ffmpeg / libavcodec / arm / ac3dsp_neon.S @ d782bca4
History | View | Annotate | Download (3.08 KB)
| 1 |
/* |
|---|---|
| 2 |
* Copyright (c) 2011 Mans Rullgard <mans@mansr.com> |
| 3 |
* |
| 4 |
* This file is part of Libav. |
| 5 |
* |
| 6 |
* Libav is free software; you can redistribute it and/or |
| 7 |
* modify it under the terms of the GNU Lesser General Public |
| 8 |
* License as published by the Free Software Foundation; either |
| 9 |
* version 2.1 of the License, or (at your option) any later version. |
| 10 |
* |
| 11 |
* Libav is distributed in the hope that it will be useful, |
| 12 |
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 13 |
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 14 |
* Lesser General Public License for more details. |
| 15 |
* |
| 16 |
* You should have received a copy of the GNU Lesser General Public |
| 17 |
* License along with Libav; if not, write to the Free Software |
| 18 |
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
| 19 |
*/ |
| 20 |
|
| 21 |
#include "asm.S" |
| 22 |
|
| 23 |
function ff_ac3_max_msb_abs_int16_neon, export=1 |
| 24 |
vmov.i16 q0, #0 |
| 25 |
vmov.i16 q2, #0 |
| 26 |
1: vld1.16 {q1}, [r0,:128]!
|
| 27 |
vabs.s16 q1, q1 |
| 28 |
vld1.16 {q3}, [r0,:128]!
|
| 29 |
vabs.s16 q3, q3 |
| 30 |
vorr q0, q0, q1 |
| 31 |
vorr q2, q2, q3 |
| 32 |
subs r1, r1, #16 |
| 33 |
bgt 1b |
| 34 |
vorr q0, q0, q2 |
| 35 |
vorr d0, d0, d1 |
| 36 |
vpmax.u16 d0, d0, d0 |
| 37 |
vpmax.u16 d0, d0, d0 |
| 38 |
vmov.u16 r0, d0[0] |
| 39 |
bx lr |
| 40 |
endfunc |
| 41 |
|
| 42 |
function ff_ac3_exponent_min_neon, export=1 |
| 43 |
cmp r1, #0 |
| 44 |
bxeq lr |
| 45 |
push {lr}
|
| 46 |
mov r12, #256 |
| 47 |
1: |
| 48 |
vld1.8 {q0}, [r0,:128]
|
| 49 |
mov lr, r1 |
| 50 |
add r3, r0, #256 |
| 51 |
2: vld1.8 {q1}, [r3,:128], r12
|
| 52 |
subs lr, lr, #1 |
| 53 |
vmin.u8 q0, q0, q1 |
| 54 |
bgt 2b |
| 55 |
subs r2, r2, #16 |
| 56 |
vst1.8 {q0}, [r0,:128]!
|
| 57 |
bgt 1b |
| 58 |
pop {pc}
|
| 59 |
endfunc |
| 60 |
|
| 61 |
function ff_ac3_lshift_int16_neon, export=1 |
| 62 |
vdup.16 q0, r2 |
| 63 |
1: vld1.16 {q1}, [r0,:128]
|
| 64 |
vshl.s16 q1, q1, q0 |
| 65 |
vst1.16 {q1}, [r0,:128]!
|
| 66 |
subs r1, r1, #8 |
| 67 |
bgt 1b |
| 68 |
bx lr |
| 69 |
endfunc |
| 70 |
|
| 71 |
function ff_ac3_rshift_int32_neon, export=1 |
| 72 |
rsb r2, r2, #0 |
| 73 |
vdup.32 q0, r2 |
| 74 |
1: vld1.32 {q1}, [r0,:128]
|
| 75 |
vshl.s32 q1, q1, q0 |
| 76 |
vst1.32 {q1}, [r0,:128]!
|
| 77 |
subs r1, r1, #4 |
| 78 |
bgt 1b |
| 79 |
bx lr |
| 80 |
endfunc |
| 81 |
|
| 82 |
function ff_float_to_fixed24_neon, export=1 |
| 83 |
1: vld1.32 {q0-q1}, [r1,:128]!
|
| 84 |
vcvt.s32.f32 q0, q0, #24 |
| 85 |
vld1.32 {q2-q3}, [r1,:128]!
|
| 86 |
vcvt.s32.f32 q1, q1, #24 |
| 87 |
vcvt.s32.f32 q2, q2, #24 |
| 88 |
vst1.32 {q0-q1}, [r0,:128]!
|
| 89 |
vcvt.s32.f32 q3, q3, #24 |
| 90 |
vst1.32 {q2-q3}, [r0,:128]!
|
| 91 |
subs r2, r2, #16 |
| 92 |
bgt 1b |
| 93 |
bx lr |
| 94 |
endfunc |