yading@10: /*
yading@10:  * Copyright (c) 2012 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 flac_lpc_16_1_arm
yading@10:         ldr             r12, [sp]
yading@10:         push            {r4, lr}
yading@10:         ldr             r1,  [r1]
yading@10:         subs            r12, r12, #2
yading@10:         ldr             lr,  [r0], #4
yading@10:         beq             2f
yading@10:         it              lt
yading@10:         poplt           {r4, pc}
yading@10: 1:
yading@10:         mul             r4,  lr,  r1
yading@10:         ldm             r0,  {r2, lr}
yading@10:         add_sh          r2,  r2,  r4,  asr r3
yading@10:         mul             r4,  r2,  r1
yading@10:         subs            r12, r12, #2
yading@10:         add_sh          lr,  lr,  r4,  asr r3
yading@10:         stm             r0!, {r2, lr}
yading@10:         bgt             1b
yading@10:         it              lt
yading@10:         poplt           {r4, pc}
yading@10: 2:
yading@10:         mul             r4,  lr,  r1
yading@10:         ldr             r2,  [r0]
yading@10:         add_sh          r2,  r2,  r4,  asr r3
yading@10:         str             r2,  [r0]
yading@10:         pop             {r4, pc}
yading@10: endfunc
yading@10: 
yading@10: function flac_lpc_16_2_arm
yading@10:         ldr             r12, [sp]
yading@10:         subs            r12, r12, r2
yading@10:         it              le
yading@10:         bxle            lr
yading@10: 
yading@10:         push            {r4-r9, lr}
yading@10:         ldm             r0!, {r6, r7}
yading@10:         ldm             r1,  {r8, r9}
yading@10:         subs            r12, r12, #1
yading@10:         beq             2f
yading@10: 1:
yading@10:         mul             r4,  r6,  r8
yading@10:         mul             r5,  r7,  r8
yading@10:         mla             r4,  r7,  r9,  r4
yading@10:         ldm             r0,  {r6, r7}
yading@10:         add_sh          r6,  r6,  r4,  asr r3
yading@10:         mla             r5,  r6,  r9,  r5
yading@10:         add_sh          r7,  r7,  r5,  asr r3
yading@10:         stm             r0!, {r6, r7}
yading@10:         subs            r12, r12, #2
yading@10:         bgt             1b
yading@10:         it              lt
yading@10:         poplt           {r4-r9, pc}
yading@10: 2:
yading@10:         mul             r4,  r6,  r8
yading@10:         mla             r4,  r7,  r9,  r4
yading@10:         ldr             r5,  [r0]
yading@10:         add_sh          r5,  r5,  r4,  asr r3
yading@10:         str             r5,  [r0]
yading@10:         pop             {r4-r9, pc}
yading@10: endfunc
yading@10: 
yading@10: function ff_flac_lpc_16_arm, export=1
yading@10:         cmp             r2,  #2
yading@10:         blt             flac_lpc_16_1_arm
yading@10:         beq             flac_lpc_16_2_arm
yading@10: 
yading@10:         ldr             r12, [sp]
yading@10:         subs            r12, r12, r2
yading@10:         it              le
yading@10:         bxle            lr
yading@10: 
yading@10:         push            {r4-r9, lr}
yading@10: 
yading@10:         subs            r12, r12, #1
yading@10:         beq             3f
yading@10: 1:
yading@10:         sub             lr,  r2,  #2
yading@10:         mov             r4,  #0
yading@10:         mov             r5,  #0
yading@10: 
yading@10:         ldr             r7,  [r0], #4
yading@10:         ldr             r9,  [r1], #4
yading@10: 2:
yading@10:         mla             r4,  r7,  r9,  r4
yading@10:         ldm             r0!, {r6, r7}
yading@10:         mla             r5,  r6,  r9,  r5
yading@10:         ldm             r1!, {r8, r9}
yading@10:         mla             r4,  r6,  r8,  r4
yading@10:         subs            lr,  lr,  #2
yading@10:         mla             r5,  r7,  r8,  r5
yading@10:         bgt             2b
yading@10:         blt             6f
yading@10: 
yading@10:         mla             r4,  r7,  r9,  r4
yading@10:         ldr             r7,  [r0], #4
yading@10:         mla             r5,  r7,  r9,  r5
yading@10:         ldr             r9,  [r1], #4
yading@10: 6:
yading@10:         mla             r4,  r7,  r9,  r4
yading@10:         ldm             r0,  {r6, r7}
yading@10:         add_sh          r6,  r6,  r4,  asr r3
yading@10:         mla             r5,  r6,  r9,  r5
yading@10:         add_sh          r7,  r7,  r5,  asr r3
yading@10:         stm             r0!, {r6, r7}
yading@10:         sub             r0,  r0,  r2,  lsl #2
yading@10:         sub             r1,  r1,  r2,  lsl #2
yading@10: 
yading@10:         subs            r12, r12, #2
yading@10:         bgt             1b
yading@10:         it              lt
yading@10:         poplt           {r4-r9, pc}
yading@10: 3:
yading@10:         mov             r4,  #0
yading@10: 4:
yading@10:         ldr             r5,  [r1], #4
yading@10:         ldr             r6,  [r0], #4
yading@10:         mla             r4,  r5,  r6,  r4
yading@10:         subs            r2,  r2,  #1
yading@10:         bgt             4b
yading@10:         ldr             r5,  [r0]
yading@10:         add_sh          r5,  r5,  r4,  asr r3
yading@10:         str             r5,  [r0]
yading@10:         pop             {r4-r9, pc}
yading@10: endfunc