simdeez 3.0.1

SIMD library to abstract over different instruction sets and widths
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317

use crate::{engines, Simd};

#[cfg(target_arch = "aarch64")]
use std::arch::is_aarch64_feature_detected;

#[macro_export]
macro_rules! fix_tuple_type {
    (()) => {
        ()
    };
    (($typ:ty)) => {
        ($typ,)
    };
    (($($typ:ty),*)) => {
        (($($typ),*))
    };
}

#[macro_export]
macro_rules! __simd_generate_base {
    ($(#[$meta:meta])* $vis:vis fn $fn_name:ident $(<$($lt:lifetime),+>)? ($($arg:ident:$typ:ty),* ) -> $rt:ty $body:block  ) => {
        simdeez_paste_item! {
            // In order to pass arguments via generics like this, we need to convert the arguments
            // into tuples. This is part of the reason for the mess below.

            #[inline(always)]
            $vis unsafe fn [<__ $fn_name _generic>]<$($($lt,)+)? S: 'static + Simd>(args_tuple: ($($typ,)*)) -> $rt {
                let ($($arg,)*) = args_tuple;
                S::invoke(#[inline(always)] || $body)
            }

            $(#[$meta])*
            #[inline(always)]
            $vis fn [<$fn_name _generic>] <$($($lt),+,)? S: Simd>($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_generic::<S, [<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            #[allow(non_camel_case_types)]
            struct [<__ $fn_name _dispatch_struct>];

            impl$(<$($lt),+>)? __SimdRunner<fix_tuple_type!(($($typ),*)), $rt> for [<__ $fn_name _dispatch_struct>] {
                unsafe fn run<S: Simd>(args_tuple: fix_tuple_type!(($($typ),*))) -> $rt {
                    [<__ $fn_name _generic>]::<S>(args_tuple)
                }
            }
        }
    };
}

#[macro_export]
macro_rules! simd_runtime_generate {
    ($(#[$meta:meta])* $vis:vis fn $fn_name:ident $(<$($lt:lifetime),+>)? ($($arg:ident:$typ:ty),* $(,)? ) -> $rt:ty $body:block  ) => {
        simdeez_paste_item! {
            // In order to pass arguments via generics like this, we need to convert the arguments
            // into tuples. This is part of the reason for the mess below.

            $(#[$meta])*
            #[inline(always)]
            $vis fn $fn_name $(<$($lt),+>)?($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_runtime_decide::<[<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            $(#[$meta])*
            #[inline(always)]
            $vis fn [<$fn_name _scalar>] $(<$($lt),+>)?($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_invoke_scalar::<[<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            __simd_generate_base!($(#[$meta])* $vis fn $fn_name $(<$($lt),+>)? ($($arg:$typ),* ) -> $rt $body);
        }
    };
    ($(#[$meta:meta])* $vis:vis fn $fn_name:ident ($($arg:ident:$typ:ty),* $(,)? ) $body:block  ) => {
        simd_runtime_generate!($(#[$meta])* $vis fn $fn_name ($($arg:$typ),*) -> () $body);
    };
}

#[macro_export]
macro_rules! simd_compiletime_select {
    ($(#[$meta:meta])* $vis:vis fn $fn_name:ident $(<$($lt:lifetime),+>)? ($($arg:ident:$typ:ty),* $(,)? ) -> $rt:ty $body:block  ) => {
        simdeez_paste_item! {
            $(#[$meta])*
            #[inline(always)]
            $vis fn $fn_name $(<$($lt),+>)?($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_compiletime_select::<[<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            $(#[$meta])*
            #[inline(always)]
            $vis fn [<$fn_name _scalar>] $(<$($lt),+>)?($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_invoke_scalar::<[<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            __simd_generate_base!($(#[$meta])* $vis fn $fn_name $(<$($lt),+>)? ($($arg:$typ),* ) -> $rt $body);
        }
    };
    ($(#[$meta:meta])* $vis:vis fn $fn_name:ident ($($arg:ident:$typ:ty),* $(,)? ) $body:block  ) => {
        simd_compiletime_select!($(#[$meta])* $vis fn $fn_name ($($arg:$typ),*) -> () $body);
    };
}

#[macro_export]
macro_rules! simd_unsafe_generate_all {
    ($(#[$meta:meta])* $vis:vis fn $fn_name:ident $(<$($lt:lifetime),+>)? ($($arg:ident:$typ:ty),* $(,)? ) -> $rt:ty $body:block  ) => {
        simdeez_paste_item! {
            $(#[$meta])*
            #[inline(always)]
            $vis fn $fn_name $(<$($lt),+>)?($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_runtime_decide::<[<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            $(#[$meta])*
            #[inline(always)]
            #[cfg(any(target_arch = "x86_64", target_arch = "x86"))]
            $vis fn [<$fn_name _scalar>] $(<$($lt),+>)?($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_invoke_scalar::<[<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            $(#[$meta])*
            #[inline(always)]
            #[cfg(any(target_arch = "x86_64", target_arch = "x86"))]
            $vis unsafe fn [<$fn_name _sse2>] $(<$($lt),+>)?($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_invoke_sse2::<[<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            $(#[$meta])*
            #[inline(always)]
            #[cfg(any(target_arch = "x86_64", target_arch = "x86"))]
            $vis unsafe fn [<$fn_name _sse41>] $(<$($lt),+>)?($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_invoke_sse41::<[<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            $(#[$meta])*
            #[inline(always)]
            #[cfg(any(target_arch = "x86_64", target_arch = "x86"))]
            $vis unsafe fn [<$fn_name _avx2>] $(<$($lt),+>)?($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_invoke_avx2::<[<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            $(#[$meta])*
            #[inline(always)]
            #[cfg(any(target_arch = "x86_64", target_arch = "x86"))]
            $vis unsafe fn [<$fn_name _avx512>] $(<$($lt),+>)?($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_invoke_avx512::<[<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            $(#[$meta])*
            #[inline(always)]
            #[cfg(target_arch = "aarch64")]
            $vis unsafe fn [<$fn_name _neon>] $(<$($lt),+>)?($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_invoke_neon::<[<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            $(#[$meta])*
            #[inline(always)]
            #[cfg(target_arch = "wasm32")]
            $vis unsafe fn [<$fn_name _wasm>] $(<$($lt),+>)?($($arg:$typ,)*) -> $rt {
                let args_tuple = ($($arg,)*);
                __run_simd_invoke_wasm::<[<__ $fn_name _dispatch_struct>], fix_tuple_type!(($($typ),*)), $rt>(args_tuple)
            }

            __simd_generate_base!($(#[$meta])* $vis fn $fn_name $(<$($lt),+>)? ($($arg:$typ),* ) -> $rt $body);
        }
    };
    ($(#[$meta:meta])* $vis:vis fn $fn_name:ident ($($arg:ident:$typ:ty),* $(,)? ) $body:block  ) => {
        simd_unsafe_generate_all!($(#[$meta])* $vis fn $fn_name ($($arg:$typ),*) -> () $body);
    };
}

pub trait __SimdRunner<A, R> {
    unsafe fn run<S: Simd>(args: A) -> R;
}

#[inline(always)]
pub fn __run_simd_runtime_decide<S: __SimdRunner<A, R>, A, R>(args: A) -> R {
    #![allow(unreachable_code)]

    #[cfg(any(target_arch = "x86_64", target_arch = "x86"))]
    {
        if is_x86_feature_detected!("avx512f")
            && is_x86_feature_detected!("avx512bw")
            && is_x86_feature_detected!("avx512dq")
        {
            return unsafe { S::run::<engines::avx512::Avx512>(args) };
        }

        if is_x86_feature_detected!("avx2") && is_x86_feature_detected!("fma") {
            return unsafe { S::run::<engines::avx2::Avx2>(args) };
        }

        if is_x86_feature_detected!("sse4.1") {
            return unsafe { S::run::<engines::sse41::Sse41>(args) };
        }

        if is_x86_feature_detected!("sse2") {
            return unsafe { S::run::<engines::sse2::Sse2>(args) };
        }
    }

    #[cfg(target_arch = "aarch64")]
    if is_aarch64_feature_detected!("neon") {
        return unsafe { S::run::<engines::neon::Neon>(args) };
    }

    #[cfg(target_arch = "wasm32")]
    {
        // Note: there's currently no way to detect SIMD support in WebAssembly at runtime
        return unsafe { S::run::<engines::wasm32::Wasm>(args) };
    }

    unsafe { S::run::<engines::scalar::Scalar>(args) }
}

#[inline(always)]
pub fn __run_simd_generic<E: Simd, S: __SimdRunner<A, R>, A, R>(args: A) -> R {
    unsafe { S::run::<E>(args) }
}

#[inline(always)]
pub fn __run_simd_compiletime_select<S: __SimdRunner<A, R>, A, R>(args: A) -> R {
    #![allow(unreachable_code)]
    #![allow(clippy::needless_return)]

    #[cfg(any(target_arch = "x86_64", target_arch = "x86"))]
    {
        #[cfg(all(
            target_feature = "avx512f",
            target_feature = "avx512bw",
            target_feature = "avx512dq"
        ))]
        return unsafe { S::run::<engines::avx512::Avx512>(args) };

        #[cfg(all(target_feature = "avx2", target_feature = "fma"))]
        return unsafe { S::run::<engines::avx2::Avx2>(args) };

        #[cfg(target_feature = "sse4.1")]
        return unsafe { S::run::<engines::sse41::Sse41>(args) };

        #[cfg(target_feature = "sse2")]
        return unsafe { S::run::<engines::sse2::Sse2>(args) };
    }

    #[cfg(target_arch = "aarch64")]
    {
        #[cfg(target_feature = "neon")]
        return unsafe { S::run::<engines::neon::Neon>(args) };
    }

    #[cfg(target_arch = "wasm32")]
    {
        return unsafe { S::run::<engines::wasm32::Wasm>(args) };
    }

    return unsafe { S::run::<engines::scalar::Scalar>(args) };
}

#[inline(always)]
pub fn __run_simd_invoke_scalar<S: __SimdRunner<A, R>, A, R>(args: A) -> R {
    unsafe { S::run::<engines::scalar::Scalar>(args) }
}

#[inline(always)]
#[cfg(any(target_arch = "x86_64", target_arch = "x86"))]
pub unsafe fn __run_simd_invoke_sse2<S: __SimdRunner<A, R>, A, R>(args: A) -> R {
    unsafe { S::run::<engines::sse2::Sse2>(args) }
}

#[inline(always)]
#[cfg(any(target_arch = "x86_64", target_arch = "x86"))]
pub unsafe fn __run_simd_invoke_sse41<S: __SimdRunner<A, R>, A, R>(args: A) -> R {
    unsafe { S::run::<engines::sse41::Sse41>(args) }
}

#[inline(always)]
#[cfg(any(target_arch = "x86_64", target_arch = "x86"))]
pub unsafe fn __run_simd_invoke_avx2<S: __SimdRunner<A, R>, A, R>(args: A) -> R {
    unsafe { S::run::<engines::avx2::Avx2>(args) }
}

#[inline(always)]
#[cfg(any(target_arch = "x86_64", target_arch = "x86"))]
pub unsafe fn __run_simd_invoke_avx512<S: __SimdRunner<A, R>, A, R>(args: A) -> R {
    unsafe { S::run::<engines::avx512::Avx512>(args) }
}

#[inline(always)]
#[cfg(target_feature = "neon")]
pub unsafe fn __run_simd_invoke_neon<S: __SimdRunner<A, R>, A, R>(args: A) -> R {
    unsafe { S::run::<engines::neon::Neon>(args) }
}

#[inline(always)]
#[cfg(target_arch = "wasm32")]
pub unsafe fn __run_simd_invoke_wasm<S: __SimdRunner<A, R>, A, R>(args: A) -> R {
    unsafe { S::run::<engines::wasm32::Wasm>(args) }
}

#[macro_export]
macro_rules! simd_invoke {
    ($g:ident, $($r:tt)+) => {
        $g::invoke(
            #[inline(always)]
            || $($r)+
        )
    }
}