2011-03-29 18:48:51 +03:00
|
|
|
/*
|
2014-01-08 16:00:10 +03:00
|
|
|
* quarterpel DSP function templates
|
2011-03-29 18:48:51 +03:00
|
|
|
*
|
2011-05-11 14:56:12 +03:00
|
|
|
* This file is part of FFmpeg.
|
2011-03-29 18:48:51 +03:00
|
|
|
*
|
2011-05-11 14:56:12 +03:00
|
|
|
* FFmpeg is free software; you can redistribute it and/or
|
2011-03-29 18:48:51 +03:00
|
|
|
* 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.
|
|
|
|
*
|
2011-05-11 14:56:12 +03:00
|
|
|
* FFmpeg is distributed in the hope that it will be useful,
|
2011-03-29 18:48:51 +03:00
|
|
|
* 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
|
2011-05-11 14:56:12 +03:00
|
|
|
* License along with FFmpeg; if not, write to the Free Software
|
2011-03-29 18:48:51 +03:00
|
|
|
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
|
|
|
|
*/
|
|
|
|
|
|
|
|
/**
|
|
|
|
* @file
|
2014-01-08 16:00:10 +03:00
|
|
|
* quarterpel DSP function templates
|
2011-03-29 18:48:51 +03:00
|
|
|
*/
|
|
|
|
|
2013-12-21 14:24:12 +03:00
|
|
|
#define PIXOP2(OPNAME, OP) \
|
2014-01-09 14:09:35 +03:00
|
|
|
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) \
|
2013-12-21 14:24:12 +03:00
|
|
|
{ \
|
|
|
|
int i; \
|
|
|
|
\
|
|
|
|
for (i = 0; i < h; i++) { \
|
2014-01-09 14:09:35 +03:00
|
|
|
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)); \
|
2013-12-21 14:24:12 +03:00
|
|
|
} \
|
|
|
|
} \
|
|
|
|
\
|
2014-01-09 14:09:35 +03:00
|
|
|
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) \
|
2013-12-21 14:24:12 +03:00
|
|
|
{ \
|
2014-01-09 14:09:35 +03:00
|
|
|
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); \
|
2013-12-21 14:24:12 +03:00
|
|
|
} \
|
|
|
|
\
|
2014-01-09 14:09:35 +03:00
|
|
|
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) \
|
2013-12-21 14:24:12 +03:00
|
|
|
{ \
|
|
|
|
/* 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)); \
|
|
|
|
} \
|
|
|
|
} \
|
|
|
|
\
|
2014-01-09 14:09:35 +03:00
|
|
|
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) \
|
2013-12-21 14:24:12 +03:00
|
|
|
{ \
|
|
|
|
/* 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)); \
|
|
|
|
} \
|
|
|
|
} \
|
2014-01-09 14:09:35 +03:00
|
|
|
\
|
|
|
|
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) \
|
2013-12-21 14:24:12 +03:00
|
|
|
{ \
|
2014-01-09 14:09:35 +03:00
|
|
|
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); \
|
2013-12-21 14:24:12 +03:00
|
|
|
} \
|
2014-01-09 14:09:35 +03:00
|
|
|
\
|
|
|
|
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) \
|
2013-12-21 14:24:12 +03:00
|
|
|
{ \
|
2014-01-09 14:09:35 +03:00
|
|
|
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); \
|
2013-12-21 14:24:12 +03:00
|
|
|
} \
|
2011-03-29 18:48:51 +03:00
|
|
|
|
2014-01-09 14:09:35 +03:00
|
|
|
#define op_avg(a, b) a = rnd_avg32(a, b)
|
2011-03-29 18:48:51 +03:00
|
|
|
#define op_put(a, b) a = b
|
2013-01-24 07:24:53 +03:00
|
|
|
#define put_no_rnd_pixels8_8_c put_pixels8_8_c
|
2011-03-29 18:48:51 +03:00
|
|
|
PIXOP2(avg, op_avg)
|
|
|
|
PIXOP2(put, op_put)
|
|
|
|
#undef op_avg
|
|
|
|
#undef op_put
|