yading@10: /*
yading@10:  * Copyright (c) 2010 Mans Rullgard <mans@mansr.com>
yading@10:  *
yading@10:  * This file is part of FFmpeg.
yading@10:  *
yading@10:  * FFmpeg is free software; you can redistribute it and/or
yading@10:  * modify it under the terms of the GNU Lesser General Public
yading@10:  * License as published by the Free Software Foundation; either
yading@10:  * version 2.1 of the License, or (at your option) any later version.
yading@10:  *
yading@10:  * FFmpeg is distributed in the hope that it will be useful,
yading@10:  * but WITHOUT ANY WARRANTY; without even the implied warranty of
yading@10:  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
yading@10:  * Lesser General Public License for more details.
yading@10:  *
yading@10:  * You should have received a copy of the GNU Lesser General Public
yading@10:  * License along with FFmpeg; if not, write to the Free Software
yading@10:  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
yading@10:  */
yading@10: 
yading@10: #include "libavutil/arm/asm.S"
yading@10: 
yading@10: function ff_dca_lfe_fir_neon, export=1
yading@10:         push            {r4-r6,lr}
yading@10: 
yading@10:         add             r4,  r0,  r3,  lsl #2   @ out2
yading@10:         add             r5,  r2,  #256*4-16     @ cf1
yading@10:         sub             r1,  r1,  #12
yading@10:         cmp             r3,  #32
yading@10:         ite             eq
yading@10:         moveq           r6,  #256/32
yading@10:         movne           r6,  #256/64
yading@10: NOVFP   vldr            s0,  [sp, #16]          @ scale
yading@10:         mov             lr,  #-16
yading@10: 1:
yading@10:         vmov.f32        q2,  #0.0               @ v0
yading@10:         vmov.f32        q3,  #0.0               @ v1
yading@10:         mov             r12, r6
yading@10: 2:
yading@10:         vld1.32         {q8},     [r2,:128]!    @ cf0
yading@10:         vld1.32         {q9},     [r5,:128], lr @ cf1
yading@10:         vld1.32         {q1},     [r1], lr      @ in
yading@10:         subs            r12, r12, #4
yading@10:         vrev64.32       q10, q8
yading@10:         vmla.f32        q3,  q1,  q9
yading@10:         vmla.f32        d4,  d2,  d21
yading@10:         vmla.f32        d5,  d3,  d20
yading@10:         bne             2b
yading@10: 
yading@10:         add             r1,  r1,  r6,  lsl #2
yading@10:         subs            r3,  r3,  #1
yading@10:         vadd.f32        d4,  d4,  d5
yading@10:         vadd.f32        d6,  d6,  d7
yading@10:         vpadd.f32       d4,  d4,  d6
yading@10:         vmul.f32        d5,  d4,  d0[0]
yading@10:         vst1.32         {d5[0]},  [r0,:32]!
yading@10:         vst1.32         {d5[1]},  [r4,:32]!
yading@10:         bne             1b
yading@10: 
yading@10:         pop             {r4-r6,pc}
yading@10: endfunc