arm: hpeldsp: fix put_pixels8_y2_{,no_rnd_}armv6
[ffmpeg.git] / libavcodec / arm / hpeldsp_armv6.S
1 /*
2  * Copyright (c) 2009 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 "libavutil/arm/asm.S"
22
23 .macro  call_2x_pixels  type, subp
24 function ff_\type\()_pixels16\subp\()_armv6, export=1
25         push            {r0-r3, lr}
26         bl              X(ff_\type\()_pixels8\subp\()_armv6)
27         pop             {r0-r3, lr}
28         add             r0,  r0,  #8
29         add             r1,  r1,  #8
30         b               X(ff_\type\()_pixels8\subp\()_armv6)
31 endfunc
32 .endm
33
34 call_2x_pixels          avg
35 call_2x_pixels          put, _x2
36 call_2x_pixels          put, _y2
37 call_2x_pixels          put, _x2_no_rnd
38 call_2x_pixels          put, _y2_no_rnd
39
40 function ff_put_pixels16_armv6, export=1
41         push            {r4-r11}
42 1:
43         ldr             r5,  [r1, #4]
44         ldr             r6,  [r1, #8]
45         ldr             r7,  [r1, #12]
46         ldr_post        r4,  r1,  r2
47         strd            r6,  r7,  [r0, #8]
48         ldr             r9,  [r1, #4]
49         strd_post       r4,  r5,  r0,  r2
50         ldr             r10, [r1, #8]
51         ldr             r11, [r1, #12]
52         ldr_post        r8,  r1,  r2
53         strd            r10, r11, [r0, #8]
54         subs            r3,  r3,  #2
55         strd_post       r8,  r9,  r0,  r2
56         bne             1b
57
58         pop             {r4-r11}
59         bx              lr
60 endfunc
61
62 function ff_put_pixels8_armv6, export=1
63         push            {r4-r7}
64 1:
65         ldr             r5,  [r1, #4]
66         ldr_post        r4,  r1,  r2
67         ldr             r7,  [r1, #4]
68         strd_post       r4,  r5,  r0,  r2
69         ldr_post        r6,  r1,  r2
70         subs            r3,  r3,  #2
71         strd_post       r6,  r7,  r0,  r2
72         bne             1b
73
74         pop             {r4-r7}
75         bx              lr
76 endfunc
77
78 function ff_put_pixels8_x2_armv6, export=1
79         push            {r4-r11, lr}
80         mov             r12, #1
81         orr             r12, r12, r12, lsl #8
82         orr             r12, r12, r12, lsl #16
83 1:
84         ldr             r4,  [r1]
85         subs            r3,  r3,  #2
86         ldr             r5,  [r1, #4]
87         ldr             r7,  [r1, #5]
88         lsr             r6,  r4,  #8
89         ldr_pre         r8,  r1,  r2
90         orr             r6,  r6,  r5,  lsl #24
91         ldr             r9,  [r1, #4]
92         ldr             r11, [r1, #5]
93         lsr             r10, r8,  #8
94         add             r1,  r1,  r2
95         orr             r10, r10, r9,  lsl #24
96         eor             r14, r4,  r6
97         uhadd8          r4,  r4,  r6
98         eor             r6,  r5,  r7
99         uhadd8          r5,  r5,  r7
100         and             r14, r14, r12
101         and             r6,  r6,  r12
102         uadd8           r4,  r4,  r14
103         eor             r14, r8,  r10
104         uadd8           r5,  r5,  r6
105         eor             r6,  r9,  r11
106         uhadd8          r8,  r8,  r10
107         and             r14, r14, r12
108         uhadd8          r9,  r9,  r11
109         and             r6,  r6,  r12
110         uadd8           r8,  r8,  r14
111         strd_post       r4,  r5,  r0,  r2
112         uadd8           r9,  r9,  r6
113         strd_post       r8,  r9,  r0,  r2
114         bne             1b
115
116         pop             {r4-r11, pc}
117 endfunc
118
119 function ff_put_pixels8_y2_armv6, export=1
120         push            {r4-r11}
121         mov             r12, #1
122         orr             r12, r12, r12, lsl #8
123         orr             r12, r12, r12, lsl #16
124         ldr             r4,  [r1]
125         ldr             r5,  [r1, #4]
126         ldr_pre         r6,  r1,  r2
127         ldr             r7,  [r1, #4]
128 1:
129         subs            r3,  r3,  #2
130         uhadd8          r8,  r4,  r6
131         eor             r10, r4,  r6
132         uhadd8          r9,  r5,  r7
133         eor             r11, r5,  r7
134         and             r10, r10, r12
135         ldr_pre         r4,  r1,  r2
136         uadd8           r8,  r8,  r10
137         and             r11, r11, r12
138         uadd8           r9,  r9,  r11
139         ldr             r5,  [r1, #4]
140         uhadd8          r10, r4,  r6
141         eor             r6,  r4,  r6
142         uhadd8          r11, r5,  r7
143         and             r6,  r6,  r12
144         eor             r7,  r5,  r7
145         uadd8           r10, r10, r6
146         and             r7,  r7,  r12
147         ldrc_pre        ne,  r6,  r1,  r2
148         uadd8           r11, r11, r7
149         strd_post       r8,  r9,  r0,  r2
150         it              ne
151         ldrne           r7,  [r1, #4]
152         strd_post       r10, r11, r0,  r2
153         bne             1b
154
155         pop             {r4-r11}
156         bx              lr
157 endfunc
158
159 function ff_put_pixels8_x2_no_rnd_armv6, export=1
160         push            {r4-r9, lr}
161 1:
162         subs            r3,  r3,  #2
163         ldr             r4,  [r1]
164         ldr             r5,  [r1, #4]
165         ldr             r7,  [r1, #5]
166         ldr_pre         r8,  r1,  r2
167         ldr             r9,  [r1, #4]
168         ldr             r14, [r1, #5]
169         add             r1,  r1,  r2
170         lsr             r6,  r4,  #8
171         orr             r6,  r6,  r5,  lsl #24
172         lsr             r12, r8,  #8
173         orr             r12, r12, r9,  lsl #24
174         uhadd8          r4,  r4,  r6
175         uhadd8          r5,  r5,  r7
176         uhadd8          r8,  r8,  r12
177         uhadd8          r9,  r9,  r14
178         stm             r0,  {r4,r5}
179         add             r0,  r0,  r2
180         stm             r0,  {r8,r9}
181         add             r0,  r0,  r2
182         bne             1b
183
184         pop             {r4-r9, pc}
185 endfunc
186
187 function ff_put_pixels8_y2_no_rnd_armv6, export=1
188         push            {r4-r9, lr}
189         ldr             r4,  [r1]
190         ldr             r5,  [r1, #4]
191         ldr_pre         r6,  r1,  r2
192         ldr             r7,  [r1, #4]
193 1:
194         subs            r3,  r3,  #2
195         uhadd8          r8,  r4,  r6
196         ldr_pre         r4,  r1,  r2
197         uhadd8          r9,  r5,  r7
198         ldr             r5,  [r1, #4]
199         uhadd8          r12, r4,  r6
200         ldrc_pre        ne,  r6,  r1,  r2
201         uhadd8          r14, r5,  r7
202         it              ne
203         ldrne           r7,  [r1, #4]
204         stm             r0,  {r8,r9}
205         add             r0,  r0,  r2
206         stm             r0,  {r12,r14}
207         add             r0,  r0,  r2
208         bne             1b
209
210         pop             {r4-r9, pc}
211 endfunc
212
213 function ff_avg_pixels8_armv6, export=1
214         pld             [r1, r2]
215         push            {r4-r10, lr}
216         mov             lr,  #1
217         orr             lr,  lr,  lr,  lsl #8
218         orr             lr,  lr,  lr,  lsl #16
219         ldrd            r4,  r5,  [r0]
220         ldr             r10, [r1, #4]
221         ldr_post        r9,  r1,  r2
222         subs            r3,  r3,  #2
223 1:
224         pld             [r1, r2]
225         eor             r8,  r4,  r9
226         uhadd8          r4,  r4,  r9
227         eor             r12, r5,  r10
228         ldrd_reg        r6,  r7,  r0,  r2
229         uhadd8          r5,  r5,  r10
230         and             r8,  r8,  lr
231         ldr             r10, [r1, #4]
232         and             r12, r12, lr
233         uadd8           r4,  r4,  r8
234         ldr_post        r9,  r1,  r2
235         eor             r8,  r6,  r9
236         uadd8           r5,  r5,  r12
237         pld             [r1, r2,  lsl #1]
238         eor             r12, r7,  r10
239         uhadd8          r6,  r6,  r9
240         strd_post       r4,  r5,  r0,  r2
241         uhadd8          r7,  r7,  r10
242         beq             2f
243         and             r8,  r8,  lr
244         ldrd_reg        r4,  r5,  r0,  r2
245         uadd8           r6,  r6,  r8
246         ldr             r10, [r1, #4]
247         and             r12, r12, lr
248         subs            r3,  r3,  #2
249         uadd8           r7,  r7,  r12
250         ldr_post        r9,  r1,  r2
251         strd_post       r6,  r7,  r0,  r2
252         b               1b
253 2:
254         and             r8,  r8,  lr
255         and             r12, r12, lr
256         uadd8           r6,  r6,  r8
257         uadd8           r7,  r7,  r12
258         strd_post       r6,  r7,  r0,  r2
259
260         pop             {r4-r10, pc}
261 endfunc