1 2Platform: AMD Accelerated Parallel Processing 3 Device: Kalindi 4 Driver version : 1214.3 (VM) (Linux x64) 5 Compute units : 2 6 7 Global memory bandwidth (GBPS) 8 float : 6.60 9 float2 : 6.71 10 float4 : 6.45 11 float8 : 3.51 12 float16 : 1.83 13 14 Single-precision compute (GFLOPS) 15 float : 100.63 16 float2 : 101.26 17 float4 : 100.94 18 float8 : 100.32 19 float16 : 99.08 20 21 Double-precision compute (GFLOPS) 22 double : 6.35 23 double2 : 6.37 24 double4 : 6.36 25 double8 : 6.34 26 double16 : 6.32 27 28 Integer compute (GIOPS) 29 int : 20.33 30 int2 : 20.39 31 int4 : 20.36 32 int8 : 20.33 33 int16 : 20.32 34 35 Transfer bandwidth (GBPS) 36 enqueueWriteBuffer : 1.80 37 enqueueReadBuffer : 1.98 38 enqueueMapBuffer(for read) : 84.42 39 memcpy from mapped ptr : 1.81 40 enqueueUnmap(after write) : 54.32 41 memcpy to mapped ptr : 1.87 42 43 Kernel launch latency : 138.08 us 44 45 Device: AMD A6-1450 APU with Radeon(TM) HD Graphics 46 Driver version : 1214.3 (sse2,avx) (Linux x64) 47 Compute units : 4 48 49 Global memory bandwidth (GBPS) 50 float : 1.97 51 float2 : 2.51 52 float4 : 1.95 53 float8 : 2.79 54 float16 : 3.54 55 56 Single-precision compute (GFLOPS) 57 float : 1.30 58 float2 : 2.50 59 float4 : 5.01 60 float8 : 9.21 61 float16 : 1.07 62 63 Double-precision compute (GFLOPS) 64 double : 0.62 65 double2 : 1.35 66 double4 : 2.56 67 double8 : 6.27 68 double16 : 2.44 69 70 Integer compute (GIOPS) 71 int : 1.60 72 int2 : 1.22 73 int4 : 4.70 74 int8 : 8.08 75 int16 : 7.91 76 77 Transfer bandwidth (GBPS) 78 enqueueWriteBuffer : 2.67 79 enqueueReadBuffer : 2.03 80 enqueueMapBuffer(for read) : 13489.22 81 memcpy from mapped ptr : 2.02 82 enqueueUnmap(after write) : 26446.84 83 memcpy to mapped ptr : 2.03 84 85 Kernel launch latency : 32.74 us 86 87