enqueueWriteBuffer              : 16.47
      enqueueReadBuffer               : 16.79
      enqueueWriteBuffer non-blocking : 18.28
      enqueueReadBuffer non-blocking  : 18.62
      enqueueMapBuffer(for read)      : 5142.44
        memcpy from mapped ptr        : 15.46
      enqueueUnmap(after write)       : 6599.52
        memcpy to mapped ptr          : 20.98

    Kernel launch latency : 107.54 us