1 /* SPDX-License-Identifier: MIT
2 *
3 * Permission is hereby granted, free of charge, to any person
4 * obtaining a copy of this software and associated documentation
5 * files (the "Software"), to deal in the Software without
6 * restriction, including without limitation the rights to use, copy,
7 * modify, merge, publish, distribute, sublicense, and/or sell copies
8 * of the Software, and to permit persons to whom the Software is
9 * furnished to do so, subject to the following conditions:
10 *
11 * The above copyright notice and this permission notice shall be
12 * included in all copies or substantial portions of the Software.
13 *
14 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
15 * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
16 * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
17 * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
18 * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
19 * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
20 * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21 * SOFTWARE.
22 *
23 * Copyright:
24 * 2020 Evan Nemerson <evan@nemerson.com>
25 * 2020 Sean Maher <seanptmaher@gmail.com> (Copyright owned by Google, LLC)
26 */
27
28 #if !defined(SIMDE_ARM_NEON_QMOVUN_H)
29 #define SIMDE_ARM_NEON_QMOVUN_H
30
31 #include "types.h"
32 #include "dup_n.h"
33 #include "min.h"
34 #include "max.h"
35 #include "movn.h"
36
37 HEDLEY_DIAGNOSTIC_PUSH
38 SIMDE_DISABLE_UNWANTED_DIAGNOSTICS
39 SIMDE_BEGIN_DECLS_
40
41 SIMDE_FUNCTION_ATTRIBUTES
42 uint8_t
simde_vqmovunh_s16(int16_t a)43 simde_vqmovunh_s16(int16_t a) {
44 #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
45 return HEDLEY_STATIC_CAST(uint8_t, vqmovunh_s16(a));
46 #else
47 return (a > UINT8_MAX) ? UINT8_MAX : ((a < 0) ? 0 : HEDLEY_STATIC_CAST(uint8_t, a));
48 #endif
49 }
50 #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
51 #undef vqmovunh_s16
52 #define vqmovunh_s16(a) simde_vqmovunh_s16((a))
53 #endif
54
55 SIMDE_FUNCTION_ATTRIBUTES
56 uint16_t
simde_vqmovuns_s32(int32_t a)57 simde_vqmovuns_s32(int32_t a) {
58 #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
59 return HEDLEY_STATIC_CAST(uint16_t, vqmovuns_s32(a));
60 #else
61 return (a > UINT16_MAX) ? UINT16_MAX : ((a < 0) ? 0 : HEDLEY_STATIC_CAST(uint16_t, a));
62 #endif
63 }
64 #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
65 #undef vqmovuns_s32
66 #define vqmovuns_s32(a) simde_vqmovuns_s32((a))
67 #endif
68
69 SIMDE_FUNCTION_ATTRIBUTES
70 uint32_t
simde_vqmovund_s64(int64_t a)71 simde_vqmovund_s64(int64_t a) {
72 #if defined(SIMDE_ARM_NEON_A64V8_NATIVE)
73 return HEDLEY_STATIC_CAST(uint32_t, vqmovund_s64(a));
74 #else
75 return (a > UINT32_MAX) ? UINT32_MAX : ((a < 0) ? 0 : HEDLEY_STATIC_CAST(uint32_t, a));
76 #endif
77 }
78 #if defined(SIMDE_ARM_NEON_A64V8_ENABLE_NATIVE_ALIASES)
79 #undef vqmovund_s64
80 #define vqmovund_s64(a) simde_vqmovund_s64((a))
81 #endif
82
83 SIMDE_FUNCTION_ATTRIBUTES
84 simde_uint8x8_t
simde_vqmovun_s16(simde_int16x8_t a)85 simde_vqmovun_s16(simde_int16x8_t a) {
86 #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
87 return vqmovun_s16(a);
88 #elif SIMDE_NATURAL_VECTOR_SIZE > 0
89 return simde_vmovn_u16(simde_vreinterpretq_u16_s16(simde_vmaxq_s16(simde_vdupq_n_s16(0), simde_vminq_s16(simde_vdupq_n_s16(UINT8_MAX), a))));
90 #else
91 simde_uint8x8_private r_;
92 simde_int16x8_private a_ = simde_int16x8_to_private(a);
93
94 SIMDE_VECTORIZE
95 for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
96 r_.values[i] = simde_vqmovunh_s16(a_.values[i]);
97 }
98
99 return simde_uint8x8_from_private(r_);
100 #endif
101 }
102 #if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
103 #undef vqmovun_s16
104 #define vqmovun_s16(a) simde_vqmovun_s16((a))
105 #endif
106
107 SIMDE_FUNCTION_ATTRIBUTES
108 simde_uint16x4_t
simde_vqmovun_s32(simde_int32x4_t a)109 simde_vqmovun_s32(simde_int32x4_t a) {
110 #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
111 return vqmovun_s32(a);
112 #elif SIMDE_NATURAL_VECTOR_SIZE > 0
113 return simde_vmovn_u32(simde_vreinterpretq_u32_s32(simde_vmaxq_s32(simde_vdupq_n_s32(0), simde_vminq_s32(simde_vdupq_n_s32(UINT16_MAX), a))));
114 #else
115 simde_uint16x4_private r_;
116 simde_int32x4_private a_ = simde_int32x4_to_private(a);
117
118 SIMDE_VECTORIZE
119 for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
120 r_.values[i] = simde_vqmovuns_s32(a_.values[i]);
121 }
122
123 return simde_uint16x4_from_private(r_);
124 #endif
125 }
126 #if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
127 #undef vqmovun_s32
128 #define vqmovun_s32(a) simde_vqmovun_s32((a))
129 #endif
130
131 SIMDE_FUNCTION_ATTRIBUTES
132 simde_uint32x2_t
simde_vqmovun_s64(simde_int64x2_t a)133 simde_vqmovun_s64(simde_int64x2_t a) {
134 #if defined(SIMDE_ARM_NEON_A32V7_NATIVE)
135 return vqmovun_s64(a);
136 #elif SIMDE_NATURAL_VECTOR_SIZE > 0
137 return simde_vmovn_u64(simde_vreinterpretq_u64_s64(simde_x_vmaxq_s64(simde_vdupq_n_s64(0), simde_x_vminq_s64(simde_vdupq_n_s64(UINT32_MAX), a))));
138 #else
139 simde_uint32x2_private r_;
140 simde_int64x2_private a_ = simde_int64x2_to_private(a);
141
142 SIMDE_VECTORIZE
143 for (size_t i = 0 ; i < (sizeof(r_.values) / sizeof(r_.values[0])) ; i++) {
144 r_.values[i] = simde_vqmovund_s64(a_.values[i]);
145 }
146
147 return simde_uint32x2_from_private(r_);
148 #endif
149 }
150 #if defined(SIMDE_ARM_NEON_A32V7_ENABLE_NATIVE_ALIASES)
151 #undef vqmovun_s64
152 #define vqmovun_s64(a) simde_vqmovun_s64((a))
153 #endif
154
155
156 SIMDE_END_DECLS_
157 HEDLEY_DIAGNOSTIC_POP
158
159 #endif /* !defined(SIMDE_ARM_NEON_QMOVUN_H) */
160