1 /* Copyright (C) 2010-2020 The RetroArch team
3 * ---------------------------------------------------------------------------------------
4 * The following license statement only applies to this file (s16_to_float_neon.S).
5 * ---------------------------------------------------------------------------------------
7 * Permission is hereby granted, free of charge,
8 * to any person obtaining a copy of this software and associated documentation files (the "Software"),
9 * to deal in the Software without restriction, including without limitation the rights to
10 * use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software,
11 * and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
13 * The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
15 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED,
16 * INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
18 * IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,
19 * WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20 * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
22 #if defined(__ARM_NEON__) && defined(HAVE_ARM_NEON_ASM_OPTIMIZATIONS)
24 #if defined(__thumb__)
25 #define DECL_ARMMODE(x) " .align 2\n" " .global " x "\n" " .thumb\n" " .thumb_func\n" " .type " x ", %function\n" x ":\n"
27 #define DECL_ARMMODE(x) " .align 4\n" " .global " x "\n" " .arm\n" x ":\n"
31 DECL_ARMMODE("convert_s16_float_asm")
32 DECL_ARMMODE("_convert_s16_float_asm")
33 "# convert_s16_float_asm(float *out, const int16_t *in, size_t samples, const float *gain)\n"
34 " # Hacky way to get a constant of 2^-15.\n"
35 " # Might be faster to just load a constant from memory.\n"
36 " # It's just done once however ...\n"
37 " vmov.f32 q8, #0.25\n"
38 " vmul.f32 q8, q8, q8\n"
39 " vmul.f32 q8, q8, q8\n"
40 " vmul.f32 q8, q8, q8\n"
41 " vadd.f32 q8, q8, q8\n"
44 " vld1.f32 {d6[0]}, [r3]\n"
45 " vmul.f32 q8, q8, d6[0]\n"
49 " vld1.s16 {q0}, [r1]!\n"
51 " # Widen to 32-bit\n"
55 " # Convert to float\n"
56 " vcvt.f32.s32 q1, q1\n"
57 " vcvt.f32.s32 q2, q2\n"
59 " vmul.f32 q1, q1, q8\n"
60 " vmul.f32 q2, q2, q8\n"
62 " vst1.f32 {q1-q2}, [r0]!\n"
64 " # Guaranteed to get samples in multiples of 8.\n"