generic-library/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
3a6ddfb874
ffmpeg/libavcodec/qpel_template.c

220 lines
14 KiB
C
Raw Normal View History

Move some functions in dsputil.c into a new file dsputil_template.c. The functions moved are used when decoding h264. Preparatory patch for high bit depth h264 decoding support. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2011-03-29 17:48:51 +02:00
/*
dsputil: Split off quarterpel bits into their own context
2014-01-08 14:00:10 +01:00
* quarterpel DSP function templates
Move some functions in dsputil.c into a new file dsputil_template.c. The functions moved are used when decoding h264. Preparatory patch for high bit depth h264 decoding support. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2011-03-29 17:48:51 +02:00
*
* This file is part of Libav.
*
* 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.
*
* 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 Libav; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
/**
* @file
dsputil: Split off quarterpel bits into their own context
2014-01-08 14:00:10 +01:00
* quarterpel DSP function templates
Move some functions in dsputil.c into a new file dsputil_template.c. The functions moved are used when decoding h264. Preparatory patch for high bit depth h264 decoding support. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2011-03-29 17:48:51 +02:00
*/
dsputil_template: K&R formatting cosmetics
2013-12-21 12:24:12 +01:00
#define PIXOP2(OPNAME, OP) \
dsputil_template: Detemplatize the code The indirection makes no sense without multiple instantiation.
2014-01-09 12:09:35 +01:00
static inline void OPNAME ## _no_rnd_pixels8_l2_8(uint8_t *dst, \
const uint8_t *src1, \
const uint8_t *src2, \
int dst_stride, \
int src_stride1, \
int src_stride2, \
int h) \
dsputil_template: K&R formatting cosmetics
2013-12-21 12:24:12 +01:00
{ \
int i; \
\
for (i = 0; i < h; i++) { \
dsputil_template: Detemplatize the code The indirection makes no sense without multiple instantiation.
2014-01-09 12:09:35 +01:00
uint32_t a, b; \
a = AV_RN32(&src1[i * src_stride1]); \
b = AV_RN32(&src2[i * src_stride2]); \
OP(*((uint32_t *) &dst[i * dst_stride]), \
no_rnd_avg32(a, b)); \
a = AV_RN32(&src1[i * src_stride1 + 4]); \
b = AV_RN32(&src2[i * src_stride2 + 4]); \
OP(*((uint32_t *) &dst[i * dst_stride + 4]), \
no_rnd_avg32(a, b)); \
dsputil_template: K&R formatting cosmetics
2013-12-21 12:24:12 +01:00
} \
} \
\
dsputil_template: Detemplatize the code The indirection makes no sense without multiple instantiation.
2014-01-09 12:09:35 +01:00
static inline void OPNAME ## _no_rnd_pixels16_l2_8(uint8_t *dst, \
const uint8_t *src1, \
const uint8_t *src2, \
int dst_stride, \
int src_stride1, \
int src_stride2, \
int h) \
dsputil_template: K&R formatting cosmetics
2013-12-21 12:24:12 +01:00
{ \
dsputil_template: Detemplatize the code The indirection makes no sense without multiple instantiation.
2014-01-09 12:09:35 +01:00
OPNAME ## _no_rnd_pixels8_l2_8(dst, src1, src2, dst_stride, \
src_stride1, src_stride2, h); \
OPNAME ## _no_rnd_pixels8_l2_8(dst + 8, \
src1 + 8, \
src2 + 8, \
dst_stride, src_stride1, \
src_stride2, h); \
dsputil_template: K&R formatting cosmetics
2013-12-21 12:24:12 +01:00
} \
\
dsputil_template: Detemplatize the code The indirection makes no sense without multiple instantiation.
2014-01-09 12:09:35 +01:00
static inline void OPNAME ## _pixels8_l4_8(uint8_t *dst, \
const uint8_t *src1, \
const uint8_t *src2, \
const uint8_t *src3, \
const uint8_t *src4, \
int dst_stride, \
int src_stride1, \
int src_stride2, \
int src_stride3, \
int src_stride4, \
int h) \
dsputil_template: K&R formatting cosmetics
2013-12-21 12:24:12 +01:00
{ \
/* FIXME HIGH BIT DEPTH */ \
int i; \
\
for (i = 0; i < h; i++) { \
uint32_t a, b, c, d, l0, l1, h0, h1; \
a = AV_RN32(&src1[i * src_stride1]); \
b = AV_RN32(&src2[i * src_stride2]); \
c = AV_RN32(&src3[i * src_stride3]); \
d = AV_RN32(&src4[i * src_stride4]); \
l0 = (a & 0x03030303UL) + \
(b & 0x03030303UL) + \
0x02020202UL; \
h0 = ((a & 0xFCFCFCFCUL) >> 2) + \
((b & 0xFCFCFCFCUL) >> 2); \
l1 = (c & 0x03030303UL) + \
(d & 0x03030303UL); \
h1 = ((c & 0xFCFCFCFCUL) >> 2) + \
((d & 0xFCFCFCFCUL) >> 2); \
OP(*((uint32_t *) &dst[i * dst_stride]), \
h0 + h1 + (((l0 + l1) >> 2) & 0x0F0F0F0FUL)); \
a = AV_RN32(&src1[i * src_stride1 + 4]); \
b = AV_RN32(&src2[i * src_stride2 + 4]); \
c = AV_RN32(&src3[i * src_stride3 + 4]); \
d = AV_RN32(&src4[i * src_stride4 + 4]); \
l0 = (a & 0x03030303UL) + \
(b & 0x03030303UL) + \
0x02020202UL; \
h0 = ((a & 0xFCFCFCFCUL) >> 2) + \
((b & 0xFCFCFCFCUL) >> 2); \
l1 = (c & 0x03030303UL) + \
(d & 0x03030303UL); \
h1 = ((c & 0xFCFCFCFCUL) >> 2) + \
((d & 0xFCFCFCFCUL) >> 2); \
OP(*((uint32_t *) &dst[i * dst_stride + 4]), \
h0 + h1 + (((l0 + l1) >> 2) & 0x0F0F0F0FUL)); \
} \
} \
\
dsputil_template: Detemplatize the code The indirection makes no sense without multiple instantiation.
2014-01-09 12:09:35 +01:00
static inline void OPNAME ## _no_rnd_pixels8_l4_8(uint8_t *dst, \
const uint8_t *src1, \
const uint8_t *src2, \
const uint8_t *src3, \
const uint8_t *src4, \
int dst_stride, \
int src_stride1, \
int src_stride2, \
int src_stride3, \
int src_stride4, \
int h) \
dsputil_template: K&R formatting cosmetics
2013-12-21 12:24:12 +01:00
{ \
/* FIXME HIGH BIT DEPTH */ \
int i; \
\
for (i = 0; i < h; i++) { \
uint32_t a, b, c, d, l0, l1, h0, h1; \
a = AV_RN32(&src1[i * src_stride1]); \
b = AV_RN32(&src2[i * src_stride2]); \
c = AV_RN32(&src3[i * src_stride3]); \
d = AV_RN32(&src4[i * src_stride4]); \
l0 = (a & 0x03030303UL) + \
(b & 0x03030303UL) + \
0x01010101UL; \
h0 = ((a & 0xFCFCFCFCUL) >> 2) + \
((b & 0xFCFCFCFCUL) >> 2); \
l1 = (c & 0x03030303UL) + \
(d & 0x03030303UL); \
h1 = ((c & 0xFCFCFCFCUL) >> 2) + \
((d & 0xFCFCFCFCUL) >> 2); \
OP(*((uint32_t *) &dst[i * dst_stride]), \
h0 + h1 + (((l0 + l1) >> 2) & 0x0F0F0F0FUL)); \
a = AV_RN32(&src1[i * src_stride1 + 4]); \
b = AV_RN32(&src2[i * src_stride2 + 4]); \
c = AV_RN32(&src3[i * src_stride3 + 4]); \
d = AV_RN32(&src4[i * src_stride4 + 4]); \
l0 = (a & 0x03030303UL) + \
(b & 0x03030303UL) + \
0x01010101UL; \
h0 = ((a & 0xFCFCFCFCUL) >> 2) + \
((b & 0xFCFCFCFCUL) >> 2); \
l1 = (c & 0x03030303UL) + \
(d & 0x03030303UL); \
h1 = ((c & 0xFCFCFCFCUL) >> 2) + \
((d & 0xFCFCFCFCUL) >> 2); \
OP(*((uint32_t *) &dst[i * dst_stride + 4]), \
h0 + h1 + (((l0 + l1) >> 2) & 0x0F0F0F0FUL)); \
} \
} \
dsputil_template: Detemplatize the code The indirection makes no sense without multiple instantiation.
2014-01-09 12:09:35 +01:00
\
static inline void OPNAME ## _pixels16_l4_8(uint8_t *dst, \
const uint8_t *src1, \
const uint8_t *src2, \
const uint8_t *src3, \
const uint8_t *src4, \
int dst_stride, \
int src_stride1, \
int src_stride2, \
int src_stride3, \
int src_stride4, \
int h) \
dsputil_template: K&R formatting cosmetics
2013-12-21 12:24:12 +01:00
{ \
dsputil_template: Detemplatize the code The indirection makes no sense without multiple instantiation.
2014-01-09 12:09:35 +01:00
OPNAME ## _pixels8_l4_8(dst, src1, src2, src3, src4, dst_stride, \
src_stride1, src_stride2, src_stride3, \
src_stride4, h); \
OPNAME ## _pixels8_l4_8(dst + 8, \
src1 + 8, src2 + 8, \
src3 + 8, src4 + 8, \
dst_stride, src_stride1, src_stride2, \
src_stride3, src_stride4, h); \
dsputil_template: K&R formatting cosmetics
2013-12-21 12:24:12 +01:00
} \
dsputil_template: Detemplatize the code The indirection makes no sense without multiple instantiation.
2014-01-09 12:09:35 +01:00
\
static inline void OPNAME ## _no_rnd_pixels16_l4_8(uint8_t *dst, \
const uint8_t *src1, \
const uint8_t *src2, \
const uint8_t *src3, \
const uint8_t *src4, \
int dst_stride, \
int src_stride1, \
int src_stride2, \
int src_stride3, \
int src_stride4, \
int h) \
dsputil_template: K&R formatting cosmetics
2013-12-21 12:24:12 +01:00
{ \
dsputil_template: Detemplatize the code The indirection makes no sense without multiple instantiation.
2014-01-09 12:09:35 +01:00
OPNAME ## _no_rnd_pixels8_l4_8(dst, src1, src2, src3, src4, \
dst_stride, src_stride1, \
src_stride2, src_stride3, \
src_stride4, h); \
OPNAME ## _no_rnd_pixels8_l4_8(dst + 8, \
src1 + 8, src2 + 8, \
src3 + 8, src4 + 8, \
dst_stride, src_stride1, \
src_stride2, src_stride3, \
src_stride4, h); \
dsputil_template: K&R formatting cosmetics
2013-12-21 12:24:12 +01:00
} \
Move some functions in dsputil.c into a new file dsputil_template.c. The functions moved are used when decoding h264. Preparatory patch for high bit depth h264 decoding support. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2011-03-29 17:48:51 +02:00
dsputil_template: Detemplatize the code The indirection makes no sense without multiple instantiation.
2014-01-09 12:09:35 +01:00
#define op_avg(a, b) a = rnd_avg32(a, b)
Move some functions in dsputil.c into a new file dsputil_template.c. The functions moved are used when decoding h264. Preparatory patch for high bit depth h264 decoding support. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2011-03-29 17:48:51 +02:00
#define op_put(a, b) a = b
dsputil: Separate h264 qpel The sh4 optimizations are removed, because the code is 100% identical to the C code, so it is unlikely to provide any real practical benefit. Signed-off-by: Diego Biurrun <diego@biurrun.de> Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com> Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
2013-01-24 05:24:53 +01:00
#define put_no_rnd_pixels8_8_c put_pixels8_8_c
Move some functions in dsputil.c into a new file dsputil_template.c. The functions moved are used when decoding h264. Preparatory patch for high bit depth h264 decoding support. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2011-03-29 17:48:51 +02:00
PIXOP2(avg, op_avg)
PIXOP2(put, op_put)
#undef op_avg
#undef op_put
Reference in New Issue Copy Permalink