Platform: Intel(R) OpenCL Graphics
  Device: Intel(R) Arc(TM) Graphics
    Driver version  : 24.52.032224 (Linux x64)
    Compute units   : 128
    Clock frequency : 2350 MHz

    Global memory bandwidth (GBPS)
      float   : 65.34
      float2  : 70.95
      float4  : 69.16
      float8  : 72.31
      float16 : 55.66

    Single-precision compute (GFLOPS)
      float   : 4766.24
      float2  : 4719.63
      float4  : 4753.81
      float8  : 4728.60
      float16 : 4486.08

    Half-precision compute (GFLOPS)
      half   : 9462.56
      half2  : 9396.65
      half4  : 9368.41
      half8  : 9377.19
      half16 : 9276.17

    Double-precision compute (GFLOPS)
      double   : 149.36
      double2  : 146.96
      double4  : 148.73
      double8  : 147.58
      double16 : 145.10

    Integer compute (GIOPS)
      int   : 1248.38
      int2  : 1195.06
      int4  : 1195.74
      int8  : 1203.09
      int16 : 1194.81

    Integer compute Fast 24bit (GIOPS)
      int   : 1197.94
      int2  : 1197.13
      int4  : 1197.40
      int8  : 1205.89
      int16 : 1189.96

    Integer char (8bit) compute (GIOPS)
      char   : 2885.35
      char2  : 2888.01
      char4  : 2893.66
      char8  : 2824.15
      char16 : 2709.35

    Integer short (16bit) compute (GIOPS)
      short   : 7428.87
      short2  : 7079.24
      short4  : 7150.68
      short8  : 7161.08
      short16 : 6937.62

    Transfer bandwidth (GBPS)
      enqueueWriteBuffer              : 22.87
      enqueueReadBuffer               : 22.78
      enqueueWriteBuffer non-blocking : 31.54
      enqueueReadBuffer non-blocking  : 31.37
      enqueueMapBuffer(for read)      : 25.32
        memcpy from mapped ptr        : 21.26
      enqueueUnmap(after write)       : 32.01
        memcpy to mapped ptr          : 20.99

    Kernel launch latency : 38.31 us