comparison arm/intreadwrite.h @ 728:1fa3820b1a84 libavutil

ARM asm for AV_RN*() ARMv6 and later support unaligned loads and stores for single word/halfword but not double/multiple. GCC is ignorant of this and will always use bytewise accesses for unaligned data. Casting to an int32_t pointer is dangerous since a load/store double or multiple instruction might be used (this happens with some code in FFmpeg). Implementing the AV_[RW]* macros with inline asm using only supported instructions gives fast and safe unaligned accesses. ARM RVCT does the right thing with generic code. This gives an overall speedup of up to 10%.
author mru
date Sat, 18 Apr 2009 00:00:28 +0000
parents
children 2d52bcf3e4e6
comparison
equal deleted inserted replaced
727:98b64f65be0d 728:1fa3820b1a84
1 /*
2 * This file is part of FFmpeg.
3 *
4 * FFmpeg is free software; you can redistribute it and/or
5 * modify it under the terms of the GNU Lesser General Public
6 * License as published by the Free Software Foundation; either
7 * version 2.1 of the License, or (at your option) any later version.
8 *
9 * FFmpeg is distributed in the hope that it will be useful,
10 * but WITHOUT ANY WARRANTY; without even the implied warranty of
11 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
12 * Lesser General Public License for more details.
13 *
14 * You should have received a copy of the GNU Lesser General Public
15 * License along with FFmpeg; if not, write to the Free Software
16 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
17 */
18
19 #ifndef AVUTIL_ARM_INTREADWRITE_H
20 #define AVUTIL_ARM_INTREADWRITE_H
21
22 #include <stdint.h>
23 #include "config.h"
24
25 #if HAVE_FAST_UNALIGNED && HAVE_INLINE_ASM
26
27 #define AV_RN16 AV_RN16
28 static inline uint16_t AV_RN16(const void *p)
29 {
30 uint16_t v;
31 __asm__ ("ldrh %0, %1" : "=r"(v) : "m"(*(const uint16_t *)p));
32 return v;
33 }
34
35 #define AV_WN16 AV_WN16
36 static inline void AV_WN16(void *p, uint16_t v)
37 {
38 __asm__ ("strh %1, %0" : "=m"(*(uint16_t *)p) : "r"(v));
39 }
40
41 #define AV_RN32 AV_RN32
42 static inline uint32_t AV_RN32(const void *p)
43 {
44 uint32_t v;
45 __asm__ ("ldr %0, %1" : "=r"(v) : "m"(*(const uint32_t *)p));
46 return v;
47 }
48
49 #define AV_WN32 AV_WN32
50 static inline void AV_WN32(void *p, uint32_t v)
51 {
52 __asm__ ("str %1, %0" : "=m"(*(uint32_t *)p) : "r"(v));
53 }
54
55 #define AV_RN64 AV_RN64
56 static inline uint64_t AV_RN64(const void *p)
57 {
58 union { uint64_t v; uint32_t hl[2]; } v;
59 __asm__ ("ldr %0, %2 \n\t"
60 "ldr %1, %3 \n\t"
61 : "=r"(v.hl[0]), "=r"(v.hl[1])
62 : "m"(*(const uint32_t*)p), "m"(*((const uint32_t*)p+1)));
63 return v.v;
64 }
65
66 #define AV_WN64 AV_WN64
67 static inline void AV_WN64(void *p, uint64_t v)
68 {
69 union { uint64_t v; uint32_t hl[2]; } vv = { v };
70 __asm__ ("str %2, %0 \n\t"
71 "str %3, %1 \n\t"
72 : "=m"(*(uint32_t*)p), "=m"(*((uint32_t*)p+1))
73 : "r"(vv.hl[0]), "r"(vv.hl[1]));
74 }
75
76 #endif /* HAVE_INLINE_ASM */
77
78 #endif /* AVUTIL_ARM_INTREADWRITE_H */