* Copyright (c) 2001 Michael Niedermayer <michaelni@gmx.at>
* Copyright (c) 2007 Mans Rullgard <mans@mansr.com>
*
- * This file is part of FFmpeg.
+ * This file is part of Libav.
*
- * FFmpeg is free software; you can redistribute it and/or
+ * Libav is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
- * FFmpeg is distributed in the hope that it will be useful,
+ * Libav is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
- * License along with FFmpeg; if not, write to the Free Software
+ * License along with Libav; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
-#include "asm.S"
+#include "libavutil/arm/asm.S"
#define W1 22725 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */
#define W2 21407 /* cos(i*M_PI/16)*sqrt(2)*(1<<14) + 0.5 */
#define W46 (W4 | (W6 << 16))
#define W57 (W5 | (W7 << 16))
- .text
- .align
-w13: .long W13
-w26: .long W26
-w42: .long W42
-w42n: .long W42n
-w46: .long W46
-w57: .long W57
-
/*
Compute partial IDCT of single row.
shift = left-shift amount
Output in registers r4--r11
*/
.macro idct_row shift
- ldr lr, w46 /* lr = W4 | (W6 << 16) */
+ ldr lr, =W46 /* lr = W4 | (W6 << 16) */
mov r1, #(1<<(\shift-1))
smlad r4, r2, ip, r1
smlsd r7, r2, ip, r1
- ldr ip, w13 /* ip = W1 | (W3 << 16) */
- ldr r10,w57 /* r10 = W5 | (W7 << 16) */
+ ldr ip, =W13 /* ip = W1 | (W3 << 16) */
+ ldr r10,=W57 /* r10 = W5 | (W7 << 16) */
smlad r5, r2, lr, r1
smlsd r6, r2, lr, r1
smlad r8, lr, r10,r8 /* B0 += W5*row[5] + W7*row[7] */
smusdx r10,r3, r1 /* r10 = B2 = W5*row[1] - W1*row[3] */
- ldr r3, w42n /* r3 = -W4 | (-W2 << 16) */
+ ldr r3, =W42n /* r3 = -W4 | (-W2 << 16) */
smlad r10,lr, r2, r10 /* B2 += W7*row[5] + W3*row[7] */
ldr r2, [r0, #4] /* r2 = row[6,4] */
smlsdx r11,lr, ip, r11 /* B3 += W3*row[5] - W1*row[7] */
- ldr ip, w46 /* ip = W4 | (W6 << 16) */
+ ldr ip, =W46 /* ip = W4 | (W6 << 16) */
smlad r9, lr, r1, r9 /* B1 -= W1*row[5] + W5*row[7] */
smlad r5, r2, r3, r5 /* A1 += -W4*row[4] - W2*row[6] */
Output in registers r4--r11
*/
.macro idct_row4 shift
- ldr lr, w46 /* lr = W4 | (W6 << 16) */
- ldr r10,w57 /* r10 = W5 | (W7 << 16) */
+ ldr lr, =W46 /* lr = W4 | (W6 << 16) */
+ ldr r10,=W57 /* r10 = W5 | (W7 << 16) */
mov r1, #(1<<(\shift-1))
smlad r4, r2, ip, r1
smlsd r7, r2, ip, r1
- ldr ip, w13 /* ip = W1 | (W3 << 16) */
+ ldr ip, =W13 /* ip = W1 | (W3 << 16) */
smlad r5, r2, lr, r1
smlsd r6, r2, lr, r1
smusdx r11,r3, r10 /* r11 = B3 = W7*row[1] - W5*row[3] */
ldr r3, [r0, #8] /* r3 = row[3,1] */
ldr r2, [r0] /* r2 = row[2,0] */
orrs lr, lr, ip
+ itt eq
cmpeq lr, r3
cmpeq lr, r2, lsr #16
beq 1f
push {r1}
- ldr ip, w42 /* ip = W4 | (W2 << 16) */
+ ldr ip, =W42 /* ip = W4 | (W2 << 16) */
cmp lr, #0
beq 2f
push {r1, lr}
ldr r2, [r0] /* r2 = row[2,0] */
- ldr ip, w42 /* ip = W4 | (W2 << 16) */
+ ldr ip, =W42 /* ip = W4 | (W2 << 16) */
ldr r3, [r0, #8] /* r3 = row[3,1] */
idct_row COL_SHIFT
pop {r1}
push {r1, r2, lr}
ldr r2, [r0] /* r2 = row[2,0] */
- ldr ip, w42 /* ip = W4 | (W2 << 16) */
+ ldr ip, =W42 /* ip = W4 | (W2 << 16) */
ldr r3, [r0, #8] /* r3 = row[3,1] */
idct_row COL_SHIFT
pop {r1, r2}
idct_finish_shift_sat COL_SHIFT
- strb r4, [r1], r2
- strb r5, [r1], r2
- strb r6, [r1], r2
- strb r7, [r1], r2
- strb r11,[r1], r2
- strb r10,[r1], r2
- strb r9, [r1], r2
- strb r8, [r1], r2
+ strb_post r4, r1, r2
+ strb_post r5, r1, r2
+ strb_post r6, r1, r2
+ strb_post r7, r1, r2
+ strb_post r11,r1, r2
+ strb_post r10,r1, r2
+ strb_post r9, r1, r2
+ strb_post r8, r1, r2
sub r1, r1, r2, lsl #3
push {r1, r2, lr}
ldr r2, [r0] /* r2 = row[2,0] */
- ldr ip, w42 /* ip = W4 | (W2 << 16) */
+ ldr ip, =W42 /* ip = W4 | (W2 << 16) */
ldr r3, [r0, #8] /* r3 = row[3,1] */
idct_row COL_SHIFT
pop {r1, r2}
add ip, r3, ip, asr #COL_SHIFT
usat ip, #8, ip
add r4, r7, r4, asr #COL_SHIFT
- strb ip, [r1], r2
+ strb_post ip, r1, r2
ldrb ip, [r1, r2]
usat r4, #8, r4
ldrb r11,[r1, r2, lsl #2]
add r5, ip, r5, asr #COL_SHIFT
usat r5, #8, r5
- strb r4, [r1], r2
+ strb_post r4, r1, r2
ldrb r3, [r1, r2]
ldrb ip, [r1, r2, lsl #2]
- strb r5, [r1], r2
+ strb_post r5, r1, r2
ldrb r7, [r1, r2]
ldrb r4, [r1, r2, lsl #2]
add r6, r3, r6, asr #COL_SHIFT
usat r8, #8, r8
add lr, r4, lr, asr #COL_SHIFT
usat lr, #8, lr
- strb r6, [r1], r2
- strb r10,[r1], r2
- strb r9, [r1], r2
- strb r8, [r1], r2
- strb lr, [r1], r2
+ strb_post r6, r1, r2
+ strb_post r10,r1, r2
+ strb_post r9, r1, r2
+ strb_post r8, r1, r2
+ strb_post lr, r1, r2
sub r1, r1, r2, lsl #3
sub r0, r0, #(16*7)
.endm
-/* void ff_simple_idct_armv6(DCTELEM *data); */
+/* void ff_simple_idct_armv6(int16_t *data); */
function ff_simple_idct_armv6, export=1
push {r4-r11, lr}
sub sp, sp, #128
pop {r4-r11, pc}
endfunc
-/* ff_simple_idct_add_armv6(uint8_t *dest, int line_size, DCTELEM *data); */
+/* ff_simple_idct_add_armv6(uint8_t *dest, int line_size, int16_t *data); */
function ff_simple_idct_add_armv6, export=1
push {r0, r1, r4-r11, lr}
sub sp, sp, #128
pop {r4-r11, pc}
endfunc
-/* ff_simple_idct_put_armv6(uint8_t *dest, int line_size, DCTELEM *data); */
+/* ff_simple_idct_put_armv6(uint8_t *dest, int line_size, int16_t *data); */
function ff_simple_idct_put_armv6, export=1
push {r0, r1, r4-r11, lr}
sub sp, sp, #128
projects / ffmpeg / blobdiff