diff --git a/bench/data.js b/bench/data.js index 86bb8164f4..5f8dbdbfd8 100644 --- a/bench/data.js +++ b/bench/data.js @@ -92881,8 +92881,368 @@ window.BENCHMARK_DATA = { "extra": "gctime=0\nmemory=0\nallocs=0\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":52,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" } ] + }, + { + "commit": { + "author": { + "email": "61313342+luraess@users.noreply.github.com", + "name": "Ludovic Räss", + "username": "luraess" + }, + "committer": { + "email": "noreply@github.com", + "name": "GitHub", + "username": "web-flow" + }, + "distinct": true, + "id": "774abc6a096390ff1c1038b4754bff246c9d979a", + "message": "Fix `resize!` when `pool=none` is in use (#2613)", + "timestamp": "2025-01-10T08:34:26+01:00", + "tree_id": "87425b9c8a180fa413d1fff2783d759570ee3b32", + "url": "https://github.com/JuliaGPU/CUDA.jl/commit/774abc6a096390ff1c1038b4754bff246c9d979a" + }, + "date": 1736502050345, + "tool": "julia", + "benches": [ + { + "name": "latency/precompile", + "value": 45532671418, + "unit": "ns", + "extra": "gctime=0\nmemory=1360\nallocs=46\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":60,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "latency/ttfp", + "value": 6382276443.5, + "unit": "ns", + "extra": "gctime=0\nmemory=1360\nallocs=46\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":60,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "latency/import", + "value": 3039078540.5, + "unit": "ns", + "extra": "gctime=0\nmemory=1360\nallocs=46\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":30,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "integration/volumerhs", + "value": 9567627, + "unit": "ns", + "extra": "gctime=0\nmemory=432\nallocs=22\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "integration/byval/slices=1", + "value": 146713, + "unit": "ns", + "extra": "gctime=0\nmemory=1136\nallocs=36\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "integration/byval/slices=3", + "value": 425286, + "unit": "ns", + "extra": "gctime=0\nmemory=2304\nallocs=72\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "integration/byval/reference", + "value": 144622, + "unit": "ns", + "extra": "gctime=0\nmemory=336\nallocs=17\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "integration/byval/slices=2", + "value": 286077, + "unit": "ns", + "extra": "gctime=0\nmemory=1728\nallocs=54\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "integration/cudadevrt", + "value": 103283, + "unit": "ns", + "extra": "gctime=0\nmemory=336\nallocs=17\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "kernel/indexing", + "value": 14073, + "unit": "ns", + "extra": "gctime=0\nmemory=480\nallocs=19\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "kernel/indexing_checked", + "value": 15126, + "unit": "ns", + "extra": "gctime=0\nmemory=480\nallocs=19\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "kernel/occupancy", + "value": 710.5460992907801, + "unit": "ns", + "extra": "gctime=0\nmemory=0\nallocs=0\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":141,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "kernel/launch", + "value": 2120.3, + "unit": "ns", + "extra": "gctime=0\nmemory=64\nallocs=2\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":10,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "kernel/rand", + "value": 14743, + "unit": "ns", + "extra": "gctime=0\nmemory=352\nallocs=13\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/reverse/1d", + "value": 19325.5, + "unit": "ns", + "extra": "gctime=0\nmemory=1184\nallocs=31\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/reverse/2d", + "value": 24669, + "unit": "ns", + "extra": "gctime=0\nmemory=1488\nallocs=40\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/reverse/1d_inplace", + "value": 10913.666666666666, + "unit": "ns", + "extra": "gctime=0\nmemory=752\nallocs=18\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":3,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/reverse/2d_inplace", + "value": 11253, + "unit": "ns", + "extra": "gctime=0\nmemory=800\nallocs=16\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/copy", + "value": 20229, + "unit": "ns", + "extra": "gctime=0\nmemory=432\nallocs=19\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/iteration/findall/int", + "value": 157863.5, + "unit": "ns", + "extra": "gctime=0\nmemory=11280\nallocs=391\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/iteration/findall/bool", + "value": 138404.5, + "unit": "ns", + "extra": "gctime=0\nmemory=8608\nallocs=290\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/iteration/findfirst/int", + "value": 153375, + "unit": "ns", + "extra": "gctime=0\nmemory=3520\nallocs=111\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/iteration/findfirst/bool", + "value": 154273, + "unit": "ns", + "extra": "gctime=0\nmemory=3520\nallocs=111\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/iteration/scalar", + "value": 75697, + "unit": "ns", + "extra": "gctime=0\nmemory=1552\nallocs=73\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/iteration/logical", + "value": 212853.5, + "unit": "ns", + "extra": "gctime=0\nmemory=14784\nallocs=529\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/iteration/findmin/1d", + "value": 41543, + "unit": "ns", + "extra": "gctime=0\nmemory=2064\nallocs=79\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/iteration/findmin/2d", + "value": 93933.5, + "unit": "ns", + "extra": "gctime=0\nmemory=7840\nallocs=259\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/reductions/reduce/1d", + "value": 35999, + "unit": "ns", + "extra": "gctime=0\nmemory=2000\nallocs=78\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/reductions/reduce/2d", + "value": 41907.5, + "unit": "ns", + "extra": "gctime=0\nmemory=928\nallocs=35\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/reductions/mapreduce/1d", + "value": 33891.5, + "unit": "ns", + "extra": "gctime=0\nmemory=2000\nallocs=78\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/reductions/mapreduce/2d", + "value": 41528, + "unit": "ns", + "extra": "gctime=0\nmemory=928\nallocs=35\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/broadcast", + "value": 21376, + "unit": "ns", + "extra": "gctime=0\nmemory=1920\nallocs=59\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/copyto!/gpu_to_gpu", + "value": 11516, + "unit": "ns", + "extra": "gctime=0\nmemory=224\nallocs=14\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/copyto!/cpu_to_gpu", + "value": 210665, + "unit": "ns", + "extra": "gctime=0\nmemory=80\nallocs=5\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/copyto!/gpu_to_cpu", + "value": 243223.5, + "unit": "ns", + "extra": "gctime=0\nmemory=80\nallocs=5\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/accumulate/1d", + "value": 108164, + "unit": "ns", + "extra": "gctime=0\nmemory=6640\nallocs=212\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/accumulate/2d", + "value": 79823.5, + "unit": "ns", + "extra": "gctime=0\nmemory=2048\nallocs=72\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/construct", + "value": 1284.3, + "unit": "ns", + "extra": "gctime=0\nmemory=208\nallocs=5\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":10,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/random/randn/Float32", + "value": 49740, + "unit": "ns", + "extra": "gctime=0\nmemory=3024\nallocs=106\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/random/randn!/Float32", + "value": 26117, + "unit": "ns", + "extra": "gctime=0\nmemory=512\nallocs=24\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/random/rand!/Int64", + "value": 27030, + "unit": "ns", + "extra": "gctime=0\nmemory=176\nallocs=7\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/random/rand!/Float32", + "value": 8836.333333333334, + "unit": "ns", + "extra": "gctime=0\nmemory=80\nallocs=5\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":3,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/random/rand/Int64", + "value": 37762.5, + "unit": "ns", + "extra": "gctime=0\nmemory=384\nallocs=12\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/random/rand/Float32", + "value": 13046, + "unit": "ns", + "extra": "gctime=0\nmemory=288\nallocs=10\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/permutedims/4d", + "value": 66810, + "unit": "ns", + "extra": "gctime=0\nmemory=3584\nallocs=87\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/permutedims/2d", + "value": 56518, + "unit": "ns", + "extra": "gctime=0\nmemory=2464\nallocs=81\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/permutedims/3d", + "value": 59273.5, + "unit": "ns", + "extra": "gctime=0\nmemory=2800\nallocs=80\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/sorting/1d", + "value": 2933200.5, + "unit": "ns", + "extra": "gctime=0\nmemory=53520\nallocs=1673\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/sorting/by", + "value": 3500043, + "unit": "ns", + "extra": "gctime=0\nmemory=53520\nallocs=1673\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "array/sorting/2d", + "value": 1084935, + "unit": "ns", + "extra": "gctime=0\nmemory=9600\nallocs=298\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":1,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "cuda/synchronization/stream/auto", + "value": 1035.9, + "unit": "ns", + "extra": "gctime=0\nmemory=0\nallocs=0\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":10,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "cuda/synchronization/stream/nonblocking", + "value": 6536.8, + "unit": "ns", + "extra": "gctime=0\nmemory=16\nallocs=1\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":5,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "cuda/synchronization/stream/blocking", + "value": 791.2244897959183, + "unit": "ns", + "extra": "gctime=0\nmemory=0\nallocs=0\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":98,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "cuda/synchronization/context/auto", + "value": 1182.9, + "unit": "ns", + "extra": "gctime=0\nmemory=48\nallocs=1\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":10,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "cuda/synchronization/context/nonblocking", + "value": 6769.6, + "unit": "ns", + "extra": "gctime=0\nmemory=48\nallocs=2\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":5,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + }, + { + "name": "cuda/synchronization/context/blocking", + "value": 915.2666666666667, + "unit": "ns", + "extra": "gctime=0\nmemory=0\nallocs=0\nparams={\"gctrial\":true,\"time_tolerance\":0.05,\"evals_set\":false,\"samples\":10000,\"evals\":45,\"gcsample\":false,\"seconds\":5,\"overhead\":0,\"memory_tolerance\":0.01}" + } + ] } ] }, - "lastUpdate": 1736444676693 + "lastUpdate": 1736502062497 } \ No newline at end of file