diff --git a/Cargo.toml b/Cargo.toml index 98ba373c68f..7c3c52bb9da 100644 --- a/Cargo.toml +++ b/Cargo.toml @@ -54,6 +54,11 @@ small_rng = [] # using min-const-generics min_const_gen = [] +# Option: use unbiased sampling. By default, bias affecting no more than one in +# 2^48 samples (for uniform distributions) is accepted. +# Note: enabling this option is expected to affect reproducibility of results. +unbiased = [] + [workspace] members = [ "rand_core", @@ -80,6 +85,12 @@ features = ["into_bits"] libc = { version = "0.2.22", optional = true, default-features = false } [dev-dependencies] +criterion = "0.4.0" +num-traits = "0.2.14" rand_pcg = { path = "rand_pcg", version = "0.3.0" } # Only to test serde1 bincode = "1.2.1" + +[[bench]] +name = "uniform" +harness = false diff --git a/benches/uniform.rs b/benches/uniform.rs new file mode 100644 index 00000000000..dc51e7e40ed --- /dev/null +++ b/benches/uniform.rs @@ -0,0 +1,78 @@ +// Copyright 2021 Developers of the Rand project. +// +// Licensed under the Apache License, Version 2.0 or the MIT license +// , at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +//! Implement benchmarks for uniform distributions over integer types + +use core::time::Duration; +use criterion::{criterion_group, criterion_main, BenchmarkId, Criterion}; +use rand::distributions::uniform::{SampleRange, Uniform}; +use rand::prelude::*; +use rand_chacha::ChaCha8Rng; +use rand_pcg::{Pcg32, Pcg64}; + +const WARM_UP_TIME: Duration = Duration::from_millis(1000); +const MEASUREMENT_TIME: Duration = Duration::from_secs(3); +const SAMPLE_SIZE: usize = 100_000; +const N_RESAMPLES: usize = 10_000; + +macro_rules! sample { + ($R:ty, $T:ty, $U:ty, $g:expr) => { + $g.bench_function(BenchmarkId::new(stringify!($R), "single"), |b| { + let mut rng = <$R>::from_entropy(); + let x = rng.gen::<$U>(); + let bits = (<$T>::BITS / 2); + let mask = (1 as $U).wrapping_neg() >> bits; + let range = (x >> bits) * (x & mask); + let low = <$T>::MIN; + let high = low.wrapping_add(range as $T); + + b.iter(|| (low..=high).sample_single(&mut rng)); + }); + + $g.bench_function(BenchmarkId::new(stringify!($R), "distr"), |b| { + let mut rng = <$R>::from_entropy(); + let x = rng.gen::<$U>(); + let bits = (<$T>::BITS / 2); + let mask = (1 as $U).wrapping_neg() >> bits; + let range = (x >> bits) * (x & mask); + let low = <$T>::MIN; + let high = low.wrapping_add(range as $T); + let dist = Uniform::<$T>::new_inclusive(<$T>::MIN, high); + + b.iter(|| dist.sample(&mut rng)); + }); + }; + + ($c:expr, $T:ty, $U:ty) => {{ + let mut g = $c.benchmark_group(concat!("sample", stringify!($T))); + g.sample_size(SAMPLE_SIZE); + g.warm_up_time(WARM_UP_TIME); + g.measurement_time(MEASUREMENT_TIME); + g.nresamples(N_RESAMPLES); + sample!(SmallRng, $T, $U, g); + sample!(ChaCha8Rng, $T, $U, g); + sample!(Pcg32, $T, $U, g); + sample!(Pcg64, $T, $U, g); + g.finish(); + }}; +} + +fn sample(c: &mut Criterion) { + sample!(c, i8, u8); + sample!(c, i16, u16); + sample!(c, i32, u32); + sample!(c, i64, u64); + sample!(c, i128, u128); +} + +criterion_group! { + name = benches; + config = Criterion::default(); + targets = sample +} +criterion_main!(benches); diff --git a/results-uniform-int-5800X-4/distr-random-i128.svg b/results-uniform-int-5800X-4/distr-random-i128.svg new file mode 100644 index 00000000000..3a74e0e0c25 --- /dev/null +++ b/results-uniform-int-5800X-4/distr-random-i128.svg @@ -0,0 +1,1658 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i128/Pcg64/Lemire + + + + + distr_random_i128/Pcg64/sample-unbiased + + + + + distr_random_i128/Pcg64/Canon-Red-Un + + + + + distr_random_i128/Pcg64/Canon-Red + + + + + distr_random_i128/Pcg64/sample + + + + + distr_random_i128/Pcg32/Lemire + + + + + distr_random_i128/Pcg32/sample-unbiased + + + + + distr_random_i128/Pcg32/Canon-Red-Un + + + + + distr_random_i128/Pcg32/Canon-Red + + + + + distr_random_i128/Pcg32/sample + + + + + distr_random_i128/ChaCha8Rng/Lemire + + + + + distr_random_i128/ChaCha8Rng/sample-unbiased + + + + + distr_random_i128/ChaCha8Rng/Canon-Red-Un + + + + + distr_random_i128/ChaCha8Rng/Canon-Red + + + + + distr_random_i128/ChaCha8Rng/sample + + + + + distr_random_i128/SmallRng/Lemire + + + + + distr_random_i128/SmallRng/sample-unbiased + + + + + distr_random_i128/SmallRng/Canon-Red-Un + + + + + distr_random_i128/SmallRng/Canon-Red + + + + + distr_random_i128/SmallRng/sample + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 40 + + + + + + + + + + + + + 60 + + + + + + + + + + + + + 80 + + + + + + + + + + + + + 100 + + + + + + + + + + + + + 120 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + gnuplot_plot_2 + + + + + + + + + gnuplot_plot_3 + + + + + + + gnuplot_plot_4 + + + + + + + gnuplot_plot_5 + + + + + + + + + gnuplot_plot_6 + + + + + + + gnuplot_plot_7 + + + + + + + gnuplot_plot_8 + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + gnuplot_plot_11 + + + + + + + gnuplot_plot_12 + + + + + + + + + gnuplot_plot_13 + + + + + + + gnuplot_plot_14 + + + + + + + + + gnuplot_plot_15 + + + + + + + gnuplot_plot_16 + + + + + + + + + gnuplot_plot_17 + + + + + + + gnuplot_plot_18 + + + + + + + gnuplot_plot_19 + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + distr_random_i128: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-4/distr-random-i16.svg b/results-uniform-int-5800X-4/distr-random-i16.svg new file mode 100644 index 00000000000..e817308cb15 --- /dev/null +++ b/results-uniform-int-5800X-4/distr-random-i16.svg @@ -0,0 +1,1737 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i16/Pcg64/Lemire + + + + + distr_random_i16/Pcg64/sample-unbiased + + + + + distr_random_i16/Pcg64/Canon32-2 + + + + + distr_random_i16/Pcg64/sample + + + + + distr_random_i16/Pcg64/Biased64 + + + + + distr_random_i16/Pcg32/Lemire + + + + + distr_random_i16/Pcg32/sample-unbiased + + + + + distr_random_i16/Pcg32/Canon32-2 + + + + + distr_random_i16/Pcg32/sample + + + + + distr_random_i16/Pcg32/Biased64 + + + + + distr_random_i16/ChaCha8Rng/Lemire + + + + + distr_random_i16/ChaCha8Rng/sample-unbiased + + + + + distr_random_i16/ChaCha8Rng/Canon32-2 + + + + + distr_random_i16/ChaCha8Rng/sample + + + + + distr_random_i16/ChaCha8Rng/Biased64 + + + + + distr_random_i16/SmallRng/Lemire + + + + + distr_random_i16/SmallRng/sample-unbiased + + + + + distr_random_i16/SmallRng/Canon32-2 + + + + + distr_random_i16/SmallRng/sample + + + + + distr_random_i16/SmallRng/Biased64 + + + + + + + + + + + + + 2 + + + + + + + + + + + + + 4 + + + + + + + + + + + + + 6 + + + + + + + + + + + + + 8 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 12 + + + + + + + + + + + + + 14 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + + + + + + + + + gnuplot_plot_2 + + + + + + + + + + + + + + + + + + + gnuplot_plot_3 + + + + + + + + + + + + + + + gnuplot_plot_4 + + + + + + + + + + + gnuplot_plot_5 + + + + + + + + + + + + + gnuplot_plot_6 + + + + + + + + + + + + + + + gnuplot_plot_7 + + + + + + + + + + + gnuplot_plot_8 + + + + + + + + + + + + + gnuplot_plot_9 + + + + + + + + + + + gnuplot_plot_10 + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_11 + + + + + + + + + gnuplot_plot_12 + + + + + + + + + gnuplot_plot_13 + + + + + + + + + + + + + + + gnuplot_plot_14 + + + + + + + + + + + + + gnuplot_plot_15 + + + + + + + + + gnuplot_plot_16 + + + + + + + + + + + + + + + gnuplot_plot_17 + + + + + + + + + + + + + + + + + gnuplot_plot_18 + + + + + + + + + + + + + + + + + gnuplot_plot_19 + + + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + + + + + distr_random_i16: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-4/distr-random-i32.svg b/results-uniform-int-5800X-4/distr-random-i32.svg new file mode 100644 index 00000000000..f696dba175d --- /dev/null +++ b/results-uniform-int-5800X-4/distr-random-i32.svg @@ -0,0 +1,1728 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i32/Pcg64/Lemire + + + + + distr_random_i32/Pcg64/sample-unbiased + + + + + distr_random_i32/Pcg64/sample + + + + + distr_random_i32/Pcg64/Canon32 + + + + + distr_random_i32/Pcg64/Biased64 + + + + + distr_random_i32/Pcg32/Lemire + + + + + distr_random_i32/Pcg32/sample-unbiased + + + + + distr_random_i32/Pcg32/sample + + + + + distr_random_i32/Pcg32/Canon32 + + + + + distr_random_i32/Pcg32/Biased64 + + + + + distr_random_i32/ChaCha8Rng/Lemire + + + + + distr_random_i32/ChaCha8Rng/sample-unbiased + + + + + distr_random_i32/ChaCha8Rng/sample + + + + + distr_random_i32/ChaCha8Rng/Canon32 + + + + + distr_random_i32/ChaCha8Rng/Biased64 + + + + + distr_random_i32/SmallRng/Lemire + + + + + distr_random_i32/SmallRng/sample-unbiased + + + + + distr_random_i32/SmallRng/sample + + + + + distr_random_i32/SmallRng/Canon32 + + + + + distr_random_i32/SmallRng/Biased64 + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 5 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 15 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 25 + + + + + + + + + + + + + 30 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + gnuplot_plot_2 + + + + + + + + + + + + + + + gnuplot_plot_3 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_4 + + + + + + + gnuplot_plot_5 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_6 + + + + + + + gnuplot_plot_7 + + + + + + + + + + + gnuplot_plot_8 + + + + + + + + + + + + + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_11 + + + + + + + gnuplot_plot_12 + + + + + + + + + + + + + gnuplot_plot_13 + + + + + + + + + + + + + + + gnuplot_plot_14 + + + + + + + gnuplot_plot_15 + + + + + + + + + + + + + + + + + + + gnuplot_plot_16 + + + + + + + gnuplot_plot_17 + + + + + + + + + + + + + gnuplot_plot_18 + + + + + + + + + + + + + gnuplot_plot_19 + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + distr_random_i32: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-4/distr-random-i64.svg b/results-uniform-int-5800X-4/distr-random-i64.svg new file mode 100644 index 00000000000..cc39416bff0 --- /dev/null +++ b/results-uniform-int-5800X-4/distr-random-i64.svg @@ -0,0 +1,1674 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i64/Pcg64/Lemire + + + + + distr_random_i64/Pcg64/sample-unbiased + + + + + distr_random_i64/Pcg64/Canon-Red-Un + + + + + distr_random_i64/Pcg64/Canon-Red + + + + + distr_random_i64/Pcg64/sample + + + + + distr_random_i64/Pcg32/Lemire + + + + + distr_random_i64/Pcg32/sample-unbiased + + + + + distr_random_i64/Pcg32/Canon-Red-Un + + + + + distr_random_i64/Pcg32/Canon-Red + + + + + distr_random_i64/Pcg32/sample + + + + + distr_random_i64/ChaCha8Rng/Lemire + + + + + distr_random_i64/ChaCha8Rng/sample-unbiased + + + + + distr_random_i64/ChaCha8Rng/Canon-Red-Un + + + + + distr_random_i64/ChaCha8Rng/Canon-Red + + + + + distr_random_i64/ChaCha8Rng/sample + + + + + distr_random_i64/SmallRng/Lemire + + + + + distr_random_i64/SmallRng/sample-unbiased + + + + + distr_random_i64/SmallRng/Canon-Red-Un + + + + + distr_random_i64/SmallRng/Canon-Red + + + + + distr_random_i64/SmallRng/sample + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 30 + + + + + + + + + + + + + 40 + + + + + + + + + + + + + 50 + + + + + + + + + + + + + 60 + + + + + + + + + + + + + 70 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + gnuplot_plot_2 + + + + + + + gnuplot_plot_3 + + + + + + + gnuplot_plot_4 + + + + + + + + + gnuplot_plot_5 + + + + + + + gnuplot_plot_6 + + + + + + + + + gnuplot_plot_7 + + + + + + + gnuplot_plot_8 + + + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + gnuplot_plot_11 + + + + + + + gnuplot_plot_12 + + + + + + + gnuplot_plot_13 + + + + + + + gnuplot_plot_14 + + + + + + + gnuplot_plot_15 + + + + + + + gnuplot_plot_16 + + + + + + + + + gnuplot_plot_17 + + + + + + + gnuplot_plot_18 + + + + + + + gnuplot_plot_19 + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + + + + + distr_random_i64: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-4/distr-random-i8.svg b/results-uniform-int-5800X-4/distr-random-i8.svg new file mode 100644 index 00000000000..7901323d11f --- /dev/null +++ b/results-uniform-int-5800X-4/distr-random-i8.svg @@ -0,0 +1,1796 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i8/Pcg64/Lemire + + + + + distr_random_i8/Pcg64/sample-unbiased + + + + + distr_random_i8/Pcg64/Canon32-2 + + + + + distr_random_i8/Pcg64/sample + + + + + distr_random_i8/Pcg64/Biased64 + + + + + distr_random_i8/Pcg32/Lemire + + + + + distr_random_i8/Pcg32/sample-unbiased + + + + + distr_random_i8/Pcg32/Canon32-2 + + + + + distr_random_i8/Pcg32/sample + + + + + distr_random_i8/Pcg32/Biased64 + + + + + distr_random_i8/ChaCha8Rng/Lemire + + + + + distr_random_i8/ChaCha8Rng/sample-unbiased + + + + + distr_random_i8/ChaCha8Rng/Canon32-2 + + + + + distr_random_i8/ChaCha8Rng/sample + + + + + distr_random_i8/ChaCha8Rng/Biased64 + + + + + distr_random_i8/SmallRng/Lemire + + + + + distr_random_i8/SmallRng/sample-unbiased + + + + + distr_random_i8/SmallRng/Canon32-2 + + + + + distr_random_i8/SmallRng/sample + + + + + distr_random_i8/SmallRng/Biased64 + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 2 + + + + + + + + + + + + + 4 + + + + + + + + + + + + + 6 + + + + + + + + + + + + + 8 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 12 + + + + + + + + + + + + + 14 + + + + + + + + + + + + + 16 + + + + + + + + + + + + + 18 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + + + + + + + + + + + gnuplot_plot_2 + + + + + + + + + + + + + gnuplot_plot_3 + + + + + + + + + + + gnuplot_plot_4 + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_5 + + + + + + + + + + + + + + + + + gnuplot_plot_6 + + + + + + + gnuplot_plot_7 + + + + + + + + + + + gnuplot_plot_8 + + + + + + + + + + + + + gnuplot_plot_9 + + + + + + + + + + + + + + + + + gnuplot_plot_10 + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_11 + + + + + + + + + + + + + + + + + + + gnuplot_plot_12 + + + + + + + + + + + + + + + gnuplot_plot_13 + + + + + + + + + gnuplot_plot_14 + + + + + + + + + gnuplot_plot_15 + + + + + + + + + gnuplot_plot_16 + + + + + + + + + + + gnuplot_plot_17 + + + + + + + + + + + + + + + + + gnuplot_plot_18 + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_19 + + + + + + + + + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i8: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-4/single-random-i128.svg b/results-uniform-int-5800X-4/single-random-i128.svg new file mode 100644 index 00000000000..700f7c27304 --- /dev/null +++ b/results-uniform-int-5800X-4/single-random-i128.svg @@ -0,0 +1,1670 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + single_random_i128/Pcg64/ONeill + + + + + single_random_i128/Pcg64/Canon-Red-Un + + + + + single_random_i128/Pcg64/Canon-Red + + + + + single_random_i128/Pcg64/sample-unbiased + + + + + single_random_i128/Pcg64/sample + + + + + single_random_i128/Pcg32/ONeill + + + + + single_random_i128/Pcg32/Canon-Red-Un + + + + + single_random_i128/Pcg32/Canon-Red + + + + + single_random_i128/Pcg32/sample-unbiased + + + + + single_random_i128/Pcg32/sample + + + + + single_random_i128/ChaCha8Rng/ONeill + + + + + single_random_i128/ChaCha8Rng/Canon-Red-Un + + + + + single_random_i128/ChaCha8Rng/Canon-Red + + + + + single_random_i128/ChaCha8Rng/sample-unbiased + + + + + single_random_i128/ChaCha8Rng/sample + + + + + single_random_i128/SmallRng/ONeill + + + + + single_random_i128/SmallRng/Canon-Red-Un + + + + + single_random_i128/SmallRng/Canon-Red + + + + + single_random_i128/SmallRng/sample-unbiased + + + + + single_random_i128/SmallRng/sample + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 30 + + + + + + + + + + + + + 40 + + + + + + + + + + + + + 50 + + + + + + + + + + + + + 60 + + + + + + + + + + + + + 70 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + + + gnuplot_plot_2 + + + + + + + gnuplot_plot_3 + + + + + + + gnuplot_plot_4 + + + + + + + gnuplot_plot_5 + + + + + + + gnuplot_plot_6 + + + + + + + gnuplot_plot_7 + + + + + + + gnuplot_plot_8 + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + gnuplot_plot_11 + + + + + + + + + gnuplot_plot_12 + + + + + + + gnuplot_plot_13 + + + + + + + gnuplot_plot_14 + + + + + + + gnuplot_plot_15 + + + + + + + gnuplot_plot_16 + + + + + + + gnuplot_plot_17 + + + + + + + gnuplot_plot_18 + + + + + + + + + gnuplot_plot_19 + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + single_random_i128: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-4/single-random-i16.svg b/results-uniform-int-5800X-4/single-random-i16.svg new file mode 100644 index 00000000000..8c19942b9d3 --- /dev/null +++ b/results-uniform-int-5800X-4/single-random-i16.svg @@ -0,0 +1,1744 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + single_random_i16/Pcg64/ONeill + + + + + single_random_i16/Pcg64/sample-unbiased + + + + + single_random_i16/Pcg64/Canon32-2 + + + + + single_random_i16/Pcg64/sample + + + + + single_random_i16/Pcg64/Biased64 + + + + + single_random_i16/Pcg32/ONeill + + + + + single_random_i16/Pcg32/sample-unbiased + + + + + single_random_i16/Pcg32/Canon32-2 + + + + + single_random_i16/Pcg32/sample + + + + + single_random_i16/Pcg32/Biased64 + + + + + single_random_i16/ChaCha8Rng/ONeill + + + + + single_random_i16/ChaCha8Rng/sample-unbiased + + + + + single_random_i16/ChaCha8Rng/Canon32-2 + + + + + single_random_i16/ChaCha8Rng/sample + + + + + single_random_i16/ChaCha8Rng/Biased64 + + + + + single_random_i16/SmallRng/ONeill + + + + + single_random_i16/SmallRng/sample-unbiased + + + + + single_random_i16/SmallRng/Canon32-2 + + + + + single_random_i16/SmallRng/sample + + + + + single_random_i16/SmallRng/Biased64 + + + + + + + + + + + + + 2 + + + + + + + + + + + + + 3 + + + + + + + + + + + + + 4 + + + + + + + + + + + + + 5 + + + + + + + + + + + + + 6 + + + + + + + + + + + + + 7 + + + + + + + + + + + + + 8 + + + + + + + + + + + + + 9 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 11 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + gnuplot_plot_2 + + + + + + + gnuplot_plot_3 + + + + + + + + + + + + + + + gnuplot_plot_4 + + + + + + + + + + + gnuplot_plot_5 + + + + + + + + + gnuplot_plot_6 + + + + + + + + + + + gnuplot_plot_7 + + + + + + + + + gnuplot_plot_8 + + + + + + + + + gnuplot_plot_9 + + + + + + + + + + + + + gnuplot_plot_10 + + + + + + + + + gnuplot_plot_11 + + + + + + + + + + + + + + + gnuplot_plot_12 + + + + + + + + + + + + + + + + + gnuplot_plot_13 + + + + + + + + + gnuplot_plot_14 + + + + + + + + + + + + + gnuplot_plot_15 + + + + + + + + + + + + + gnuplot_plot_16 + + + + + + + + + + + gnuplot_plot_17 + + + + + + + + + gnuplot_plot_18 + + + + + + + + + + + gnuplot_plot_19 + + + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + single_random_i16: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-4/single-random-i32.svg b/results-uniform-int-5800X-4/single-random-i32.svg new file mode 100644 index 00000000000..dd0b1a21b8b --- /dev/null +++ b/results-uniform-int-5800X-4/single-random-i32.svg @@ -0,0 +1,1737 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + single_random_i32/Pcg64/ONeill + + + + + single_random_i32/Pcg64/sample-unbiased + + + + + single_random_i32/Pcg64/sample + + + + + single_random_i32/Pcg64/Canon32 + + + + + single_random_i32/Pcg64/Biased64 + + + + + single_random_i32/Pcg32/ONeill + + + + + single_random_i32/Pcg32/sample-unbiased + + + + + single_random_i32/Pcg32/sample + + + + + single_random_i32/Pcg32/Canon32 + + + + + single_random_i32/Pcg32/Biased64 + + + + + single_random_i32/ChaCha8Rng/ONeill + + + + + single_random_i32/ChaCha8Rng/sample-unbiased + + + + + single_random_i32/ChaCha8Rng/sample + + + + + single_random_i32/ChaCha8Rng/Canon32 + + + + + single_random_i32/ChaCha8Rng/Biased64 + + + + + single_random_i32/SmallRng/ONeill + + + + + single_random_i32/SmallRng/sample-unbiased + + + + + single_random_i32/SmallRng/sample + + + + + single_random_i32/SmallRng/Canon32 + + + + + single_random_i32/SmallRng/Biased64 + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 5 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 15 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 25 + + + + + + + + + + + + + 30 + + + + + + + + + + + + + 35 + + + + + + + + + + + + + 40 + + + + + + + + + + + + + 45 + + + + + + + + + + + + + 50 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + gnuplot_plot_2 + + + + + + + + + + + + + gnuplot_plot_3 + + + + + + + gnuplot_plot_4 + + + + + + + gnuplot_plot_5 + + + + + + + + + + + gnuplot_plot_6 + + + + + + + gnuplot_plot_7 + + + + + + + + + + + gnuplot_plot_8 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + gnuplot_plot_11 + + + + + + + gnuplot_plot_12 + + + + + + + + + gnuplot_plot_13 + + + + + + + gnuplot_plot_14 + + + + + + + gnuplot_plot_15 + + + + + + + + + + + + + + + gnuplot_plot_16 + + + + + + + gnuplot_plot_17 + + + + + + + + + gnuplot_plot_18 + + + + + + + gnuplot_plot_19 + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + + + single_random_i32: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-4/single-random-i64.svg b/results-uniform-int-5800X-4/single-random-i64.svg new file mode 100644 index 00000000000..36723cd65c3 --- /dev/null +++ b/results-uniform-int-5800X-4/single-random-i64.svg @@ -0,0 +1,1682 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + single_random_i64/Pcg64/ONeill + + + + + single_random_i64/Pcg64/Canon-Red-Un + + + + + single_random_i64/Pcg64/Canon-Red + + + + + single_random_i64/Pcg64/sample-unbiased + + + + + single_random_i64/Pcg64/sample + + + + + single_random_i64/Pcg32/ONeill + + + + + single_random_i64/Pcg32/Canon-Red-Un + + + + + single_random_i64/Pcg32/Canon-Red + + + + + single_random_i64/Pcg32/sample-unbiased + + + + + single_random_i64/Pcg32/sample + + + + + single_random_i64/ChaCha8Rng/ONeill + + + + + single_random_i64/ChaCha8Rng/Canon-Red-Un + + + + + single_random_i64/ChaCha8Rng/Canon-Red + + + + + single_random_i64/ChaCha8Rng/sample-unbiased + + + + + single_random_i64/ChaCha8Rng/sample + + + + + single_random_i64/SmallRng/ONeill + + + + + single_random_i64/SmallRng/Canon-Red-Un + + + + + single_random_i64/SmallRng/Canon-Red + + + + + single_random_i64/SmallRng/sample-unbiased + + + + + single_random_i64/SmallRng/sample + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 30 + + + + + + + + + + + + + 40 + + + + + + + + + + + + + 50 + + + + + + + + + + + + + 60 + + + + + + + + + + + + + 70 + + + + + + + + + + + + + 80 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + gnuplot_plot_2 + + + + + + + gnuplot_plot_3 + + + + + + + gnuplot_plot_4 + + + + + + + gnuplot_plot_5 + + + + + + + gnuplot_plot_6 + + + + + + + gnuplot_plot_7 + + + + + + + gnuplot_plot_8 + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + + + gnuplot_plot_11 + + + + + + + gnuplot_plot_12 + + + + + + + gnuplot_plot_13 + + + + + + + gnuplot_plot_14 + + + + + + + + + gnuplot_plot_15 + + + + + + + gnuplot_plot_16 + + + + + + + gnuplot_plot_17 + + + + + + + gnuplot_plot_18 + + + + + + + gnuplot_plot_19 + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + single_random_i64: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-4/single-random-i8.svg b/results-uniform-int-5800X-4/single-random-i8.svg new file mode 100644 index 00000000000..09594e7e025 --- /dev/null +++ b/results-uniform-int-5800X-4/single-random-i8.svg @@ -0,0 +1,1741 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + single_random_i8/Pcg64/ONeill + + + + + single_random_i8/Pcg64/sample-unbiased + + + + + single_random_i8/Pcg64/Canon32-2 + + + + + single_random_i8/Pcg64/sample + + + + + single_random_i8/Pcg64/Biased64 + + + + + single_random_i8/Pcg32/ONeill + + + + + single_random_i8/Pcg32/sample-unbiased + + + + + single_random_i8/Pcg32/Canon32-2 + + + + + single_random_i8/Pcg32/sample + + + + + single_random_i8/Pcg32/Biased64 + + + + + single_random_i8/ChaCha8Rng/ONeill + + + + + single_random_i8/ChaCha8Rng/sample-unbiased + + + + + single_random_i8/ChaCha8Rng/Canon32-2 + + + + + single_random_i8/ChaCha8Rng/sample + + + + + single_random_i8/ChaCha8Rng/Biased64 + + + + + single_random_i8/SmallRng/ONeill + + + + + single_random_i8/SmallRng/sample-unbiased + + + + + single_random_i8/SmallRng/Canon32-2 + + + + + single_random_i8/SmallRng/sample + + + + + single_random_i8/SmallRng/Biased64 + + + + + + + + + + + + + 2 + + + + + + + + + + + + + 4 + + + + + + + + + + + + + 6 + + + + + + + + + + + + + 8 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 12 + + + + + + + + + + + + + 14 + + + + + + + + + + + + + 16 + + + + + + + + + + + + + 18 + + + + + + + + + + + + + 20 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + + + gnuplot_plot_2 + + + + + + + + + gnuplot_plot_3 + + + + + + + + + gnuplot_plot_4 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_5 + + + + + + + + + + + + + gnuplot_plot_6 + + + + + + + + + + + gnuplot_plot_7 + + + + + + + gnuplot_plot_8 + + + + + + + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + + + + + + + gnuplot_plot_11 + + + + + + + + + + + + + + + gnuplot_plot_12 + + + + + + + + + + + + + + + gnuplot_plot_13 + + + + + + + + + + + gnuplot_plot_14 + + + + + + + gnuplot_plot_15 + + + + + + + + + + + + + + + gnuplot_plot_16 + + + + + + + + + + + gnuplot_plot_17 + + + + + + + + + gnuplot_plot_18 + + + + + + + + + gnuplot_plot_19 + + + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + single_random_i8: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-5/distr-random-i128.svg b/results-uniform-int-5800X-5/distr-random-i128.svg new file mode 100644 index 00000000000..7b8eebeb91a --- /dev/null +++ b/results-uniform-int-5800X-5/distr-random-i128.svg @@ -0,0 +1,1703 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i128/Pcg64/Lemire + + + + + distr_random_i128/Pcg64/sample-unbiased + + + + + distr_random_i128/Pcg64/Canon-Red-Un + + + + + distr_random_i128/Pcg64/Canon-Red + + + + + distr_random_i128/Pcg64/sample + + + + + distr_random_i128/Pcg32/Lemire + + + + + distr_random_i128/Pcg32/sample-unbiased + + + + + distr_random_i128/Pcg32/Canon-Red-Un + + + + + distr_random_i128/Pcg32/Canon-Red + + + + + distr_random_i128/Pcg32/sample + + + + + distr_random_i128/ChaCha8Rng/Lemire + + + + + distr_random_i128/ChaCha8Rng/sample-unbiased + + + + + distr_random_i128/ChaCha8Rng/Canon-Red-Un + + + + + distr_random_i128/ChaCha8Rng/Canon-Red + + + + + distr_random_i128/ChaCha8Rng/sample + + + + + distr_random_i128/SmallRng/Lemire + + + + + distr_random_i128/SmallRng/sample-unbiased + + + + + distr_random_i128/SmallRng/Canon-Red-Un + + + + + distr_random_i128/SmallRng/Canon-Red + + + + + distr_random_i128/SmallRng/sample + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 30 + + + + + + + + + + + + + 40 + + + + + + + + + + + + + 50 + + + + + + + + + + + + + 60 + + + + + + + + + + + + + 70 + + + + + + + + + + + + + 80 + + + + + + + + + + + + + 90 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + gnuplot_plot_2 + + + + + + + gnuplot_plot_3 + + + + + + + gnuplot_plot_4 + + + + + + + gnuplot_plot_5 + + + + + + + gnuplot_plot_6 + + + + + + + gnuplot_plot_7 + + + + + + + + + + + gnuplot_plot_8 + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + gnuplot_plot_11 + + + + + + + + + gnuplot_plot_12 + + + + + + + + + gnuplot_plot_13 + + + + + + + gnuplot_plot_14 + + + + + + + gnuplot_plot_15 + + + + + + + + + gnuplot_plot_16 + + + + + + + gnuplot_plot_17 + + + + + + + gnuplot_plot_18 + + + + + + + gnuplot_plot_19 + + + + + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + + + distr_random_i128: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-5/distr-random-i16.svg b/results-uniform-int-5800X-5/distr-random-i16.svg new file mode 100644 index 00000000000..d5dc1bae601 --- /dev/null +++ b/results-uniform-int-5800X-5/distr-random-i16.svg @@ -0,0 +1,1840 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i16/Pcg64/Lemire + + + + + distr_random_i16/Pcg64/sample-unbiased + + + + + distr_random_i16/Pcg64/Canon32-2 + + + + + distr_random_i16/Pcg64/sample + + + + + distr_random_i16/Pcg64/Biased64 + + + + + distr_random_i16/Pcg32/Lemire + + + + + distr_random_i16/Pcg32/sample-unbiased + + + + + distr_random_i16/Pcg32/Canon32-2 + + + + + distr_random_i16/Pcg32/sample + + + + + distr_random_i16/Pcg32/Biased64 + + + + + distr_random_i16/ChaCha8Rng/Lemire + + + + + distr_random_i16/ChaCha8Rng/sample-unbiased + + + + + distr_random_i16/ChaCha8Rng/Canon32-2 + + + + + distr_random_i16/ChaCha8Rng/sample + + + + + distr_random_i16/ChaCha8Rng/Biased64 + + + + + distr_random_i16/SmallRng/Lemire + + + + + distr_random_i16/SmallRng/sample-unbiased + + + + + distr_random_i16/SmallRng/Canon32-2 + + + + + distr_random_i16/SmallRng/sample + + + + + distr_random_i16/SmallRng/Biased64 + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 5 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 15 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 25 + + + + + + + + + + + + + 30 + + + + + + + + + + + + + 35 + + + + + + + + + + + + + 40 + + + + + + + + + + + + + 45 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + gnuplot_plot_2 + + + + + + + + + + + gnuplot_plot_3 + + + + + + + + + + + + + gnuplot_plot_4 + + + + + + + + + + + gnuplot_plot_5 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_6 + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_7 + + + + + + + + + + + + + + + + + + + gnuplot_plot_8 + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_9 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_10 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_11 + + + + + + + + + gnuplot_plot_12 + + + + + + + + + gnuplot_plot_13 + + + + + + + gnuplot_plot_14 + + + + + + + + + + + gnuplot_plot_15 + + + + + + + gnuplot_plot_16 + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_17 + + + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_18 + + + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_19 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i16: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-5/distr-random-i32-2.svg b/results-uniform-int-5800X-5/distr-random-i32-2.svg new file mode 100644 index 00000000000..1bb65a83d10 --- /dev/null +++ b/results-uniform-int-5800X-5/distr-random-i32-2.svg @@ -0,0 +1,2096 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i32/Pcg64/Lemire64 + + + + + distr_random_i32/Pcg64/Lemire + + + + + distr_random_i32/Pcg64/sample-unbiased + + + + + distr_random_i32/Pcg64/sample + + + + + distr_random_i32/Pcg64/Canon32 + + + + + distr_random_i32/Pcg64/Biased64 + + + + + distr_random_i32/Pcg32/Lemire64 + + + + + distr_random_i32/Pcg32/Lemire + + + + + distr_random_i32/Pcg32/sample-unbiased + + + + + distr_random_i32/Pcg32/sample + + + + + distr_random_i32/Pcg32/Canon32 + + + + + distr_random_i32/Pcg32/Biased64 + + + + + distr_random_i32/ChaCha8Rng/Lemire64 + + + + + distr_random_i32/ChaCha8Rng/Lemire + + + + + distr_random_i32/ChaCha8Rng/sample-unbiased + + + + + distr_random_i32/ChaCha8Rng/sample + + + + + distr_random_i32/ChaCha8Rng/Canon32 + + + + + distr_random_i32/ChaCha8Rng/Biased64 + + + + + distr_random_i32/SmallRng/Lemire64 + + + + + distr_random_i32/SmallRng/Lemire + + + + + distr_random_i32/SmallRng/sample-unbiased + + + + + distr_random_i32/SmallRng/sample + + + + + distr_random_i32/SmallRng/Canon32 + + + + + distr_random_i32/SmallRng/Biased64 + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 5 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 15 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 25 + + + + + + + + + + + + + 30 + + + + + + + + + + + + + 35 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + + + + + + + + + + + gnuplot_plot_2 + + + + + + + gnuplot_plot_3 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_4 + + + + + + + + + + + + + + + + + + + gnuplot_plot_5 + + + + + + + gnuplot_plot_6 + + + + + + + + + + + + + gnuplot_plot_7 + + + + + + + + + + + + + + + + + gnuplot_plot_8 + + + + + + + + + gnuplot_plot_9 + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_10 + + + + + + + + + + + + + + + gnuplot_plot_11 + + + + + + + gnuplot_plot_12 + + + + + + + + + + + + + + + gnuplot_plot_13 + + + + + + + + + + + gnuplot_plot_14 + + + + + + + gnuplot_plot_15 + + + + + + + + + + + + + + + gnuplot_plot_16 + + + + + + + + + + + + + gnuplot_plot_17 + + + + + + + gnuplot_plot_18 + + + + + + + + + + + gnuplot_plot_19 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_20 + + + + + + + gnuplot_plot_21 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_22 + + + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_23 + + + + + + + gnuplot_plot_24 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i32: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-5/distr-random-i32.svg b/results-uniform-int-5800X-5/distr-random-i32.svg new file mode 100644 index 00000000000..03b8f210f86 --- /dev/null +++ b/results-uniform-int-5800X-5/distr-random-i32.svg @@ -0,0 +1,1788 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i32/Pcg64/Lemire + + + + + distr_random_i32/Pcg64/sample-unbiased + + + + + distr_random_i32/Pcg64/sample + + + + + distr_random_i32/Pcg64/Canon32 + + + + + distr_random_i32/Pcg64/Biased64 + + + + + distr_random_i32/Pcg32/Lemire + + + + + distr_random_i32/Pcg32/sample-unbiased + + + + + distr_random_i32/Pcg32/sample + + + + + distr_random_i32/Pcg32/Canon32 + + + + + distr_random_i32/Pcg32/Biased64 + + + + + distr_random_i32/ChaCha8Rng/Lemire + + + + + distr_random_i32/ChaCha8Rng/sample-unbiased + + + + + distr_random_i32/ChaCha8Rng/sample + + + + + distr_random_i32/ChaCha8Rng/Canon32 + + + + + distr_random_i32/ChaCha8Rng/Biased64 + + + + + distr_random_i32/SmallRng/Lemire + + + + + distr_random_i32/SmallRng/sample-unbiased + + + + + distr_random_i32/SmallRng/sample + + + + + distr_random_i32/SmallRng/Canon32 + + + + + distr_random_i32/SmallRng/Biased64 + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 5 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 15 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 25 + + + + + + + + + + + + + 30 + + + + + + + + + + + + + 35 + + + + + + + + + + + + + 40 + + + + + + + + + + + + + 45 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + gnuplot_plot_2 + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_3 + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_4 + + + + + + + gnuplot_plot_5 + + + + + + + + + + + + + + + + + gnuplot_plot_6 + + + + + + + gnuplot_plot_7 + + + + + + + + + + + + + gnuplot_plot_8 + + + + + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + + + + + + + + + + + + + gnuplot_plot_11 + + + + + + + gnuplot_plot_12 + + + + + + + + + gnuplot_plot_13 + + + + + + + + + + + + + + + + + gnuplot_plot_14 + + + + + + + gnuplot_plot_15 + + + + + + + + + + + + + + + gnuplot_plot_16 + + + + + + + gnuplot_plot_17 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_18 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_19 + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i32: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-5/distr-random-i64.svg b/results-uniform-int-5800X-5/distr-random-i64.svg new file mode 100644 index 00000000000..5f0770680fc --- /dev/null +++ b/results-uniform-int-5800X-5/distr-random-i64.svg @@ -0,0 +1,1669 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i64/Pcg64/Lemire + + + + + distr_random_i64/Pcg64/sample-unbiased + + + + + distr_random_i64/Pcg64/Canon-Red-Un + + + + + distr_random_i64/Pcg64/Canon-Red + + + + + distr_random_i64/Pcg64/sample + + + + + distr_random_i64/Pcg32/Lemire + + + + + distr_random_i64/Pcg32/sample-unbiased + + + + + distr_random_i64/Pcg32/Canon-Red-Un + + + + + distr_random_i64/Pcg32/Canon-Red + + + + + distr_random_i64/Pcg32/sample + + + + + distr_random_i64/ChaCha8Rng/Lemire + + + + + distr_random_i64/ChaCha8Rng/sample-unbiased + + + + + distr_random_i64/ChaCha8Rng/Canon-Red-Un + + + + + distr_random_i64/ChaCha8Rng/Canon-Red + + + + + distr_random_i64/ChaCha8Rng/sample + + + + + distr_random_i64/SmallRng/Lemire + + + + + distr_random_i64/SmallRng/sample-unbiased + + + + + distr_random_i64/SmallRng/Canon-Red-Un + + + + + distr_random_i64/SmallRng/Canon-Red + + + + + distr_random_i64/SmallRng/sample + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 30 + + + + + + + + + + + + + 40 + + + + + + + + + + + + + 50 + + + + + + + + + + + + + 60 + + + + + + + + + + + + + 70 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + gnuplot_plot_2 + + + + + + + gnuplot_plot_3 + + + + + + + gnuplot_plot_4 + + + + + + + + + gnuplot_plot_5 + + + + + + + gnuplot_plot_6 + + + + + + + gnuplot_plot_7 + + + + + + + gnuplot_plot_8 + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + gnuplot_plot_11 + + + + + + + gnuplot_plot_12 + + + + + + + gnuplot_plot_13 + + + + + + + gnuplot_plot_14 + + + + + + + gnuplot_plot_15 + + + + + + + gnuplot_plot_16 + + + + + + + gnuplot_plot_17 + + + + + + + gnuplot_plot_18 + + + + + + + gnuplot_plot_19 + + + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + distr_random_i64: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-5/distr-random-i8.svg b/results-uniform-int-5800X-5/distr-random-i8.svg new file mode 100644 index 00000000000..275addd25a8 --- /dev/null +++ b/results-uniform-int-5800X-5/distr-random-i8.svg @@ -0,0 +1,1849 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i8/Pcg64/Lemire + + + + + distr_random_i8/Pcg64/sample-unbiased + + + + + distr_random_i8/Pcg64/Canon32-2 + + + + + distr_random_i8/Pcg64/sample + + + + + distr_random_i8/Pcg64/Biased64 + + + + + distr_random_i8/Pcg32/Lemire + + + + + distr_random_i8/Pcg32/sample-unbiased + + + + + distr_random_i8/Pcg32/Canon32-2 + + + + + distr_random_i8/Pcg32/sample + + + + + distr_random_i8/Pcg32/Biased64 + + + + + distr_random_i8/ChaCha8Rng/Lemire + + + + + distr_random_i8/ChaCha8Rng/sample-unbiased + + + + + distr_random_i8/ChaCha8Rng/Canon32-2 + + + + + distr_random_i8/ChaCha8Rng/sample + + + + + distr_random_i8/ChaCha8Rng/Biased64 + + + + + distr_random_i8/SmallRng/Lemire + + + + + distr_random_i8/SmallRng/sample-unbiased + + + + + distr_random_i8/SmallRng/Canon32-2 + + + + + distr_random_i8/SmallRng/sample + + + + + distr_random_i8/SmallRng/Biased64 + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 5 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 15 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 25 + + + + + + + + + + + + + 30 + + + + + + + + + + + + + 35 + + + + + + + + + + + + + 40 + + + + + + + + + + + + + 45 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + + + + + + + gnuplot_plot_2 + + + + + + + + + gnuplot_plot_3 + + + + + + + + + gnuplot_plot_4 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_5 + + + + + + + + + + + + + gnuplot_plot_6 + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_7 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_8 + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_9 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_10 + + + + + + + + + + + + + gnuplot_plot_11 + + + + + + + + + gnuplot_plot_12 + + + + + + + + + + + gnuplot_plot_13 + + + + + + + + + + + + + + + + + + + gnuplot_plot_14 + + + + + + + + + + + gnuplot_plot_15 + + + + + + + + + + + gnuplot_plot_16 + + + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_17 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_18 + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_19 + + + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + distr_random_i8: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-5/single-random-i128.svg b/results-uniform-int-5800X-5/single-random-i128.svg new file mode 100644 index 00000000000..700f7c27304 --- /dev/null +++ b/results-uniform-int-5800X-5/single-random-i128.svg @@ -0,0 +1,1670 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + single_random_i128/Pcg64/ONeill + + + + + single_random_i128/Pcg64/Canon-Red-Un + + + + + single_random_i128/Pcg64/Canon-Red + + + + + single_random_i128/Pcg64/sample-unbiased + + + + + single_random_i128/Pcg64/sample + + + + + single_random_i128/Pcg32/ONeill + + + + + single_random_i128/Pcg32/Canon-Red-Un + + + + + single_random_i128/Pcg32/Canon-Red + + + + + single_random_i128/Pcg32/sample-unbiased + + + + + single_random_i128/Pcg32/sample + + + + + single_random_i128/ChaCha8Rng/ONeill + + + + + single_random_i128/ChaCha8Rng/Canon-Red-Un + + + + + single_random_i128/ChaCha8Rng/Canon-Red + + + + + single_random_i128/ChaCha8Rng/sample-unbiased + + + + + single_random_i128/ChaCha8Rng/sample + + + + + single_random_i128/SmallRng/ONeill + + + + + single_random_i128/SmallRng/Canon-Red-Un + + + + + single_random_i128/SmallRng/Canon-Red + + + + + single_random_i128/SmallRng/sample-unbiased + + + + + single_random_i128/SmallRng/sample + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 20 + + + + + + + + + + + + + 30 + + + + + + + + + + + + + 40 + + + + + + + + + + + + + 50 + + + + + + + + + + + + + 60 + + + + + + + + + + + + + 70 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + + + gnuplot_plot_2 + + + + + + + gnuplot_plot_3 + + + + + + + gnuplot_plot_4 + + + + + + + gnuplot_plot_5 + + + + + + + gnuplot_plot_6 + + + + + + + gnuplot_plot_7 + + + + + + + gnuplot_plot_8 + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + gnuplot_plot_11 + + + + + + + + + gnuplot_plot_12 + + + + + + + gnuplot_plot_13 + + + + + + + gnuplot_plot_14 + + + + + + + gnuplot_plot_15 + + + + + + + gnuplot_plot_16 + + + + + + + gnuplot_plot_17 + + + + + + + gnuplot_plot_18 + + + + + + + + + gnuplot_plot_19 + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + single_random_i128: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-5/single-random-i16.svg b/results-uniform-int-5800X-5/single-random-i16.svg new file mode 100644 index 00000000000..baf54a26a81 --- /dev/null +++ b/results-uniform-int-5800X-5/single-random-i16.svg @@ -0,0 +1,1723 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + single_random_i16/Pcg64/ONeill + + + + + single_random_i16/Pcg64/sample-unbiased + + + + + single_random_i16/Pcg64/Canon32-2 + + + + + single_random_i16/Pcg64/sample + + + + + single_random_i16/Pcg64/Biased64 + + + + + single_random_i16/Pcg32/ONeill + + + + + single_random_i16/Pcg32/sample-unbiased + + + + + single_random_i16/Pcg32/Canon32-2 + + + + + single_random_i16/Pcg32/sample + + + + + single_random_i16/Pcg32/Biased64 + + + + + single_random_i16/ChaCha8Rng/ONeill + + + + + single_random_i16/ChaCha8Rng/sample-unbiased + + + + + single_random_i16/ChaCha8Rng/Canon32-2 + + + + + single_random_i16/ChaCha8Rng/sample + + + + + single_random_i16/ChaCha8Rng/Biased64 + + + + + single_random_i16/SmallRng/ONeill + + + + + single_random_i16/SmallRng/sample-unbiased + + + + + single_random_i16/SmallRng/Canon32-2 + + + + + single_random_i16/SmallRng/sample + + + + + single_random_i16/SmallRng/Biased64 + + + + + + + + + + + + + 1 + + + + + + + + + + + + + 2 + + + + + + + + + + + + + 3 + + + + + + + + + + + + + 4 + + + + + + + + + + + + + 5 + + + + + + + + + + + + + 6 + + + + + + + + + + + + + 7 + + + + + + + + + + + + + 8 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + + + + + gnuplot_plot_2 + + + + + + + gnuplot_plot_3 + + + + + + + + + + + + + + + + + gnuplot_plot_4 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_5 + + + + + + + gnuplot_plot_6 + + + + + + + + + gnuplot_plot_7 + + + + + + + + + + + + + + + gnuplot_plot_8 + + + + + + + + + + + + + + + gnuplot_plot_9 + + + + + + + + + gnuplot_plot_10 + + + + + + + + + gnuplot_plot_11 + + + + + + + + + + + + + + + gnuplot_plot_12 + + + + + + + + + + + gnuplot_plot_13 + + + + + + + + + + + + + gnuplot_plot_14 + + + + + + + + + gnuplot_plot_15 + + + + + + + + + + + + + gnuplot_plot_16 + + + + + + + + + gnuplot_plot_17 + + + + + + + gnuplot_plot_18 + + + + + + + + + + + gnuplot_plot_19 + + + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + single_random_i16: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-5/single-random-i32.svg b/results-uniform-int-5800X-5/single-random-i32.svg new file mode 100644 index 00000000000..3df8dda9091 --- /dev/null +++ b/results-uniform-int-5800X-5/single-random-i32.svg @@ -0,0 +1,1691 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + single_random_i32/Pcg64/ONeill + + + + + single_random_i32/Pcg64/sample-unbiased + + + + + single_random_i32/Pcg64/sample + + + + + single_random_i32/Pcg64/Canon32 + + + + + single_random_i32/Pcg64/Biased64 + + + + + single_random_i32/Pcg32/ONeill + + + + + single_random_i32/Pcg32/sample-unbiased + + + + + single_random_i32/Pcg32/sample + + + + + single_random_i32/Pcg32/Canon32 + + + + + single_random_i32/Pcg32/Biased64 + + + + + single_random_i32/ChaCha8Rng/ONeill + + + + + single_random_i32/ChaCha8Rng/sample-unbiased + + + + + single_random_i32/ChaCha8Rng/sample + + + + + single_random_i32/ChaCha8Rng/Canon32 + + + + + single_random_i32/ChaCha8Rng/Biased64 + + + + + single_random_i32/SmallRng/ONeill + + + + + single_random_i32/SmallRng/sample-unbiased + + + + + single_random_i32/SmallRng/sample + + + + + single_random_i32/SmallRng/Canon32 + + + + + single_random_i32/SmallRng/Biased64 + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 2 + + + + + + + + + + + + + 4 + + + + + + + + + + + + + 6 + + + + + + + + + + + + + 8 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 12 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + gnuplot_plot_2 + + + + + + + + + gnuplot_plot_3 + + + + + + + + + + + + + + + gnuplot_plot_4 + + + + + + + + + gnuplot_plot_5 + + + + + + + + + + + gnuplot_plot_6 + + + + + + + gnuplot_plot_7 + + + + + + + + + gnuplot_plot_8 + + + + + + + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + + + + + gnuplot_plot_11 + + + + + + + gnuplot_plot_12 + + + + + + + + + gnuplot_plot_13 + + + + + + + gnuplot_plot_14 + + + + + + + gnuplot_plot_15 + + + + + + + + + + + gnuplot_plot_16 + + + + + + + gnuplot_plot_17 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_18 + + + + + + + + + + + + + gnuplot_plot_19 + + + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + + + + + + + single_random_i32: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-5/single-random-i64.svg b/results-uniform-int-5800X-5/single-random-i64.svg new file mode 100644 index 00000000000..40ff49521f6 --- /dev/null +++ b/results-uniform-int-5800X-5/single-random-i64.svg @@ -0,0 +1,1666 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + single_random_i64/Pcg64/ONeill + + + + + single_random_i64/Pcg64/Canon-Red-Un + + + + + single_random_i64/Pcg64/Canon-Red + + + + + single_random_i64/Pcg64/sample-unbiased + + + + + single_random_i64/Pcg64/sample + + + + + single_random_i64/Pcg32/ONeill + + + + + single_random_i64/Pcg32/Canon-Red-Un + + + + + single_random_i64/Pcg32/Canon-Red + + + + + single_random_i64/Pcg32/sample-unbiased + + + + + single_random_i64/Pcg32/sample + + + + + single_random_i64/ChaCha8Rng/ONeill + + + + + single_random_i64/ChaCha8Rng/Canon-Red-Un + + + + + single_random_i64/ChaCha8Rng/Canon-Red + + + + + single_random_i64/ChaCha8Rng/sample-unbiased + + + + + single_random_i64/ChaCha8Rng/sample + + + + + single_random_i64/SmallRng/ONeill + + + + + single_random_i64/SmallRng/Canon-Red-Un + + + + + single_random_i64/SmallRng/Canon-Red + + + + + single_random_i64/SmallRng/sample-unbiased + + + + + single_random_i64/SmallRng/sample + + + + + + + + + + + + + 0 + + + + + + + + + + + + + 2 + + + + + + + + + + + + + 4 + + + + + + + + + + + + + 6 + + + + + + + + + + + + + 8 + + + + + + + + + + + + + 10 + + + + + + + + + + + + + 12 + + + + + + + + + + + + + 14 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + gnuplot_plot_2 + + + + + + + gnuplot_plot_3 + + + + + + + gnuplot_plot_4 + + + + + + + gnuplot_plot_5 + + + + + + + gnuplot_plot_6 + + + + + + + gnuplot_plot_7 + + + + + + + gnuplot_plot_8 + + + + + + + gnuplot_plot_9 + + + + + + + gnuplot_plot_10 + + + + + + + gnuplot_plot_11 + + + + + + + gnuplot_plot_12 + + + + + + + gnuplot_plot_13 + + + + + + + gnuplot_plot_14 + + + + + + + gnuplot_plot_15 + + + + + + + gnuplot_plot_16 + + + + + + + gnuplot_plot_17 + + + + + + + gnuplot_plot_18 + + + + + + + gnuplot_plot_19 + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + single_random_i64: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-5/single-random-i8.svg b/results-uniform-int-5800X-5/single-random-i8.svg new file mode 100644 index 00000000000..9c1233ff7f5 --- /dev/null +++ b/results-uniform-int-5800X-5/single-random-i8.svg @@ -0,0 +1,1763 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + single_random_i8/Pcg64/ONeill + + + + + single_random_i8/Pcg64/sample-unbiased + + + + + single_random_i8/Pcg64/Canon32-2 + + + + + single_random_i8/Pcg64/sample + + + + + single_random_i8/Pcg64/Biased64 + + + + + single_random_i8/Pcg32/ONeill + + + + + single_random_i8/Pcg32/sample-unbiased + + + + + single_random_i8/Pcg32/Canon32-2 + + + + + single_random_i8/Pcg32/sample + + + + + single_random_i8/Pcg32/Biased64 + + + + + single_random_i8/ChaCha8Rng/ONeill + + + + + single_random_i8/ChaCha8Rng/sample-unbiased + + + + + single_random_i8/ChaCha8Rng/Canon32-2 + + + + + single_random_i8/ChaCha8Rng/sample + + + + + single_random_i8/ChaCha8Rng/Biased64 + + + + + single_random_i8/SmallRng/ONeill + + + + + single_random_i8/SmallRng/sample-unbiased + + + + + single_random_i8/SmallRng/Canon32-2 + + + + + single_random_i8/SmallRng/sample + + + + + single_random_i8/SmallRng/Biased64 + + + + + + + + + + + + + 1 + + + + + + + + + + + + + 2 + + + + + + + + + + + + + 3 + + + + + + + + + + + + + 4 + + + + + + + + + + + + + 5 + + + + + + + + + + + + + 6 + + + + + + + + + + + + + 7 + + + + + + + + + + + + + 8 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_2 + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_3 + + + + + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_4 + + + + + + + gnuplot_plot_5 + + + + + + + + + + + + + gnuplot_plot_6 + + + + + + + + + gnuplot_plot_7 + + + + + + + + + + + + + gnuplot_plot_8 + + + + + + + + + gnuplot_plot_9 + + + + + + + + + gnuplot_plot_10 + + + + + + + + + + + gnuplot_plot_11 + + + + + + + + + + + + + + + gnuplot_plot_12 + + + + + + + + + + + + + + + + + gnuplot_plot_13 + + + + + + + + + + + gnuplot_plot_14 + + + + + + + + + + + + + gnuplot_plot_15 + + + + + + + + + + + gnuplot_plot_16 + + + + + + + + + + + gnuplot_plot_17 + + + + + + + + + gnuplot_plot_18 + + + + + + + gnuplot_plot_19 + + + + + + + + + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + single_random_i8: Violin plot + + + + + + + diff --git a/results-uniform-int-5800X-5/violin.svg b/results-uniform-int-5800X-5/violin.svg new file mode 100644 index 00000000000..baf54a26a81 --- /dev/null +++ b/results-uniform-int-5800X-5/violin.svg @@ -0,0 +1,1723 @@ + + + +Gnuplot +Produced by GNUPLOT 5.4 patchlevel 3 + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + single_random_i16/Pcg64/ONeill + + + + + single_random_i16/Pcg64/sample-unbiased + + + + + single_random_i16/Pcg64/Canon32-2 + + + + + single_random_i16/Pcg64/sample + + + + + single_random_i16/Pcg64/Biased64 + + + + + single_random_i16/Pcg32/ONeill + + + + + single_random_i16/Pcg32/sample-unbiased + + + + + single_random_i16/Pcg32/Canon32-2 + + + + + single_random_i16/Pcg32/sample + + + + + single_random_i16/Pcg32/Biased64 + + + + + single_random_i16/ChaCha8Rng/ONeill + + + + + single_random_i16/ChaCha8Rng/sample-unbiased + + + + + single_random_i16/ChaCha8Rng/Canon32-2 + + + + + single_random_i16/ChaCha8Rng/sample + + + + + single_random_i16/ChaCha8Rng/Biased64 + + + + + single_random_i16/SmallRng/ONeill + + + + + single_random_i16/SmallRng/sample-unbiased + + + + + single_random_i16/SmallRng/Canon32-2 + + + + + single_random_i16/SmallRng/sample + + + + + single_random_i16/SmallRng/Biased64 + + + + + + + + + + + + + 1 + + + + + + + + + + + + + 2 + + + + + + + + + + + + + 3 + + + + + + + + + + + + + 4 + + + + + + + + + + + + + 5 + + + + + + + + + + + + + 6 + + + + + + + + + + + + + 7 + + + + + + + + + + + + + 8 + + + + + + + + + Input + + + + + Average time (ns) + + + + + PDF + + + PDF + + + + + + + + + + + + + + gnuplot_plot_2 + + + + + + + gnuplot_plot_3 + + + + + + + + + + + + + + + + + gnuplot_plot_4 + + + + + + + + + + + + + + + + + + + + + gnuplot_plot_5 + + + + + + + gnuplot_plot_6 + + + + + + + + + gnuplot_plot_7 + + + + + + + + + + + + + + + gnuplot_plot_8 + + + + + + + + + + + + + + + gnuplot_plot_9 + + + + + + + + + gnuplot_plot_10 + + + + + + + + + gnuplot_plot_11 + + + + + + + + + + + + + + + gnuplot_plot_12 + + + + + + + + + + + gnuplot_plot_13 + + + + + + + + + + + + + gnuplot_plot_14 + + + + + + + + + gnuplot_plot_15 + + + + + + + + + + + + + gnuplot_plot_16 + + + + + + + + + gnuplot_plot_17 + + + + + + + gnuplot_plot_18 + + + + + + + + + + + gnuplot_plot_19 + + + + + + + + + gnuplot_plot_20 + + + + + + + + + + + + + + + + + single_random_i16: Violin plot + + + + + + + diff --git a/src/distributions/uniform.rs b/src/distributions/uniform.rs index 261357b2456..e161463d920 100644 --- a/src/distributions/uniform.rs +++ b/src/distributions/uniform.rs @@ -106,11 +106,8 @@ //! [`UniformDuration`]: crate::distributions::uniform::UniformDuration //! [`SampleBorrow::borrow`]: crate::distributions::uniform::SampleBorrow::borrow -use core::time::Duration; use core::ops::{Range, RangeInclusive}; -use crate::distributions::float::IntoFloat; -use crate::distributions::utils::{BoolAsSIMD, FloatAsSIMD, FloatSIMDUtils, WideningMultiply}; use crate::distributions::Distribution; use crate::{Rng, RngCore}; @@ -118,10 +115,16 @@ use crate::{Rng, RngCore}; #[allow(unused_imports)] // rustc doesn't detect that this is actually used use crate::distributions::utils::Float; -#[cfg(feature = "simd_support")] use packed_simd::*; +#[cfg(feature = "serde1")] use serde::{Deserialize, Serialize}; -#[cfg(feature = "serde1")] -use serde::{Serialize, Deserialize}; +mod uniform_float; +mod uniform_int; +mod uniform_other; +#[cfg(feature = "simd_support")] mod uniform_simd; + +pub use uniform_float::UniformFloat; +pub use uniform_int::UniformInt; +pub use uniform_other::{UniformChar, UniformDuration}; /// Sample values uniformly between two bounds. /// @@ -175,8 +178,12 @@ use serde::{Serialize, Deserialize}; #[derive(Clone, Copy, Debug, PartialEq)] #[cfg_attr(feature = "serde1", derive(Serialize, Deserialize))] #[cfg_attr(feature = "serde1", serde(bound(serialize = "X::Sampler: Serialize")))] -#[cfg_attr(feature = "serde1", serde(bound(deserialize = "X::Sampler: Deserialize<'de>")))] -pub struct Uniform(X::Sampler); +#[cfg_attr( + feature = "serde1", + serde(bound(deserialize = "X::Sampler: Deserialize<'de>")) +)] +// HACK: field is pub +pub struct Uniform(pub X::Sampler); impl Uniform { /// Create a new `Uniform` instance which samples uniformly from the half @@ -291,10 +298,10 @@ pub trait UniformSampler: Sized { /// some types more optimal implementations for single usage may be provided /// via this method. /// Results may not be identical. - fn sample_single_inclusive(low: B1, high: B2, rng: &mut R) - -> Self::X - where B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized + fn sample_single_inclusive(low: B1, high: B2, rng: &mut R) -> Self::X + where + B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized, { let uniform: Self = UniformSampler::new_inclusive(low, high); uniform.sample(rng) @@ -313,7 +320,6 @@ impl From> for Uniform { } } - /// Helper trait similar to [`Borrow`] but implemented /// only for SampleUniform and references to SampleUniform in /// order to resolve ambiguity issues. @@ -378,1151 +384,9 @@ impl SampleRange for RangeInclusive { } } - -//////////////////////////////////////////////////////////////////////////////// - -// What follows are all back-ends. - - -/// The back-end implementing [`UniformSampler`] for integer types. -/// -/// Unless you are implementing [`UniformSampler`] for your own type, this type -/// should not be used directly, use [`Uniform`] instead. -/// -/// # Implementation notes -/// -/// For simplicity, we use the same generic struct `UniformInt` for all -/// integer types `X`. This gives us only one field type, `X`; to store unsigned -/// values of this size, we take use the fact that these conversions are no-ops. -/// -/// For a closed range, the number of possible numbers we should generate is -/// `range = (high - low + 1)`. To avoid bias, we must ensure that the size of -/// our sample space, `zone`, is a multiple of `range`; other values must be -/// rejected (by replacing with a new random sample). -/// -/// As a special case, we use `range = 0` to represent the full range of the -/// result type (i.e. for `new_inclusive($ty::MIN, $ty::MAX)`). -/// -/// The optimum `zone` is the largest product of `range` which fits in our -/// (unsigned) target type. We calculate this by calculating how many numbers we -/// must reject: `reject = (MAX + 1) % range = (MAX - range + 1) % range`. Any (large) -/// product of `range` will suffice, thus in `sample_single` we multiply by a -/// power of 2 via bit-shifting (faster but may cause more rejections). -/// -/// The smallest integer PRNGs generate is `u32`. For 8- and 16-bit outputs we -/// use `u32` for our `zone` and samples (because it's not slower and because -/// it reduces the chance of having to reject a sample). In this case we cannot -/// store `zone` in the target type since it is too large, however we know -/// `ints_to_reject < range <= $unsigned::MAX`. -/// -/// An alternative to using a modulus is widening multiply: After a widening -/// multiply by `range`, the result is in the high word. Then comparing the low -/// word against `zone` makes sure our distribution is uniform. -#[derive(Clone, Copy, Debug, PartialEq)] -#[cfg_attr(feature = "serde1", derive(Serialize, Deserialize))] -pub struct UniformInt { - low: X, - range: X, - z: X, // either ints_to_reject or zone depending on implementation -} - -macro_rules! uniform_int_impl { - ($ty:ty, $unsigned:ident, $u_large:ident) => { - impl SampleUniform for $ty { - type Sampler = UniformInt<$ty>; - } - - impl UniformSampler for UniformInt<$ty> { - // We play free and fast with unsigned vs signed here - // (when $ty is signed), but that's fine, since the - // contract of this macro is for $ty and $unsigned to be - // "bit-equal", so casting between them is a no-op. - - type X = $ty; - - #[inline] // if the range is constant, this helps LLVM to do the - // calculations at compile-time. - fn new(low_b: B1, high_b: B2) -> Self - where - B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized, - { - let low = *low_b.borrow(); - let high = *high_b.borrow(); - assert!(low < high, "Uniform::new called with `low >= high`"); - UniformSampler::new_inclusive(low, high - 1) - } - - #[inline] // if the range is constant, this helps LLVM to do the - // calculations at compile-time. - fn new_inclusive(low_b: B1, high_b: B2) -> Self - where - B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized, - { - let low = *low_b.borrow(); - let high = *high_b.borrow(); - assert!( - low <= high, - "Uniform::new_inclusive called with `low > high`" - ); - let unsigned_max = ::core::$u_large::MAX; - - let range = high.wrapping_sub(low).wrapping_add(1) as $unsigned; - let ints_to_reject = if range > 0 { - let range = $u_large::from(range); - (unsigned_max - range + 1) % range - } else { - 0 - }; - - UniformInt { - low, - // These are really $unsigned values, but store as $ty: - range: range as $ty, - z: ints_to_reject as $unsigned as $ty, - } - } - - #[inline] - fn sample(&self, rng: &mut R) -> Self::X { - let range = self.range as $unsigned as $u_large; - if range > 0 { - let unsigned_max = ::core::$u_large::MAX; - let zone = unsigned_max - (self.z as $unsigned as $u_large); - loop { - let v: $u_large = rng.gen(); - let (hi, lo) = v.wmul(range); - if lo <= zone { - return self.low.wrapping_add(hi as $ty); - } - } - } else { - // Sample from the entire integer range. - rng.gen() - } - } - - #[inline] - fn sample_single(low_b: B1, high_b: B2, rng: &mut R) -> Self::X - where - B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized, - { - let low = *low_b.borrow(); - let high = *high_b.borrow(); - assert!(low < high, "UniformSampler::sample_single: low >= high"); - Self::sample_single_inclusive(low, high - 1, rng) - } - - #[inline] - fn sample_single_inclusive(low_b: B1, high_b: B2, rng: &mut R) -> Self::X - where - B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized, - { - let low = *low_b.borrow(); - let high = *high_b.borrow(); - assert!(low <= high, "UniformSampler::sample_single_inclusive: low > high"); - let range = high.wrapping_sub(low).wrapping_add(1) as $unsigned as $u_large; - // If the above resulted in wrap-around to 0, the range is $ty::MIN..=$ty::MAX, - // and any integer will do. - if range == 0 { - return rng.gen(); - } - - let zone = if ::core::$unsigned::MAX <= ::core::u16::MAX as $unsigned { - // Using a modulus is faster than the approximation for - // i8 and i16. I suppose we trade the cost of one - // modulus for near-perfect branch prediction. - let unsigned_max: $u_large = ::core::$u_large::MAX; - let ints_to_reject = (unsigned_max - range + 1) % range; - unsigned_max - ints_to_reject - } else { - // conservative but fast approximation. `- 1` is necessary to allow the - // same comparison without bias. - (range << range.leading_zeros()).wrapping_sub(1) - }; - - loop { - let v: $u_large = rng.gen(); - let (hi, lo) = v.wmul(range); - if lo <= zone { - return low.wrapping_add(hi as $ty); - } - } - } - } - }; -} - -uniform_int_impl! { i8, u8, u32 } -uniform_int_impl! { i16, u16, u32 } -uniform_int_impl! { i32, u32, u32 } -uniform_int_impl! { i64, u64, u64 } -uniform_int_impl! { i128, u128, u128 } -uniform_int_impl! { isize, usize, usize } -uniform_int_impl! { u8, u8, u32 } -uniform_int_impl! { u16, u16, u32 } -uniform_int_impl! { u32, u32, u32 } -uniform_int_impl! { u64, u64, u64 } -uniform_int_impl! { usize, usize, usize } -uniform_int_impl! { u128, u128, u128 } - -#[cfg(feature = "simd_support")] -macro_rules! uniform_simd_int_impl { - ($ty:ident, $unsigned:ident, $u_scalar:ident) => { - // The "pick the largest zone that can fit in an `u32`" optimization - // is less useful here. Multiple lanes complicate things, we don't - // know the PRNG's minimal output size, and casting to a larger vector - // is generally a bad idea for SIMD performance. The user can still - // implement it manually. - - // TODO: look into `Uniform::::new(0u32, 100)` functionality - // perhaps `impl SampleUniform for $u_scalar`? - impl SampleUniform for $ty { - type Sampler = UniformInt<$ty>; - } - - impl UniformSampler for UniformInt<$ty> { - type X = $ty; - - #[inline] // if the range is constant, this helps LLVM to do the - // calculations at compile-time. - fn new(low_b: B1, high_b: B2) -> Self - where B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized - { - let low = *low_b.borrow(); - let high = *high_b.borrow(); - assert!(low.lt(high).all(), "Uniform::new called with `low >= high`"); - UniformSampler::new_inclusive(low, high - 1) - } - - #[inline] // if the range is constant, this helps LLVM to do the - // calculations at compile-time. - fn new_inclusive(low_b: B1, high_b: B2) -> Self - where B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized - { - let low = *low_b.borrow(); - let high = *high_b.borrow(); - assert!(low.le(high).all(), - "Uniform::new_inclusive called with `low > high`"); - let unsigned_max = ::core::$u_scalar::MAX; - - // NOTE: these may need to be replaced with explicitly - // wrapping operations if `packed_simd` changes - let range: $unsigned = ((high - low) + 1).cast(); - // `% 0` will panic at runtime. - let not_full_range = range.gt($unsigned::splat(0)); - // replacing 0 with `unsigned_max` allows a faster `select` - // with bitwise OR - let modulo = not_full_range.select(range, $unsigned::splat(unsigned_max)); - // wrapping addition - let ints_to_reject = (unsigned_max - range + 1) % modulo; - // When `range` is 0, `lo` of `v.wmul(range)` will always be - // zero which means only one sample is needed. - let zone = unsigned_max - ints_to_reject; - - UniformInt { - low, - // These are really $unsigned values, but store as $ty: - range: range.cast(), - z: zone.cast(), - } - } - - fn sample(&self, rng: &mut R) -> Self::X { - let range: $unsigned = self.range.cast(); - let zone: $unsigned = self.z.cast(); - - // This might seem very slow, generating a whole new - // SIMD vector for every sample rejection. For most uses - // though, the chance of rejection is small and provides good - // general performance. With multiple lanes, that chance is - // multiplied. To mitigate this, we replace only the lanes of - // the vector which fail, iteratively reducing the chance of - // rejection. The replacement method does however add a little - // overhead. Benchmarking or calculating probabilities might - // reveal contexts where this replacement method is slower. - let mut v: $unsigned = rng.gen(); - loop { - let (hi, lo) = v.wmul(range); - let mask = lo.le(zone); - if mask.all() { - let hi: $ty = hi.cast(); - // wrapping addition - let result = self.low + hi; - // `select` here compiles to a blend operation - // When `range.eq(0).none()` the compare and blend - // operations are avoided. - let v: $ty = v.cast(); - return range.gt($unsigned::splat(0)).select(result, v); - } - // Replace only the failing lanes - v = mask.select(v, rng.gen()); - } - } - } - }; - - // bulk implementation - ($(($unsigned:ident, $signed:ident),)+ $u_scalar:ident) => { - $( - uniform_simd_int_impl!($unsigned, $unsigned, $u_scalar); - uniform_simd_int_impl!($signed, $unsigned, $u_scalar); - )+ - }; -} - -#[cfg(feature = "simd_support")] -uniform_simd_int_impl! { - (u64x2, i64x2), - (u64x4, i64x4), - (u64x8, i64x8), - u64 -} - -#[cfg(feature = "simd_support")] -uniform_simd_int_impl! { - (u32x2, i32x2), - (u32x4, i32x4), - (u32x8, i32x8), - (u32x16, i32x16), - u32 -} - -#[cfg(feature = "simd_support")] -uniform_simd_int_impl! { - (u16x2, i16x2), - (u16x4, i16x4), - (u16x8, i16x8), - (u16x16, i16x16), - (u16x32, i16x32), - u16 -} - -#[cfg(feature = "simd_support")] -uniform_simd_int_impl! { - (u8x2, i8x2), - (u8x4, i8x4), - (u8x8, i8x8), - (u8x16, i8x16), - (u8x32, i8x32), - (u8x64, i8x64), - u8 -} - -impl SampleUniform for char { - type Sampler = UniformChar; -} - -/// The back-end implementing [`UniformSampler`] for `char`. -/// -/// Unless you are implementing [`UniformSampler`] for your own type, this type -/// should not be used directly, use [`Uniform`] instead. -/// -/// This differs from integer range sampling since the range `0xD800..=0xDFFF` -/// are used for surrogate pairs in UCS and UTF-16, and consequently are not -/// valid Unicode code points. We must therefore avoid sampling values in this -/// range. -#[derive(Clone, Copy, Debug)] -#[cfg_attr(feature = "serde1", derive(Serialize, Deserialize))] -pub struct UniformChar { - sampler: UniformInt, -} - -/// UTF-16 surrogate range start -const CHAR_SURROGATE_START: u32 = 0xD800; -/// UTF-16 surrogate range size -const CHAR_SURROGATE_LEN: u32 = 0xE000 - CHAR_SURROGATE_START; - -/// Convert `char` to compressed `u32` -fn char_to_comp_u32(c: char) -> u32 { - match c as u32 { - c if c >= CHAR_SURROGATE_START => c - CHAR_SURROGATE_LEN, - c => c, - } -} - -impl UniformSampler for UniformChar { - type X = char; - - #[inline] // if the range is constant, this helps LLVM to do the - // calculations at compile-time. - fn new(low_b: B1, high_b: B2) -> Self - where - B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized, - { - let low = char_to_comp_u32(*low_b.borrow()); - let high = char_to_comp_u32(*high_b.borrow()); - let sampler = UniformInt::::new(low, high); - UniformChar { sampler } - } - - #[inline] // if the range is constant, this helps LLVM to do the - // calculations at compile-time. - fn new_inclusive(low_b: B1, high_b: B2) -> Self - where - B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized, - { - let low = char_to_comp_u32(*low_b.borrow()); - let high = char_to_comp_u32(*high_b.borrow()); - let sampler = UniformInt::::new_inclusive(low, high); - UniformChar { sampler } - } - - fn sample(&self, rng: &mut R) -> Self::X { - let mut x = self.sampler.sample(rng); - if x >= CHAR_SURROGATE_START { - x += CHAR_SURROGATE_LEN; - } - // SAFETY: x must not be in surrogate range or greater than char::MAX. - // This relies on range constructors which accept char arguments. - // Validity of input char values is assumed. - unsafe { core::char::from_u32_unchecked(x) } - } -} - -/// The back-end implementing [`UniformSampler`] for floating-point types. -/// -/// Unless you are implementing [`UniformSampler`] for your own type, this type -/// should not be used directly, use [`Uniform`] instead. -/// -/// # Implementation notes -/// -/// Instead of generating a float in the `[0, 1)` range using [`Standard`], the -/// `UniformFloat` implementation converts the output of an PRNG itself. This -/// way one or two steps can be optimized out. -/// -/// The floats are first converted to a value in the `[1, 2)` interval using a -/// transmute-based method, and then mapped to the expected range with a -/// multiply and addition. Values produced this way have what equals 23 bits of -/// random digits for an `f32`, and 52 for an `f64`. -/// -/// [`new`]: UniformSampler::new -/// [`new_inclusive`]: UniformSampler::new_inclusive -/// [`Standard`]: crate::distributions::Standard -#[derive(Clone, Copy, Debug, PartialEq)] -#[cfg_attr(feature = "serde1", derive(Serialize, Deserialize))] -pub struct UniformFloat { - low: X, - scale: X, -} - -macro_rules! uniform_float_impl { - ($ty:ty, $uty:ident, $f_scalar:ident, $u_scalar:ident, $bits_to_discard:expr) => { - impl SampleUniform for $ty { - type Sampler = UniformFloat<$ty>; - } - - impl UniformSampler for UniformFloat<$ty> { - type X = $ty; - - fn new(low_b: B1, high_b: B2) -> Self - where - B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized, - { - let low = *low_b.borrow(); - let high = *high_b.borrow(); - debug_assert!( - low.all_finite(), - "Uniform::new called with `low` non-finite." - ); - debug_assert!( - high.all_finite(), - "Uniform::new called with `high` non-finite." - ); - assert!(low.all_lt(high), "Uniform::new called with `low >= high`"); - let max_rand = <$ty>::splat( - (::core::$u_scalar::MAX >> $bits_to_discard).into_float_with_exponent(0) - 1.0, - ); - - let mut scale = high - low; - assert!(scale.all_finite(), "Uniform::new: range overflow"); - - loop { - let mask = (scale * max_rand + low).ge_mask(high); - if mask.none() { - break; - } - scale = scale.decrease_masked(mask); - } - - debug_assert!(<$ty>::splat(0.0).all_le(scale)); - - UniformFloat { low, scale } - } - - fn new_inclusive(low_b: B1, high_b: B2) -> Self - where - B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized, - { - let low = *low_b.borrow(); - let high = *high_b.borrow(); - debug_assert!( - low.all_finite(), - "Uniform::new_inclusive called with `low` non-finite." - ); - debug_assert!( - high.all_finite(), - "Uniform::new_inclusive called with `high` non-finite." - ); - assert!( - low.all_le(high), - "Uniform::new_inclusive called with `low > high`" - ); - let max_rand = <$ty>::splat( - (::core::$u_scalar::MAX >> $bits_to_discard).into_float_with_exponent(0) - 1.0, - ); - - let mut scale = (high - low) / max_rand; - assert!(scale.all_finite(), "Uniform::new_inclusive: range overflow"); - - loop { - let mask = (scale * max_rand + low).gt_mask(high); - if mask.none() { - break; - } - scale = scale.decrease_masked(mask); - } - - debug_assert!(<$ty>::splat(0.0).all_le(scale)); - - UniformFloat { low, scale } - } - - fn sample(&self, rng: &mut R) -> Self::X { - // Generate a value in the range [1, 2) - let value1_2 = (rng.gen::<$uty>() >> $bits_to_discard).into_float_with_exponent(0); - - // Get a value in the range [0, 1) in order to avoid - // overflowing into infinity when multiplying with scale - let value0_1 = value1_2 - 1.0; - - // We don't use `f64::mul_add`, because it is not available with - // `no_std`. Furthermore, it is slower for some targets (but - // faster for others). However, the order of multiplication and - // addition is important, because on some platforms (e.g. ARM) - // it will be optimized to a single (non-FMA) instruction. - value0_1 * self.scale + self.low - } - - #[inline] - fn sample_single(low_b: B1, high_b: B2, rng: &mut R) -> Self::X - where - B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized, - { - let low = *low_b.borrow(); - let high = *high_b.borrow(); - debug_assert!( - low.all_finite(), - "UniformSampler::sample_single called with `low` non-finite." - ); - debug_assert!( - high.all_finite(), - "UniformSampler::sample_single called with `high` non-finite." - ); - assert!( - low.all_lt(high), - "UniformSampler::sample_single: low >= high" - ); - let mut scale = high - low; - assert!(scale.all_finite(), "UniformSampler::sample_single: range overflow"); - - loop { - // Generate a value in the range [1, 2) - let value1_2 = - (rng.gen::<$uty>() >> $bits_to_discard).into_float_with_exponent(0); - - // Get a value in the range [0, 1) in order to avoid - // overflowing into infinity when multiplying with scale - let value0_1 = value1_2 - 1.0; - - // Doing multiply before addition allows some architectures - // to use a single instruction. - let res = value0_1 * scale + low; - - debug_assert!(low.all_le(res) || !scale.all_finite()); - if res.all_lt(high) { - return res; - } - - // This handles a number of edge cases. - // * `low` or `high` is NaN. In this case `scale` and - // `res` are going to end up as NaN. - // * `low` is negative infinity and `high` is finite. - // `scale` is going to be infinite and `res` will be - // NaN. - // * `high` is positive infinity and `low` is finite. - // `scale` is going to be infinite and `res` will - // be infinite or NaN (if value0_1 is 0). - // * `low` is negative infinity and `high` is positive - // infinity. `scale` will be infinite and `res` will - // be NaN. - // * `low` and `high` are finite, but `high - low` - // overflows to infinite. `scale` will be infinite - // and `res` will be infinite or NaN (if value0_1 is 0). - // So if `high` or `low` are non-finite, we are guaranteed - // to fail the `res < high` check above and end up here. - // - // While we technically should check for non-finite `low` - // and `high` before entering the loop, by doing the checks - // here instead, we allow the common case to avoid these - // checks. But we are still guaranteed that if `low` or - // `high` are non-finite we'll end up here and can do the - // appropriate checks. - // - // Likewise `high - low` overflowing to infinity is also - // rare, so handle it here after the common case. - let mask = !scale.finite_mask(); - if mask.any() { - assert!( - low.all_finite() && high.all_finite(), - "Uniform::sample_single: low and high must be finite" - ); - scale = scale.decrease_masked(mask); - } - } - } - } - }; -} - -uniform_float_impl! { f32, u32, f32, u32, 32 - 23 } -uniform_float_impl! { f64, u64, f64, u64, 64 - 52 } - -#[cfg(feature = "simd_support")] -uniform_float_impl! { f32x2, u32x2, f32, u32, 32 - 23 } -#[cfg(feature = "simd_support")] -uniform_float_impl! { f32x4, u32x4, f32, u32, 32 - 23 } -#[cfg(feature = "simd_support")] -uniform_float_impl! { f32x8, u32x8, f32, u32, 32 - 23 } -#[cfg(feature = "simd_support")] -uniform_float_impl! { f32x16, u32x16, f32, u32, 32 - 23 } - -#[cfg(feature = "simd_support")] -uniform_float_impl! { f64x2, u64x2, f64, u64, 64 - 52 } -#[cfg(feature = "simd_support")] -uniform_float_impl! { f64x4, u64x4, f64, u64, 64 - 52 } -#[cfg(feature = "simd_support")] -uniform_float_impl! { f64x8, u64x8, f64, u64, 64 - 52 } - - -/// The back-end implementing [`UniformSampler`] for `Duration`. -/// -/// Unless you are implementing [`UniformSampler`] for your own types, this type -/// should not be used directly, use [`Uniform`] instead. -#[derive(Clone, Copy, Debug)] -#[cfg_attr(feature = "serde1", derive(Serialize, Deserialize))] -pub struct UniformDuration { - mode: UniformDurationMode, - offset: u32, -} - -#[derive(Debug, Copy, Clone)] -#[cfg_attr(feature = "serde1", derive(Serialize, Deserialize))] -enum UniformDurationMode { - Small { - secs: u64, - nanos: Uniform, - }, - Medium { - nanos: Uniform, - }, - Large { - max_secs: u64, - max_nanos: u32, - secs: Uniform, - }, -} - -impl SampleUniform for Duration { - type Sampler = UniformDuration; -} - -impl UniformSampler for UniformDuration { - type X = Duration; - - #[inline] - fn new(low_b: B1, high_b: B2) -> Self - where - B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized, - { - let low = *low_b.borrow(); - let high = *high_b.borrow(); - assert!(low < high, "Uniform::new called with `low >= high`"); - UniformDuration::new_inclusive(low, high - Duration::new(0, 1)) - } - - #[inline] - fn new_inclusive(low_b: B1, high_b: B2) -> Self - where - B1: SampleBorrow + Sized, - B2: SampleBorrow + Sized, - { - let low = *low_b.borrow(); - let high = *high_b.borrow(); - assert!( - low <= high, - "Uniform::new_inclusive called with `low > high`" - ); - - let low_s = low.as_secs(); - let low_n = low.subsec_nanos(); - let mut high_s = high.as_secs(); - let mut high_n = high.subsec_nanos(); - - if high_n < low_n { - high_s -= 1; - high_n += 1_000_000_000; - } - - let mode = if low_s == high_s { - UniformDurationMode::Small { - secs: low_s, - nanos: Uniform::new_inclusive(low_n, high_n), - } - } else { - let max = high_s - .checked_mul(1_000_000_000) - .and_then(|n| n.checked_add(u64::from(high_n))); - - if let Some(higher_bound) = max { - let lower_bound = low_s * 1_000_000_000 + u64::from(low_n); - UniformDurationMode::Medium { - nanos: Uniform::new_inclusive(lower_bound, higher_bound), - } - } else { - // An offset is applied to simplify generation of nanoseconds - let max_nanos = high_n - low_n; - UniformDurationMode::Large { - max_secs: high_s, - max_nanos, - secs: Uniform::new_inclusive(low_s, high_s), - } - } - }; - UniformDuration { - mode, - offset: low_n, - } - } - - #[inline] - fn sample(&self, rng: &mut R) -> Duration { - match self.mode { - UniformDurationMode::Small { secs, nanos } => { - let n = nanos.sample(rng); - Duration::new(secs, n) - } - UniformDurationMode::Medium { nanos } => { - let nanos = nanos.sample(rng); - Duration::new(nanos / 1_000_000_000, (nanos % 1_000_000_000) as u32) - } - UniformDurationMode::Large { - max_secs, - max_nanos, - secs, - } => { - // constant folding means this is at least as fast as `Rng::sample(Range)` - let nano_range = Uniform::new(0, 1_000_000_000); - loop { - let s = secs.sample(rng); - let n = nano_range.sample(rng); - if !(s == max_secs && n > max_nanos) { - let sum = n + self.offset; - break Duration::new(s, sum); - } - } - } - } - } -} - #[cfg(test)] mod tests { use super::*; - use crate::rngs::mock::StepRng; - - #[test] - #[cfg(feature = "serde1")] - fn test_serialization_uniform_duration() { - let distr = UniformDuration::new(Duration::from_secs(10), Duration::from_secs(60)); - let de_distr: UniformDuration = bincode::deserialize(&bincode::serialize(&distr).unwrap()).unwrap(); - assert_eq!( - distr.offset, de_distr.offset - ); - match (distr.mode, de_distr.mode) { - (UniformDurationMode::Small {secs: a_secs, nanos: a_nanos}, UniformDurationMode::Small {secs, nanos}) => { - assert_eq!(a_secs, secs); - - assert_eq!(a_nanos.0.low, nanos.0.low); - assert_eq!(a_nanos.0.range, nanos.0.range); - assert_eq!(a_nanos.0.z, nanos.0.z); - } - (UniformDurationMode::Medium {nanos: a_nanos} , UniformDurationMode::Medium {nanos}) => { - assert_eq!(a_nanos.0.low, nanos.0.low); - assert_eq!(a_nanos.0.range, nanos.0.range); - assert_eq!(a_nanos.0.z, nanos.0.z); - } - (UniformDurationMode::Large {max_secs:a_max_secs, max_nanos:a_max_nanos, secs:a_secs}, UniformDurationMode::Large {max_secs, max_nanos, secs} ) => { - assert_eq!(a_max_secs, max_secs); - assert_eq!(a_max_nanos, max_nanos); - - assert_eq!(a_secs.0.low, secs.0.low); - assert_eq!(a_secs.0.range, secs.0.range); - assert_eq!(a_secs.0.z, secs.0.z); - } - _ => panic!("`UniformDurationMode` was not serialized/deserialized correctly") - } - } - - #[test] - #[cfg(feature = "serde1")] - fn test_uniform_serialization() { - let unit_box: Uniform = Uniform::new(-1, 1); - let de_unit_box: Uniform = bincode::deserialize(&bincode::serialize(&unit_box).unwrap()).unwrap(); - - assert_eq!(unit_box.0.low, de_unit_box.0.low); - assert_eq!(unit_box.0.range, de_unit_box.0.range); - assert_eq!(unit_box.0.z, de_unit_box.0.z); - - let unit_box: Uniform = Uniform::new(-1., 1.); - let de_unit_box: Uniform = bincode::deserialize(&bincode::serialize(&unit_box).unwrap()).unwrap(); - - assert_eq!(unit_box.0.low, de_unit_box.0.low); - assert_eq!(unit_box.0.scale, de_unit_box.0.scale); - } - - #[should_panic] - #[test] - fn test_uniform_bad_limits_equal_int() { - Uniform::new(10, 10); - } - - #[test] - fn test_uniform_good_limits_equal_int() { - let mut rng = crate::test::rng(804); - let dist = Uniform::new_inclusive(10, 10); - for _ in 0..20 { - assert_eq!(rng.sample(dist), 10); - } - } - - #[should_panic] - #[test] - fn test_uniform_bad_limits_flipped_int() { - Uniform::new(10, 5); - } - - #[test] - #[cfg_attr(miri, ignore)] // Miri is too slow - fn test_integers() { - use core::{i128, u128}; - use core::{i16, i32, i64, i8, isize}; - use core::{u16, u32, u64, u8, usize}; - - let mut rng = crate::test::rng(251); - macro_rules! t { - ($ty:ident, $v:expr, $le:expr, $lt:expr) => {{ - for &(low, high) in $v.iter() { - let my_uniform = Uniform::new(low, high); - for _ in 0..1000 { - let v: $ty = rng.sample(my_uniform); - assert!($le(low, v) && $lt(v, high)); - } - - let my_uniform = Uniform::new_inclusive(low, high); - for _ in 0..1000 { - let v: $ty = rng.sample(my_uniform); - assert!($le(low, v) && $le(v, high)); - } - - let my_uniform = Uniform::new(&low, high); - for _ in 0..1000 { - let v: $ty = rng.sample(my_uniform); - assert!($le(low, v) && $lt(v, high)); - } - - let my_uniform = Uniform::new_inclusive(&low, &high); - for _ in 0..1000 { - let v: $ty = rng.sample(my_uniform); - assert!($le(low, v) && $le(v, high)); - } - - for _ in 0..1000 { - let v = <$ty as SampleUniform>::Sampler::sample_single(low, high, &mut rng); - assert!($le(low, v) && $lt(v, high)); - } - - for _ in 0..1000 { - let v = <$ty as SampleUniform>::Sampler::sample_single_inclusive(low, high, &mut rng); - assert!($le(low, v) && $le(v, high)); - } - } - }}; - - // scalar bulk - ($($ty:ident),*) => {{ - $(t!( - $ty, - [(0, 10), (10, 127), ($ty::MIN, $ty::MAX)], - |x, y| x <= y, - |x, y| x < y - );)* - }}; - - // simd bulk - ($($ty:ident),* => $scalar:ident) => {{ - $(t!( - $ty, - [ - ($ty::splat(0), $ty::splat(10)), - ($ty::splat(10), $ty::splat(127)), - ($ty::splat($scalar::MIN), $ty::splat($scalar::MAX)), - ], - |x: $ty, y| x.le(y).all(), - |x: $ty, y| x.lt(y).all() - );)* - }}; - } - t!(i8, i16, i32, i64, isize, u8, u16, u32, u64, usize, i128, u128); - - #[cfg(feature = "simd_support")] - { - t!(u8x2, u8x4, u8x8, u8x16, u8x32, u8x64 => u8); - t!(i8x2, i8x4, i8x8, i8x16, i8x32, i8x64 => i8); - t!(u16x2, u16x4, u16x8, u16x16, u16x32 => u16); - t!(i16x2, i16x4, i16x8, i16x16, i16x32 => i16); - t!(u32x2, u32x4, u32x8, u32x16 => u32); - t!(i32x2, i32x4, i32x8, i32x16 => i32); - t!(u64x2, u64x4, u64x8 => u64); - t!(i64x2, i64x4, i64x8 => i64); - } - } - - #[test] - #[cfg_attr(miri, ignore)] // Miri is too slow - fn test_char() { - let mut rng = crate::test::rng(891); - let mut max = core::char::from_u32(0).unwrap(); - for _ in 0..100 { - let c = rng.gen_range('A'..='Z'); - assert!(('A'..='Z').contains(&c)); - max = max.max(c); - } - assert_eq!(max, 'Z'); - let d = Uniform::new( - core::char::from_u32(0xD7F0).unwrap(), - core::char::from_u32(0xE010).unwrap(), - ); - for _ in 0..100 { - let c = d.sample(&mut rng); - assert!((c as u32) < 0xD800 || (c as u32) > 0xDFFF); - } - } - - #[test] - #[cfg_attr(miri, ignore)] // Miri is too slow - fn test_floats() { - let mut rng = crate::test::rng(252); - let mut zero_rng = StepRng::new(0, 0); - let mut max_rng = StepRng::new(0xffff_ffff_ffff_ffff, 0); - macro_rules! t { - ($ty:ty, $f_scalar:ident, $bits_shifted:expr) => {{ - let v: &[($f_scalar, $f_scalar)] = &[ - (0.0, 100.0), - (-1e35, -1e25), - (1e-35, 1e-25), - (-1e35, 1e35), - (<$f_scalar>::from_bits(0), <$f_scalar>::from_bits(3)), - (-<$f_scalar>::from_bits(10), -<$f_scalar>::from_bits(1)), - (-<$f_scalar>::from_bits(5), 0.0), - (-<$f_scalar>::from_bits(7), -0.0), - (0.1 * ::core::$f_scalar::MAX, ::core::$f_scalar::MAX), - (-::core::$f_scalar::MAX * 0.2, ::core::$f_scalar::MAX * 0.7), - ]; - for &(low_scalar, high_scalar) in v.iter() { - for lane in 0..<$ty>::lanes() { - let low = <$ty>::splat(0.0 as $f_scalar).replace(lane, low_scalar); - let high = <$ty>::splat(1.0 as $f_scalar).replace(lane, high_scalar); - let my_uniform = Uniform::new(low, high); - let my_incl_uniform = Uniform::new_inclusive(low, high); - for _ in 0..100 { - let v = rng.sample(my_uniform).extract(lane); - assert!(low_scalar <= v && v < high_scalar); - let v = rng.sample(my_incl_uniform).extract(lane); - assert!(low_scalar <= v && v <= high_scalar); - let v = <$ty as SampleUniform>::Sampler - ::sample_single(low, high, &mut rng).extract(lane); - assert!(low_scalar <= v && v < high_scalar); - } - - assert_eq!( - rng.sample(Uniform::new_inclusive(low, low)).extract(lane), - low_scalar - ); - - assert_eq!(zero_rng.sample(my_uniform).extract(lane), low_scalar); - assert_eq!(zero_rng.sample(my_incl_uniform).extract(lane), low_scalar); - assert_eq!(<$ty as SampleUniform>::Sampler - ::sample_single(low, high, &mut zero_rng) - .extract(lane), low_scalar); - assert!(max_rng.sample(my_uniform).extract(lane) < high_scalar); - assert!(max_rng.sample(my_incl_uniform).extract(lane) <= high_scalar); - - // Don't run this test for really tiny differences between high and low - // since for those rounding might result in selecting high for a very - // long time. - if (high_scalar - low_scalar) > 0.0001 { - let mut lowering_max_rng = StepRng::new( - 0xffff_ffff_ffff_ffff, - (-1i64 << $bits_shifted) as u64, - ); - assert!( - <$ty as SampleUniform>::Sampler - ::sample_single(low, high, &mut lowering_max_rng) - .extract(lane) < high_scalar - ); - } - } - } - - assert_eq!( - rng.sample(Uniform::new_inclusive( - ::core::$f_scalar::MAX, - ::core::$f_scalar::MAX - )), - ::core::$f_scalar::MAX - ); - assert_eq!( - rng.sample(Uniform::new_inclusive( - -::core::$f_scalar::MAX, - -::core::$f_scalar::MAX - )), - -::core::$f_scalar::MAX - ); - }}; - } - - t!(f32, f32, 32 - 23); - t!(f64, f64, 64 - 52); - #[cfg(feature = "simd_support")] - { - t!(f32x2, f32, 32 - 23); - t!(f32x4, f32, 32 - 23); - t!(f32x8, f32, 32 - 23); - t!(f32x16, f32, 32 - 23); - t!(f64x2, f64, 64 - 52); - t!(f64x4, f64, 64 - 52); - t!(f64x8, f64, 64 - 52); - } - } - - #[test] - #[should_panic] - fn test_float_overflow() { - let _ = Uniform::from(::core::f64::MIN..::core::f64::MAX); - } - - #[test] - #[should_panic] - fn test_float_overflow_single() { - let mut rng = crate::test::rng(252); - rng.gen_range(::core::f64::MIN..::core::f64::MAX); - } - - #[test] - #[cfg(all( - feature = "std", - not(target_arch = "wasm32"), - not(target_arch = "asmjs") - ))] - fn test_float_assertions() { - use super::SampleUniform; - use std::panic::catch_unwind; - fn range(low: T, high: T) { - let mut rng = crate::test::rng(253); - T::Sampler::sample_single(low, high, &mut rng); - } - - macro_rules! t { - ($ty:ident, $f_scalar:ident) => {{ - let v: &[($f_scalar, $f_scalar)] = &[ - (::std::$f_scalar::NAN, 0.0), - (1.0, ::std::$f_scalar::NAN), - (::std::$f_scalar::NAN, ::std::$f_scalar::NAN), - (1.0, 0.5), - (::std::$f_scalar::MAX, -::std::$f_scalar::MAX), - (::std::$f_scalar::INFINITY, ::std::$f_scalar::INFINITY), - ( - ::std::$f_scalar::NEG_INFINITY, - ::std::$f_scalar::NEG_INFINITY, - ), - (::std::$f_scalar::NEG_INFINITY, 5.0), - (5.0, ::std::$f_scalar::INFINITY), - (::std::$f_scalar::NAN, ::std::$f_scalar::INFINITY), - (::std::$f_scalar::NEG_INFINITY, ::std::$f_scalar::NAN), - (::std::$f_scalar::NEG_INFINITY, ::std::$f_scalar::INFINITY), - ]; - for &(low_scalar, high_scalar) in v.iter() { - for lane in 0..<$ty>::lanes() { - let low = <$ty>::splat(0.0 as $f_scalar).replace(lane, low_scalar); - let high = <$ty>::splat(1.0 as $f_scalar).replace(lane, high_scalar); - assert!(catch_unwind(|| range(low, high)).is_err()); - assert!(catch_unwind(|| Uniform::new(low, high)).is_err()); - assert!(catch_unwind(|| Uniform::new_inclusive(low, high)).is_err()); - assert!(catch_unwind(|| range(low, low)).is_err()); - assert!(catch_unwind(|| Uniform::new(low, low)).is_err()); - } - } - }}; - } - - t!(f32, f32); - t!(f64, f64); - #[cfg(feature = "simd_support")] - { - t!(f32x2, f32); - t!(f32x4, f32); - t!(f32x8, f32); - t!(f32x16, f32); - t!(f64x2, f64); - t!(f64x4, f64); - t!(f64x8, f64); - } - } - - - #[test] - #[cfg_attr(miri, ignore)] // Miri is too slow - fn test_durations() { - let mut rng = crate::test::rng(253); - - let v = &[ - (Duration::new(10, 50000), Duration::new(100, 1234)), - (Duration::new(0, 100), Duration::new(1, 50)), - ( - Duration::new(0, 0), - Duration::new(u64::max_value(), 999_999_999), - ), - ]; - for &(low, high) in v.iter() { - let my_uniform = Uniform::new(low, high); - for _ in 0..1000 { - let v = rng.sample(my_uniform); - assert!(low <= v && v < high); - } - } - } #[test] fn test_custom_uniform() { @@ -1573,86 +437,21 @@ mod tests { } } - #[test] - fn test_uniform_from_std_range() { - let r = Uniform::from(2u32..7); - assert_eq!(r.0.low, 2); - assert_eq!(r.0.range, 5); - let r = Uniform::from(2.0f64..7.0); - assert_eq!(r.0.low, 2.0); - assert_eq!(r.0.scale, 5.0); - } - - #[test] - fn test_uniform_from_std_range_inclusive() { - let r = Uniform::from(2u32..=6); - assert_eq!(r.0.low, 2); - assert_eq!(r.0.range, 5); - let r = Uniform::from(2.0f64..=7.0); - assert_eq!(r.0.low, 2.0); - assert!(r.0.scale > 5.0); - assert!(r.0.scale < 5.0 + 1e-14); - } - - #[test] - fn value_stability() { - fn test_samples( - lb: T, ub: T, expected_single: &[T], expected_multiple: &[T], - ) where Uniform: Distribution { - let mut rng = crate::test::rng(897); - let mut buf = [lb; 3]; - - for x in &mut buf { - *x = T::Sampler::sample_single(lb, ub, &mut rng); - } - assert_eq!(&buf, expected_single); + pub(crate) fn test_samples( + lb: T, ub: T, expected_single: &[T], expected_multiple: &[T], + ) where Uniform: Distribution { + let mut rng = crate::test::rng(897); + let mut buf = [lb; 3]; - let distr = Uniform::new(lb, ub); - for x in &mut buf { - *x = rng.sample(&distr); - } - assert_eq!(&buf, expected_multiple); + for x in &mut buf { + *x = T::Sampler::sample_single(lb, ub, &mut rng); } + assert_eq!(&buf, expected_single); - // We test on a sub-set of types; possibly we should do more. - // TODO: SIMD types - - test_samples(11u8, 219, &[17, 66, 214], &[181, 93, 165]); - test_samples(11u32, 219, &[17, 66, 214], &[181, 93, 165]); - - test_samples(0f32, 1e-2f32, &[0.0003070104, 0.0026630748, 0.00979833], &[ - 0.008194133, - 0.00398172, - 0.007428536, - ]); - test_samples( - -1e10f64, - 1e10f64, - &[-4673848682.871551, 6388267422.932352, 4857075081.198343], - &[1173375212.1808167, 1917642852.109581, 2365076174.3153973], - ); - - test_samples( - Duration::new(2, 0), - Duration::new(4, 0), - &[ - Duration::new(2, 532615131), - Duration::new(3, 638826742), - Duration::new(3, 485707508), - ], - &[ - Duration::new(3, 117337521), - Duration::new(3, 191764285), - Duration::new(3, 236507617), - ], - ); - } - - #[test] - fn uniform_distributions_can_be_compared() { - assert_eq!(Uniform::new(1.0, 2.0), Uniform::new(1.0, 2.0)); - - // To cover UniformInt - assert_eq!(Uniform::new(1 as u32, 2 as u32), Uniform::new(1 as u32, 2 as u32)); + let distr = Uniform::new(lb, ub); + for x in &mut buf { + *x = rng.sample(&distr); + } + assert_eq!(&buf, expected_multiple); } } diff --git a/src/distributions/uniform/uniform_float.rs b/src/distributions/uniform/uniform_float.rs new file mode 100644 index 00000000000..cf4e37e067c --- /dev/null +++ b/src/distributions/uniform/uniform_float.rs @@ -0,0 +1,480 @@ +// Copyright 2018-2021 Developers of the Rand project. +// +// Licensed under the Apache License, Version 2.0 or the MIT license +// , at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +use super::{SampleBorrow, SampleUniform, UniformSampler}; +use crate::distributions::float::IntoFloat; +use crate::distributions::utils::{BoolAsSIMD, FloatAsSIMD, FloatSIMDUtils}; +use crate::Rng; +#[cfg(feature = "simd_support")] use packed_simd::*; +#[cfg(feature = "serde1")] use serde::{Deserialize, Serialize}; + +/// The back-end implementing [`UniformSampler`] for floating-point types. +/// +/// Unless you are implementing [`UniformSampler`] for your own type, this type +/// should not be used directly, use [`Uniform`] instead. +/// +/// # Implementation notes +/// +/// Instead of generating a float in the `[0, 1)` range using [`Standard`], the +/// `UniformFloat` implementation converts the output of an PRNG itself. This +/// way one or two steps can be optimized out. +/// +/// The floats are first converted to a value in the `[1, 2)` interval using a +/// transmute-based method, and then mapped to the expected range with a +/// multiply and addition. Values produced this way have what equals 23 bits of +/// random digits for an `f32`, and 52 for an `f64`. +/// +/// [`new`]: UniformSampler::new +/// [`new_inclusive`]: UniformSampler::new_inclusive +/// [`Standard`]: crate::distributions::Standard +#[derive(Clone, Copy, Debug, PartialEq)] +#[cfg_attr(feature = "serde1", derive(Serialize, Deserialize))] +pub struct UniformFloat { + low: X, + scale: X, +} + +macro_rules! uniform_float_impl { + ($ty:ty, $uty:ident, $f_scalar:ident, $u_scalar:ident, $bits_to_discard:expr) => { + impl SampleUniform for $ty { + type Sampler = UniformFloat<$ty>; + } + + impl UniformSampler for UniformFloat<$ty> { + type X = $ty; + + fn new(low_b: B1, high_b: B2) -> Self + where + B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized, + { + let low = *low_b.borrow(); + let high = *high_b.borrow(); + debug_assert!( + low.all_finite(), + "Uniform::new called with `low` non-finite." + ); + debug_assert!( + high.all_finite(), + "Uniform::new called with `high` non-finite." + ); + assert!(low.all_lt(high), "Uniform::new called with `low >= high`"); + let max_rand = <$ty>::splat( + (::core::$u_scalar::MAX >> $bits_to_discard).into_float_with_exponent(0) - 1.0, + ); + + let mut scale = high - low; + assert!(scale.all_finite(), "Uniform::new: range overflow"); + + loop { + let mask = (scale * max_rand + low).ge_mask(high); + if mask.none() { + break; + } + scale = scale.decrease_masked(mask); + } + + debug_assert!(<$ty>::splat(0.0).all_le(scale)); + + UniformFloat { low, scale } + } + + fn new_inclusive(low_b: B1, high_b: B2) -> Self + where + B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized, + { + let low = *low_b.borrow(); + let high = *high_b.borrow(); + debug_assert!( + low.all_finite(), + "Uniform::new_inclusive called with `low` non-finite." + ); + debug_assert!( + high.all_finite(), + "Uniform::new_inclusive called with `high` non-finite." + ); + assert!( + low.all_le(high), + "Uniform::new_inclusive called with `low > high`" + ); + let max_rand = <$ty>::splat( + (::core::$u_scalar::MAX >> $bits_to_discard).into_float_with_exponent(0) - 1.0, + ); + + let mut scale = (high - low) / max_rand; + assert!(scale.all_finite(), "Uniform::new_inclusive: range overflow"); + + loop { + let mask = (scale * max_rand + low).gt_mask(high); + if mask.none() { + break; + } + scale = scale.decrease_masked(mask); + } + + debug_assert!(<$ty>::splat(0.0).all_le(scale)); + + UniformFloat { low, scale } + } + + fn sample(&self, rng: &mut R) -> Self::X { + // Generate a value in the range [1, 2) + let value1_2 = (rng.gen::<$uty>() >> $bits_to_discard).into_float_with_exponent(0); + + // Get a value in the range [0, 1) in order to avoid + // overflowing into infinity when multiplying with scale + let value0_1 = value1_2 - 1.0; + + // We don't use `f64::mul_add`, because it is not available with + // `no_std`. Furthermore, it is slower for some targets (but + // faster for others). However, the order of multiplication and + // addition is important, because on some platforms (e.g. ARM) + // it will be optimized to a single (non-FMA) instruction. + value0_1 * self.scale + self.low + } + + #[inline] + fn sample_single(low_b: B1, high_b: B2, rng: &mut R) -> Self::X + where + B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized, + { + let low = *low_b.borrow(); + let high = *high_b.borrow(); + debug_assert!( + low.all_finite(), + "UniformSampler::sample_single called with `low` non-finite." + ); + debug_assert!( + high.all_finite(), + "UniformSampler::sample_single called with `high` non-finite." + ); + assert!( + low.all_lt(high), + "UniformSampler::sample_single: low >= high" + ); + let mut scale = high - low; + assert!( + scale.all_finite(), + "UniformSampler::sample_single: range overflow" + ); + + loop { + // Generate a value in the range [1, 2) + let value1_2 = + (rng.gen::<$uty>() >> $bits_to_discard).into_float_with_exponent(0); + + // Get a value in the range [0, 1) in order to avoid + // overflowing into infinity when multiplying with scale + let value0_1 = value1_2 - 1.0; + + // Doing multiply before addition allows some architectures + // to use a single instruction. + let res = value0_1 * scale + low; + + debug_assert!(low.all_le(res) || !scale.all_finite()); + if res.all_lt(high) { + return res; + } + + // This handles a number of edge cases. + // * `low` or `high` is NaN. In this case `scale` and + // `res` are going to end up as NaN. + // * `low` is negative infinity and `high` is finite. + // `scale` is going to be infinite and `res` will be + // NaN. + // * `high` is positive infinity and `low` is finite. + // `scale` is going to be infinite and `res` will + // be infinite or NaN (if value0_1 is 0). + // * `low` is negative infinity and `high` is positive + // infinity. `scale` will be infinite and `res` will + // be NaN. + // * `low` and `high` are finite, but `high - low` + // overflows to infinite. `scale` will be infinite + // and `res` will be infinite or NaN (if value0_1 is 0). + // So if `high` or `low` are non-finite, we are guaranteed + // to fail the `res < high` check above and end up here. + // + // While we technically should check for non-finite `low` + // and `high` before entering the loop, by doing the checks + // here instead, we allow the common case to avoid these + // checks. But we are still guaranteed that if `low` or + // `high` are non-finite we'll end up here and can do the + // appropriate checks. + // + // Likewise `high - low` overflowing to infinity is also + // rare, so handle it here after the common case. + let mask = !scale.finite_mask(); + if mask.any() { + assert!( + low.all_finite() && high.all_finite(), + "Uniform::sample_single: low and high must be finite" + ); + scale = scale.decrease_masked(mask); + } + } + } + } + }; +} + +uniform_float_impl! { f32, u32, f32, u32, 32 - 23 } +uniform_float_impl! { f64, u64, f64, u64, 64 - 52 } + +#[cfg(feature = "simd_support")] +uniform_float_impl! { f32x2, u32x2, f32, u32, 32 - 23 } +#[cfg(feature = "simd_support")] +uniform_float_impl! { f32x4, u32x4, f32, u32, 32 - 23 } +#[cfg(feature = "simd_support")] +uniform_float_impl! { f32x8, u32x8, f32, u32, 32 - 23 } +#[cfg(feature = "simd_support")] +uniform_float_impl! { f32x16, u32x16, f32, u32, 32 - 23 } + +#[cfg(feature = "simd_support")] +uniform_float_impl! { f64x2, u64x2, f64, u64, 64 - 52 } +#[cfg(feature = "simd_support")] +uniform_float_impl! { f64x4, u64x4, f64, u64, 64 - 52 } +#[cfg(feature = "simd_support")] +uniform_float_impl! { f64x8, u64x8, f64, u64, 64 - 52 } + +#[cfg(test)] +mod tests { + use super::*; + use crate::distributions::uniform::tests::test_samples; + use crate::distributions::Uniform; + use crate::rngs::mock::StepRng; + + #[test] + #[cfg(feature = "serde1")] + fn test_uniform_serialization() { + let unit_box: Uniform = Uniform::new(-1., 1.); + let de_unit_box: Uniform = + bincode::deserialize(&bincode::serialize(&unit_box).unwrap()).unwrap(); + + assert_eq!(unit_box.0.low, de_unit_box.0.low); + assert_eq!(unit_box.0.scale, de_unit_box.0.scale); + } + + #[test] + #[cfg_attr(miri, ignore)] // Miri is too slow + fn test_floats() { + let mut rng = crate::test::rng(252); + let mut zero_rng = StepRng::new(0, 0); + let mut max_rng = StepRng::new(0xffff_ffff_ffff_ffff, 0); + macro_rules! t { + ($ty:ty, $f_scalar:ident, $bits_shifted:expr) => {{ + let v: &[($f_scalar, $f_scalar)] = &[ + (0.0, 100.0), + (-1e35, -1e25), + (1e-35, 1e-25), + (-1e35, 1e35), + (<$f_scalar>::from_bits(0), <$f_scalar>::from_bits(3)), + (-<$f_scalar>::from_bits(10), -<$f_scalar>::from_bits(1)), + (-<$f_scalar>::from_bits(5), 0.0), + (-<$f_scalar>::from_bits(7), -0.0), + (0.1 * ::core::$f_scalar::MAX, ::core::$f_scalar::MAX), + (-::core::$f_scalar::MAX * 0.2, ::core::$f_scalar::MAX * 0.7), + ]; + for &(low_scalar, high_scalar) in v.iter() { + for lane in 0..<$ty>::lanes() { + let low = <$ty>::splat(0.0 as $f_scalar).replace(lane, low_scalar); + let high = <$ty>::splat(1.0 as $f_scalar).replace(lane, high_scalar); + let my_uniform = Uniform::new(low, high); + let my_incl_uniform = Uniform::new_inclusive(low, high); + for _ in 0..100 { + let v = rng.sample(my_uniform).extract(lane); + assert!(low_scalar <= v && v < high_scalar); + let v = rng.sample(my_incl_uniform).extract(lane); + assert!(low_scalar <= v && v <= high_scalar); + let v = + <$ty as SampleUniform>::Sampler::sample_single(low, high, &mut rng) + .extract(lane); + assert!(low_scalar <= v && v < high_scalar); + } + + assert_eq!( + rng.sample(Uniform::new_inclusive(low, low)).extract(lane), + low_scalar + ); + + assert_eq!(zero_rng.sample(my_uniform).extract(lane), low_scalar); + assert_eq!(zero_rng.sample(my_incl_uniform).extract(lane), low_scalar); + assert_eq!( + <$ty as SampleUniform>::Sampler::sample_single( + low, + high, + &mut zero_rng + ) + .extract(lane), + low_scalar + ); + assert!(max_rng.sample(my_uniform).extract(lane) < high_scalar); + assert!(max_rng.sample(my_incl_uniform).extract(lane) <= high_scalar); + + // Don't run this test for really tiny differences between high and low + // since for those rounding might result in selecting high for a very + // long time. + if (high_scalar - low_scalar) > 0.0001 { + let mut lowering_max_rng = StepRng::new( + 0xffff_ffff_ffff_ffff, + (-1i64 << $bits_shifted) as u64, + ); + assert!( + <$ty as SampleUniform>::Sampler::sample_single( + low, + high, + &mut lowering_max_rng + ) + .extract(lane) + < high_scalar + ); + } + } + } + + assert_eq!( + rng.sample(Uniform::new_inclusive( + ::core::$f_scalar::MAX, + ::core::$f_scalar::MAX + )), + ::core::$f_scalar::MAX + ); + assert_eq!( + rng.sample(Uniform::new_inclusive( + -::core::$f_scalar::MAX, + -::core::$f_scalar::MAX + )), + -::core::$f_scalar::MAX + ); + }}; + } + + t!(f32, f32, 32 - 23); + t!(f64, f64, 64 - 52); + #[cfg(feature = "simd_support")] + { + t!(f32x2, f32, 32 - 23); + t!(f32x4, f32, 32 - 23); + t!(f32x8, f32, 32 - 23); + t!(f32x16, f32, 32 - 23); + t!(f64x2, f64, 64 - 52); + t!(f64x4, f64, 64 - 52); + t!(f64x8, f64, 64 - 52); + } + } + + #[test] + #[should_panic] + fn test_float_overflow() { + let _ = Uniform::from(::core::f64::MIN..::core::f64::MAX); + } + + #[test] + #[should_panic] + fn test_float_overflow_single() { + let mut rng = crate::test::rng(252); + rng.gen_range(::core::f64::MIN..::core::f64::MAX); + } + + #[test] + #[cfg(all( + feature = "std", + not(target_arch = "wasm32"), + not(target_arch = "asmjs") + ))] + fn test_float_assertions() { + use super::SampleUniform; + use std::panic::catch_unwind; + fn range(low: T, high: T) { + let mut rng = crate::test::rng(253); + T::Sampler::sample_single(low, high, &mut rng); + } + + macro_rules! t { + ($ty:ident, $f_scalar:ident) => {{ + let v: &[($f_scalar, $f_scalar)] = &[ + (::std::$f_scalar::NAN, 0.0), + (1.0, ::std::$f_scalar::NAN), + (::std::$f_scalar::NAN, ::std::$f_scalar::NAN), + (1.0, 0.5), + (::std::$f_scalar::MAX, -::std::$f_scalar::MAX), + (::std::$f_scalar::INFINITY, ::std::$f_scalar::INFINITY), + ( + ::std::$f_scalar::NEG_INFINITY, + ::std::$f_scalar::NEG_INFINITY, + ), + (::std::$f_scalar::NEG_INFINITY, 5.0), + (5.0, ::std::$f_scalar::INFINITY), + (::std::$f_scalar::NAN, ::std::$f_scalar::INFINITY), + (::std::$f_scalar::NEG_INFINITY, ::std::$f_scalar::NAN), + (::std::$f_scalar::NEG_INFINITY, ::std::$f_scalar::INFINITY), + ]; + for &(low_scalar, high_scalar) in v.iter() { + for lane in 0..<$ty>::lanes() { + let low = <$ty>::splat(0.0 as $f_scalar).replace(lane, low_scalar); + let high = <$ty>::splat(1.0 as $f_scalar).replace(lane, high_scalar); + assert!(catch_unwind(|| range(low, high)).is_err()); + assert!(catch_unwind(|| Uniform::new(low, high)).is_err()); + assert!(catch_unwind(|| Uniform::new_inclusive(low, high)).is_err()); + assert!(catch_unwind(|| range(low, low)).is_err()); + assert!(catch_unwind(|| Uniform::new(low, low)).is_err()); + } + } + }}; + } + + t!(f32, f32); + t!(f64, f64); + #[cfg(feature = "simd_support")] + { + t!(f32x2, f32); + t!(f32x4, f32); + t!(f32x8, f32); + t!(f32x16, f32); + t!(f64x2, f64); + t!(f64x4, f64); + t!(f64x8, f64); + } + } + + #[test] + fn test_uniform_from_std_range() { + let r = Uniform::from(2.0f64..7.0); + assert_eq!(r.0.low, 2.0); + assert_eq!(r.0.scale, 5.0); + } + + #[test] + fn test_uniform_from_std_range_inclusive() { + let r = Uniform::from(2.0f64..=7.0); + assert_eq!(r.0.low, 2.0); + assert!(r.0.scale > 5.0); + assert!(r.0.scale < 5.0 + 1e-14); + } + + #[test] + fn value_stability() { + test_samples(0f32, 1000000f32, &[30701.041, 266307.47, 979833.0], &[ + 819413.3, 398172.03, 742853.6, + ]); + test_samples( + -1f64, + 1f64, + &[-0.4673848682871551, 0.6388267422932352, 0.4857075081198343], + &[0.11733752121808161, 0.191764285210958, 0.2365076174315397], + ); + + // TODO: SIMD types + } + + #[test] + fn uniform_distributions_can_be_compared() { + assert_eq!(Uniform::new(1.0, 2.0), Uniform::new(1.0, 2.0)); + } +} diff --git a/src/distributions/uniform/uniform_int.rs b/src/distributions/uniform/uniform_int.rs new file mode 100644 index 00000000000..41123333a26 --- /dev/null +++ b/src/distributions/uniform/uniform_int.rs @@ -0,0 +1,411 @@ +// Copyright 2018-2021 Developers of the Rand project. +// +// Licensed under the Apache License, Version 2.0 or the MIT license +// , at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +use super::{SampleBorrow, SampleUniform, UniformSampler}; +use crate::distributions::utils::WideningMultiply; +use crate::Rng; +#[cfg(feature = "serde1")] use serde::{Deserialize, Serialize}; + +/// The back-end implementing [`UniformSampler`] for integer types. +/// +/// Unless you are implementing [`UniformSampler`] for your own type, this type +/// should not be used directly, use [`Uniform`] instead. +/// +/// # Implementation notes +/// +/// For simplicity, we use the same generic struct `UniformInt` for all +/// integer types `X`. This gives us only one field type, `X`; to store unsigned +/// values of this size, we take use the fact that these conversions are no-ops. +/// +/// For a closed range, the number of possible numbers we should generate is +/// `range = (high - low + 1)`. To avoid bias, we must ensure that the size of +/// our sample space, `zone`, is a multiple of `range`; other values must be +/// rejected (by replacing with a new random sample). +/// +/// As a special case, we use `range = 0` to represent the full range of the +/// result type (i.e. for `new_inclusive($ty::MIN, $ty::MAX)`). +/// +/// The optimum `zone` is the largest product of `range` which fits in our +/// (unsigned) target type. We calculate this by calculating how many numbers we +/// must reject: `reject = (MAX + 1) % range = (MAX - range + 1) % range`. Any (large) +/// product of `range` will suffice, thus in `sample_single` we multiply by a +/// power of 2 via bit-shifting (faster but may cause more rejections). +/// +/// The smallest integer PRNGs generate is `u32`. For 8- and 16-bit outputs we +/// use `u32` for our `zone` and samples (because it's not slower and because +/// it reduces the chance of having to reject a sample). In this case we cannot +/// store `zone` in the target type since it is too large, however we know +/// `ints_to_reject < range <= $uty::MAX`. +/// +/// An alternative to using a modulus is widening multiply: After a widening +/// multiply by `range`, the result is in the high word. Then comparing the low +/// word against `zone` makes sure our distribution is uniform. +#[derive(Clone, Copy, Debug, PartialEq)] +#[cfg_attr(feature = "serde1", derive(Serialize, Deserialize))] +pub struct UniformInt { + low: X, + range: X, + #[cfg(feature = "unbiased")] + thresh: X, // effectively 2.pow(max(64, uty_bits)) % range +} + +macro_rules! uniform_int_impl { + ($ty:ty, $uty:ident, $sample_ty:ident) => { + impl SampleUniform for $ty { + type Sampler = UniformInt<$ty>; + } + + impl UniformSampler for UniformInt<$ty> { + // We play free and fast with unsigned vs signed here + // (when $ty is signed), but that's fine, since the + // contract of this macro is for $ty and $uty to be + // "bit-equal", so casting between them is a no-op. + + type X = $ty; + + #[inline] // if the range is constant, this helps LLVM to do the + // calculations at compile-time. + fn new(low_b: B1, high_b: B2) -> Self + where + B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized, + { + let low = *low_b.borrow(); + let high = *high_b.borrow(); + assert!(low < high, "Uniform::new called with `low >= high`"); + UniformSampler::new_inclusive(low, high - 1) + } + + #[inline] // if the range is constant, this helps LLVM to do the + // calculations at compile-time. + fn new_inclusive(low_b: B1, high_b: B2) -> Self + where + B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized, + { + let low = *low_b.borrow(); + let high = *high_b.borrow(); + assert!( + low <= high, + "Uniform::new_inclusive called with `low > high`" + ); + + let range = high.wrapping_sub(low).wrapping_add(1) as $uty; + #[cfg(feature = "unbiased")] + let thresh = if range > 0 { + let range = $sample_ty::from(range); + (range.wrapping_neg() % range) + } else { + 0 + }; + + UniformInt { + low, + range: range as $ty, // type: $uty + #[cfg(feature = "unbiased")] + thresh: thresh as $uty as $ty, // type: $sample_ty + } + } + + /// Sample from distribution, Canon's method, biased + /// + /// In the worst case, bias affects 1 in `2^n` samples where n is + /// 56 (`i8`), 48 (`i16`), 96 (`i32`), 64 (`i64`), 128 (`i128`). + #[cfg(not(feature = "unbiased"))] + #[inline] + fn sample(&self, rng: &mut R) -> Self::X { + let range = self.range as $uty as $sample_ty; + if range == 0 { + return rng.gen(); + } + + let (mut result, lo) = rng.gen::<$sample_ty>().wmul(range); + + if lo > range.wrapping_neg() { + let (new_hi, _) = (rng.gen::<$sample_ty>()).wmul(range); + let is_overflow = lo.checked_add(new_hi).is_none(); + result += is_overflow as $sample_ty; + } + + self.low.wrapping_add(result as $ty) + } + + /// Sample from distribution, Lemire's method, unbiased + #[cfg(feature = "unbiased")] + #[inline] + fn sample(&self, rng: &mut R) -> Self::X { + let range = self.range as $uty as $sample_ty; + if range == 0 { + return rng.gen(); + } + + let thresh = self.thresh as $uty as $sample_ty; + let hi = loop { + let (hi, lo) = rng.gen::<$sample_ty>().wmul(range); + if lo >= thresh { + break hi; + } + }; + self.low.wrapping_add(hi as $ty) + } + + #[inline] + fn sample_single(low_b: B1, high_b: B2, rng: &mut R) -> Self::X + where + B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized, + { + let low = *low_b.borrow(); + let high = *high_b.borrow(); + assert!(low < high, "UniformSampler::sample_single: low >= high"); + Self::sample_single_inclusive(low, high - 1, rng) + } + + /// Sample single value, Canon's method, biased + /// + /// In the worst case, bias affects 1 in `2^n` samples where n is + /// 56 (`i8`), 48 (`i16`), 96 (`i32`), 64 (`i64`), 128 (`i128`). + #[cfg(not(feature = "unbiased"))] + #[inline] + fn sample_single_inclusive( + low_b: B1, high_b: B2, rng: &mut R, + ) -> Self::X + where + B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized, + { + let low = *low_b.borrow(); + let high = *high_b.borrow(); + assert!( + low <= high, + "UniformSampler::sample_single_inclusive: low > high" + ); + let range = high.wrapping_sub(low).wrapping_add(1) as $uty as $sample_ty; + if range == 0 { + // Range is MAX+1 (unrepresentable), so we need a special case + return rng.gen(); + } + + // generate a sample using a sensible integer type + let (mut result, lo_order) = rng.gen::<$sample_ty>().wmul(range); + + // if the sample is biased... + if lo_order > range.wrapping_neg() { + // ...generate a new sample with 64 more bits, enough that bias is undetectable + let (new_hi_order, _) = (rng.gen::<$sample_ty>()).wmul(range as $sample_ty); + // and adjust if needed + result += + lo_order.checked_add(new_hi_order as $sample_ty).is_none() as $sample_ty; + } + + low.wrapping_add(result as $ty) + } + + /// Sample single value, Canon's method, unbiased + #[cfg(feature = "unbiased")] + #[inline] + fn sample_single_inclusive( + low_b: B1, high_b: B2, rng: &mut R, + ) -> Self::X + where + B1: SampleBorrow<$ty> + Sized, + B2: SampleBorrow<$ty> + Sized, + { + let low = *low_b.borrow(); + let high = *high_b.borrow(); + assert!( + low <= high, + "UniformSampler::sample_single_inclusive: low > high" + ); + let range = high.wrapping_sub(low).wrapping_add(1) as $uty as $sample_ty; + if range == 0 { + // Range is MAX+1 (unrepresentable), so we need a special case + return rng.gen(); + } + + let (mut result, mut lo) = rng.gen::<$sample_ty>().wmul(range); + + while lo > range.wrapping_neg() { + let (new_hi, new_lo) = (rng.gen::<$sample_ty>()).wmul(range); + match lo.checked_add(new_hi) { + Some(x) if x < $sample_ty::MAX => { + // Anything less than MAX: last term is 0 + break; + } + None => { + // Overflow: last term is 1 + result += 1; + break; + } + _ => { + // Unlikely case: must check next sample + lo = new_lo; + continue; + } + } + } + + low.wrapping_add(result as $ty) + } + } + }; +} +uniform_int_impl! { i8, u8, u32 } +uniform_int_impl! { i16, u16, u32 } +uniform_int_impl! { i32, u32, u64 } +uniform_int_impl! { i64, u64, u64 } +uniform_int_impl! { i128, u128, u128 } +uniform_int_impl! { u8, u8, u32 } +uniform_int_impl! { u16, u16, u32} +uniform_int_impl! { u32, u32, u64 } +uniform_int_impl! { u64, u64, u64 } +uniform_int_impl! { u128, u128, u128 } +mod isize_int_impls { + use super::*; + uniform_int_impl! { isize, usize, usize } + uniform_int_impl! { usize, usize, usize } +} + +#[cfg(test)] +mod tests { + use super::*; + use crate::distributions::uniform::{tests::test_samples, Uniform}; + + #[test] + #[cfg(feature = "serde1")] + fn test_uniform_serialization() { + let unit_box: Uniform = Uniform::new(-1, 1); + let de_unit_box: Uniform = + bincode::deserialize(&bincode::serialize(&unit_box).unwrap()).unwrap(); + + assert_eq!(unit_box.0.low, de_unit_box.0.low); + assert_eq!(unit_box.0.range, de_unit_box.0.range); + assert_eq!(unit_box.0.z, de_unit_box.0.z); + } + + #[should_panic] + #[test] + fn test_uniform_bad_limits_equal_int() { + Uniform::new(10, 10); + } + + #[test] + fn test_uniform_good_limits_equal_int() { + let mut rng = crate::test::rng(804); + let dist = Uniform::new_inclusive(10, 10); + for _ in 0..20 { + assert_eq!(rng.sample(dist), 10); + } + } + + #[should_panic] + #[test] + fn test_uniform_bad_limits_flipped_int() { + Uniform::new(10, 5); + } + + #[test] + #[cfg_attr(miri, ignore)] // Miri is too slow + fn test_integers() { + use core::{i128, u128}; + use core::{i16, i32, i64, i8, isize}; + use core::{u16, u32, u64, u8, usize}; + + let mut rng = crate::test::rng(251); + macro_rules! t { + ($ty:ident, $v:expr, $le:expr, $lt:expr) => {{ + for &(low, high) in $v.iter() { + let my_uniform = Uniform::new(low, high); + for _ in 0..1000 { + let v: $ty = rng.sample(my_uniform); + assert!($le(low, v) && $lt(v, high)); + } + + let my_uniform = Uniform::new_inclusive(low, high); + for _ in 0..1000 { + let v: $ty = rng.sample(my_uniform); + assert!($le(low, v) && $le(v, high)); + } + + let my_uniform = Uniform::new(&low, high); + for _ in 0..1000 { + let v: $ty = rng.sample(my_uniform); + assert!($le(low, v) && $lt(v, high)); + } + + let my_uniform = Uniform::new_inclusive(&low, &high); + for _ in 0..1000 { + let v: $ty = rng.sample(my_uniform); + assert!($le(low, v) && $le(v, high)); + } + + for _ in 0..1000 { + let v = <$ty as SampleUniform>::Sampler::sample_single(low, high, &mut rng); + assert!($le(low, v) && $lt(v, high)); + } + + for _ in 0..1000 { + let v = <$ty as SampleUniform>::Sampler::sample_single_inclusive(low, high, &mut rng); + assert!($le(low, v) && $le(v, high)); + } + } + }}; + + // scalar bulk + ($($ty:ident),*) => {{ + $(t!( + $ty, + [(0, 10), (10, 127), ($ty::MIN, $ty::MAX)], + |x, y| x <= y, + |x, y| x < y + );)* + }}; + } + t!(i8, i16, i32, i64, isize, u8, u16, u32, u64, usize, i128, u128); + } + + #[test] + fn test_uniform_from_std_range() { + let r = Uniform::from(2u32..7); + assert_eq!(r.0.low, 2); + assert_eq!(r.0.range, 5); + } + + #[test] + fn test_uniform_from_std_range_inclusive() { + let r = Uniform::from(2u32..=6); + assert_eq!(r.0.low, 2); + assert_eq!(r.0.range, 5); + } + + #[test] + fn value_stability() { + // We test on a sub-set of types; possibly we should do more. + // TODO: SIMD types + + test_samples(11u8, 219, &[17, 66, 214], &[181, 93, 165]); + test_samples(11u32, 219, &[17, 66, 214], &[181, 93, 165]); + + test_samples(0f32, 1e-2f32, &[0.0003070104, 0.0026630748, 0.00979833], &[ + 0.008194133, + 0.00398172, + 0.007428536, + ]); + test_samples( + -1e10f64, + 1e10f64, + &[-4673848682.871551, 6388267422.932352, 4857075081.198343], + &[1173375212.1808167, 1917642852.109581, 2365076174.3153973], + ); + } + + #[test] + fn uniform_distributions_can_be_compared() { + assert_eq!(Uniform::new(1u32, 2u32), Uniform::new(1u32, 2u32)); + } +} diff --git a/src/distributions/uniform/uniform_other.rs b/src/distributions/uniform/uniform_other.rs new file mode 100644 index 00000000000..f8316c900b9 --- /dev/null +++ b/src/distributions/uniform/uniform_other.rs @@ -0,0 +1,340 @@ +// Copyright 2018-2021 Developers of the Rand project. +// +// Licensed under the Apache License, Version 2.0 or the MIT license +// , at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +use super::{SampleBorrow, SampleUniform, Uniform, UniformInt, UniformSampler}; +use crate::distributions::Distribution; +use crate::Rng; +use core::time::Duration; +#[cfg(feature = "serde1")] use serde::{Deserialize, Serialize}; + +impl SampleUniform for char { + type Sampler = UniformChar; +} + +/// The back-end implementing [`UniformSampler`] for `char`. +/// +/// Unless you are implementing [`UniformSampler`] for your own type, this type +/// should not be used directly, use [`Uniform`] instead. +/// +/// This differs from integer range sampling since the range `0xD800..=0xDFFF` +/// are used for surrogate pairs in UCS and UTF-16, and consequently are not +/// valid Unicode code points. We must therefore avoid sampling values in this +/// range. +#[derive(Clone, Copy, Debug, PartialEq)] +#[cfg_attr(feature = "serde1", derive(Serialize, Deserialize))] +pub struct UniformChar { + sampler: UniformInt, +} + +/// UTF-16 surrogate range start +const CHAR_SURROGATE_START: u32 = 0xD800; +/// UTF-16 surrogate range size +const CHAR_SURROGATE_LEN: u32 = 0xE000 - CHAR_SURROGATE_START; + +/// Convert `char` to compressed `u32` +fn char_to_comp_u32(c: char) -> u32 { + match c as u32 { + c if c >= CHAR_SURROGATE_START => c - CHAR_SURROGATE_LEN, + c => c, + } +} + +impl UniformSampler for UniformChar { + type X = char; + + #[inline] // if the range is constant, this helps LLVM to do the + // calculations at compile-time. + fn new(low_b: B1, high_b: B2) -> Self + where + B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized, + { + let low = char_to_comp_u32(*low_b.borrow()); + let high = char_to_comp_u32(*high_b.borrow()); + let sampler = UniformInt::::new(low, high); + UniformChar { sampler } + } + + #[inline] // if the range is constant, this helps LLVM to do the + // calculations at compile-time. + fn new_inclusive(low_b: B1, high_b: B2) -> Self + where + B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized, + { + let low = char_to_comp_u32(*low_b.borrow()); + let high = char_to_comp_u32(*high_b.borrow()); + let sampler = UniformInt::::new_inclusive(low, high); + UniformChar { sampler } + } + + fn sample(&self, rng: &mut R) -> Self::X { + let mut x = self.sampler.sample(rng); + if x >= CHAR_SURROGATE_START { + x += CHAR_SURROGATE_LEN; + } + // SAFETY: x must not be in surrogate range or greater than char::MAX. + // This relies on range constructors which accept char arguments. + // Validity of input char values is assumed. + unsafe { core::char::from_u32_unchecked(x) } + } +} + +/// The back-end implementing [`UniformSampler`] for `Duration`. +/// +/// Unless you are implementing [`UniformSampler`] for your own types, this type +/// should not be used directly, use [`Uniform`] instead. +#[derive(Clone, Copy, Debug, PartialEq)] +#[cfg_attr(feature = "serde1", derive(Serialize, Deserialize))] +pub struct UniformDuration { + mode: UniformDurationMode, + offset: u32, +} + +#[derive(Debug, Copy, Clone, PartialEq)] +#[cfg_attr(feature = "serde1", derive(Serialize, Deserialize))] +enum UniformDurationMode { + Small { + secs: u64, + nanos: Uniform, + }, + Medium { + nanos: Uniform, + }, + Large { + max_secs: u64, + max_nanos: u32, + secs: Uniform, + }, +} + +impl SampleUniform for Duration { + type Sampler = UniformDuration; +} + +impl UniformSampler for UniformDuration { + type X = Duration; + + #[inline] + fn new(low_b: B1, high_b: B2) -> Self + where + B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized, + { + let low = *low_b.borrow(); + let high = *high_b.borrow(); + assert!(low < high, "Uniform::new called with `low >= high`"); + UniformDuration::new_inclusive(low, high - Duration::new(0, 1)) + } + + #[inline] + fn new_inclusive(low_b: B1, high_b: B2) -> Self + where + B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized, + { + let low = *low_b.borrow(); + let high = *high_b.borrow(); + assert!( + low <= high, + "Uniform::new_inclusive called with `low > high`" + ); + + let low_s = low.as_secs(); + let low_n = low.subsec_nanos(); + let mut high_s = high.as_secs(); + let mut high_n = high.subsec_nanos(); + + if high_n < low_n { + high_s -= 1; + high_n += 1_000_000_000; + } + + let mode = if low_s == high_s { + UniformDurationMode::Small { + secs: low_s, + nanos: Uniform::new_inclusive(low_n, high_n), + } + } else { + let max = high_s + .checked_mul(1_000_000_000) + .and_then(|n| n.checked_add(u64::from(high_n))); + + if let Some(higher_bound) = max { + let lower_bound = low_s * 1_000_000_000 + u64::from(low_n); + UniformDurationMode::Medium { + nanos: Uniform::new_inclusive(lower_bound, higher_bound), + } + } else { + // An offset is applied to simplify generation of nanoseconds + let max_nanos = high_n - low_n; + UniformDurationMode::Large { + max_secs: high_s, + max_nanos, + secs: Uniform::new_inclusive(low_s, high_s), + } + } + }; + UniformDuration { + mode, + offset: low_n, + } + } + + #[inline] + fn sample(&self, rng: &mut R) -> Duration { + match self.mode { + UniformDurationMode::Small { secs, nanos } => { + let n = nanos.sample(rng); + Duration::new(secs, n) + } + UniformDurationMode::Medium { nanos } => { + let nanos = nanos.sample(rng); + Duration::new(nanos / 1_000_000_000, (nanos % 1_000_000_000) as u32) + } + UniformDurationMode::Large { + max_secs, + max_nanos, + secs, + } => { + // constant folding means this is at least as fast as `Rng::sample(Range)` + let nano_range = Uniform::new(0, 1_000_000_000); + loop { + let s = secs.sample(rng); + let n = nano_range.sample(rng); + if !(s == max_secs && n > max_nanos) { + let sum = n + self.offset; + break Duration::new(s, sum); + } + } + } + } + } +} + +#[cfg(test)] +mod tests { + use super::*; + use crate::distributions::uniform::tests::test_samples; + + #[test] + #[cfg(feature = "serde1")] + fn test_serialization_uniform_duration() { + let distr = UniformDuration::new(Duration::from_secs(10), Duration::from_secs(60)); + let de_distr: UniformDuration = + bincode::deserialize(&bincode::serialize(&distr).unwrap()).unwrap(); + assert_eq!(distr.offset, de_distr.offset); + match (distr.mode, de_distr.mode) { + ( + UniformDurationMode::Small { + secs: a_secs, + nanos: a_nanos, + }, + UniformDurationMode::Small { secs, nanos }, + ) => { + assert_eq!(a_secs, secs); + assert_eq!(a_nanos, nanos); + } + ( + UniformDurationMode::Medium { nanos: a_nanos }, + UniformDurationMode::Medium { nanos }, + ) => { + assert_eq!(a_nanos, nanos); + } + ( + UniformDurationMode::Large { + max_secs: a_max_secs, + max_nanos: a_max_nanos, + secs: a_secs, + }, + UniformDurationMode::Large { + max_secs, + max_nanos, + secs, + }, + ) => { + assert_eq!(a_max_secs, max_secs); + assert_eq!(a_max_nanos, max_nanos); + assert_eq!(a_secs, secs); + } + _ => panic!("`UniformDurationMode` was not serialized/deserialized correctly"), + } + } + + #[test] + #[cfg_attr(miri, ignore)] // Miri is too slow + fn test_char() { + let mut rng = crate::test::rng(891); + let mut max = core::char::from_u32(0).unwrap(); + for _ in 0..100 { + let c = rng.gen_range('A'..='Z'); + assert!(('A'..='Z').contains(&c)); + max = max.max(c); + } + assert_eq!(max, 'Z'); + let d = Uniform::new( + core::char::from_u32(0xD7F0).unwrap(), + core::char::from_u32(0xE010).unwrap(), + ); + for _ in 0..100 { + let c = d.sample(&mut rng); + assert!((c as u32) < 0xD800 || (c as u32) > 0xDFFF); + } + } + + #[test] + #[cfg_attr(miri, ignore)] // Miri is too slow + fn test_durations() { + let mut rng = crate::test::rng(253); + + let v = &[ + (Duration::new(10, 50000), Duration::new(100, 1234)), + (Duration::new(0, 100), Duration::new(1, 50)), + ( + Duration::new(0, 0), + Duration::new(u64::max_value(), 999_999_999), + ), + ]; + for &(low, high) in v.iter() { + let my_uniform = Uniform::new(low, high); + for _ in 0..1000 { + let v = rng.sample(my_uniform); + assert!(low <= v && v < high); + } + } + } + + #[test] + fn value_stability() { + test_samples( + Duration::new(2, 0), + Duration::new(4, 0), + &[ + Duration::new(2, 532615131), + Duration::new(3, 638826742), + Duration::new(3, 485707508), + ], + &[ + Duration::new(3, 117337521), + Duration::new(3, 191764285), + Duration::new(3, 236507617), + ], + ); + + test_samples('a', 'z', &['a', 'g', 'y'], &['u', 'j', 's']); + } + + #[test] + fn uniform_distributions_can_be_compared() { + assert_eq!(Uniform::new('a', 'g'), Uniform::new('a', 'g')); + assert_eq!( + Uniform::new(Duration::from_millis(10), Duration::from_millis(20)), + Uniform::new(Duration::from_millis(10), Duration::from_millis(20)), + ); + } +} diff --git a/src/distributions/uniform/uniform_simd.rs b/src/distributions/uniform/uniform_simd.rs new file mode 100644 index 00000000000..99ad1b52d76 --- /dev/null +++ b/src/distributions/uniform/uniform_simd.rs @@ -0,0 +1,230 @@ +// Copyright 2018-2021 Developers of the Rand project. +// +// Licensed under the Apache License, Version 2.0 or the MIT license +// , at your +// option. This file may not be copied, modified, or distributed +// except according to those terms. + +use super::{SampleBorrow, SampleUniform, UniformInt, UniformSampler}; +use crate::distributions::utils::WideningMultiply; +use crate::Rng; +use packed_simd::*; + +macro_rules! uniform_simd_int_impl { + ($ty:ident, $unsigned:ident, $u_scalar:ident) => { + // The "pick the largest zone that can fit in an `u32`" optimization + // is less useful here. Multiple lanes complicate things, we don't + // know the PRNG's minimal output size, and casting to a larger vector + // is generally a bad idea for SIMD performance. The user can still + // implement it manually. + + // TODO: look into `Uniform::::new(0u32, 100)` functionality + // perhaps `impl SampleUniform for $u_scalar`? + impl SampleUniform for $ty { + type Sampler = UniformInt<$ty>; + } + + impl UniformSampler for UniformInt<$ty> { + type X = $ty; + + #[inline] // if the range is constant, this helps LLVM to do the + // calculations at compile-time. + fn new(low_b: B1, high_b: B2) -> Self + where B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized + { + let low = *low_b.borrow(); + let high = *high_b.borrow(); + assert!(low.lt(high).all(), "Uniform::new called with `low >= high`"); + UniformSampler::new_inclusive(low, high - 1) + } + + #[inline] // if the range is constant, this helps LLVM to do the + // calculations at compile-time. + fn new_inclusive(low_b: B1, high_b: B2) -> Self + where B1: SampleBorrow + Sized, + B2: SampleBorrow + Sized + { + let low = *low_b.borrow(); + let high = *high_b.borrow(); + assert!(low.le(high).all(), + "Uniform::new_inclusive called with `low > high`"); + let unsigned_max = ::core::$u_scalar::MAX; + + // NOTE: these may need to be replaced with explicitly + // wrapping operations if `packed_simd` changes + let range: $unsigned = ((high - low) + 1).cast(); + // `% 0` will panic at runtime. + let not_full_range = range.gt($unsigned::splat(0)); + // replacing 0 with `unsigned_max` allows a faster `select` + // with bitwise OR + let modulo = not_full_range.select(range, $unsigned::splat(unsigned_max)); + // wrapping addition + let ints_to_reject = (unsigned_max - range + 1) % modulo; + // When `range` is 0, `lo` of `v.wmul(range)` will always be + // zero which means only one sample is needed. + let zone = unsigned_max - ints_to_reject; + + UniformInt { + low, + // These are really $unsigned values, but store as $ty: + range: range.cast(), + z: zone.cast(), + nrmr: ((0 - range) % range).cast(), + } + } + + fn sample(&self, rng: &mut R) -> Self::X { + let range: $unsigned = self.range.cast(); + let zone: $unsigned = self.z.cast(); + + // This might seem very slow, generating a whole new + // SIMD vector for every sample rejection. For most uses + // though, the chance of rejection is small and provides good + // general performance. With multiple lanes, that chance is + // multiplied. To mitigate this, we replace only the lanes of + // the vector which fail, iteratively reducing the chance of + // rejection. The replacement method does however add a little + // overhead. Benchmarking or calculating probabilities might + // reveal contexts where this replacement method is slower. + let mut v: $unsigned = rng.gen(); + loop { + let (hi, lo) = v.wmul(range); + let mask = lo.le(zone); + if mask.all() { + let hi: $ty = hi.cast(); + // wrapping addition + let result = self.low + hi; + // `select` here compiles to a blend operation + // When `range.eq(0).none()` the compare and blend + // operations are avoided. + let v: $ty = v.cast(); + return range.gt($unsigned::splat(0)).select(result, v); + } + // Replace only the failing lanes + v = mask.select(v, rng.gen()); + } + } + } + }; + + // bulk implementation + ($(($unsigned:ident, $signed:ident),)+ $u_scalar:ident) => { + $( + uniform_simd_int_impl!($unsigned, $unsigned, $u_scalar); + uniform_simd_int_impl!($signed, $unsigned, $u_scalar); + )+ + }; +} + +uniform_simd_int_impl! { + (u64x2, i64x2), + (u64x4, i64x4), + (u64x8, i64x8), + u64 +} + +uniform_simd_int_impl! { + (u32x2, i32x2), + (u32x4, i32x4), + (u32x8, i32x8), + (u32x16, i32x16), + u32 +} + +uniform_simd_int_impl! { + (u16x2, i16x2), + (u16x4, i16x4), + (u16x8, i16x8), + (u16x16, i16x16), + (u16x32, i16x32), + u16 +} + +uniform_simd_int_impl! { + (u8x2, i8x2), + (u8x4, i8x4), + (u8x8, i8x8), + (u8x16, i8x16), + (u8x32, i8x32), + (u8x64, i8x64), + u8 +} + +#[cfg(test)] +mod tests { + use super::*; + use crate::distributions::uniform::Uniform; + + #[test] + #[cfg_attr(miri, ignore)] // Miri is too slow + fn test_integers() { + use core::{i16, i32, i64, i8}; + use core::{u16, u32, u64, u8}; + + let mut rng = crate::test::rng(251); + macro_rules! t { + ($ty:ident, $v:expr, $le:expr, $lt:expr) => {{ + for &(low, high) in $v.iter() { + let my_uniform = Uniform::new(low, high); + for _ in 0..1000 { + let v: $ty = rng.sample(my_uniform); + assert!($le(low, v) && $lt(v, high)); + } + + let my_uniform = Uniform::new_inclusive(low, high); + for _ in 0..1000 { + let v: $ty = rng.sample(my_uniform); + assert!($le(low, v) && $le(v, high)); + } + + let my_uniform = Uniform::new(&low, high); + for _ in 0..1000 { + let v: $ty = rng.sample(my_uniform); + assert!($le(low, v) && $lt(v, high)); + } + + let my_uniform = Uniform::new_inclusive(&low, &high); + for _ in 0..1000 { + let v: $ty = rng.sample(my_uniform); + assert!($le(low, v) && $le(v, high)); + } + + for _ in 0..1000 { + let v = <$ty as SampleUniform>::Sampler::sample_single(low, high, &mut rng); + assert!($le(low, v) && $lt(v, high)); + } + + for _ in 0..1000 { + let v = <$ty as SampleUniform>::Sampler::sample_single_inclusive(low, high, &mut rng); + assert!($le(low, v) && $le(v, high)); + } + } + }}; + + // simd bulk + ($($ty:ident),* => $scalar:ident) => {{ + $(t!( + $ty, + [ + ($ty::splat(0), $ty::splat(10)), + ($ty::splat(10), $ty::splat(127)), + ($ty::splat($scalar::MIN), $ty::splat($scalar::MAX)), + ], + |x: $ty, y| x.le(y).all(), + |x: $ty, y| x.lt(y).all() + );)* + }}; + } + + t!(u8x2, u8x4, u8x8, u8x16, u8x32, u8x64 => u8); + t!(i8x2, i8x4, i8x8, i8x16, i8x32, i8x64 => i8); + t!(u16x2, u16x4, u16x8, u16x16, u16x32 => u16); + t!(i16x2, i16x4, i16x8, i16x16, i16x32 => i16); + t!(u32x2, u32x4, u32x8, u32x16 => u32); + t!(i32x2, i32x4, i32x8, i32x16 => i32); + t!(u64x2, u64x4, u64x8 => u64); + t!(i64x2, i64x4, i64x8 => i64); + } +}