1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-11-21 10:55:51 +02:00

mlib merge

Originally committed as revision 131 to svn://svn.ffmpeg.org/ffmpeg/trunk
This commit is contained in:
Fabrice Bellard 2001-09-16 21:54:00 +00:00
parent 8d67072fea
commit c34270f5e8
4 changed files with 181 additions and 12 deletions

View File

@ -29,6 +29,13 @@ ASM_OBJS += armv4l/jrevdct_arm.o
OBJS += armv4l/dsputil_arm.o
endif
# sun mediaLib specific stuff
# currently only works when libavcodec is used in mplayer
ifeq ($(HAVE_MLIB),yes)
OBJS += mlib/dsputil_mlib.o
CFLAGS += $(MLIB_INC)
endif
SRCS = $(OBJS:.o=.c) $(ASM_OBJS:.o=.s)
LIB= libavcodec.a
@ -58,8 +65,8 @@ depend:
clean:
rm -f *.o *~ *.a i386/*.o i386/*~ \
armv4l/*.o armv4l/*~ \
mlib/*.o mlib/*~ \
libac3/*.o libac3/*~ \
mpglib/*.o mpglib/*~ \
apiexample $(TESTS)
distclean: clean

View File

@ -413,6 +413,7 @@ void block_permute(INT16 *block)
void dsputil_init(void)
{
int i, j;
int use_permuted_idct;
for(i=0;i<256;i++) cropTbl[i + MAX_NEG_CROP] = i;
for(i=0;i<MAX_NEG_CROP;i++) {
@ -435,17 +436,7 @@ void dsputil_init(void)
pix_abs16x16_xy2 = pix_abs16x16_xy2_c;
av_fdct = jpeg_fdct_ifast;
/* permute for IDCT */
for(i=0;i<64;i++) {
j = zigzag_direct[i];
zigzag_direct[i] = block_permute_op(j);
j = ff_alternate_horizontal_scan[i];
ff_alternate_horizontal_scan[i] = block_permute_op(j);
j = ff_alternate_vertical_scan[i];
ff_alternate_vertical_scan[i] = block_permute_op(j);
}
block_permute(default_intra_matrix);
block_permute(default_non_intra_matrix);
use_permuted_idct = 1;
#ifdef HAVE_MMX
dsputil_init_mmx();
@ -453,4 +444,22 @@ void dsputil_init(void)
#ifdef ARCH_ARMV4L
dsputil_init_armv4l();
#endif
#ifdef HAVE_MLIB
dsputil_init_mlib();
use_permuted_idct = 0;
#endif
if (use_permuted_idct) {
/* permute for IDCT */
for(i=0;i<64;i++) {
j = zigzag_direct[i];
zigzag_direct[i] = block_permute_op(j);
j = ff_alternate_horizontal_scan[i];
ff_alternate_horizontal_scan[i] = block_permute_op(j);
j = ff_alternate_vertical_scan[i];
ff_alternate_vertical_scan[i] = block_permute_op(j);
}
block_permute(default_intra_matrix);
block_permute(default_non_intra_matrix);
}
}

View File

@ -111,6 +111,15 @@ void dsputil_init_mmx(void);
void dsputil_init_armv4l(void);
#elif defined(HAVE_MLIB)
#define emms_c()
/* SPARC/VIS IDCT needs 8-byte aligned DCT blocks */
#define __align8 __attribute__ ((aligned (8)))
void dsputil_init_mlib(void);
#else
#define emms_c()

View File

@ -0,0 +1,144 @@
/*
* Sun mediaLib optimized DSP utils
* Copyright (c) 2001 Juergen Keil.
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* This program 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 General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*/
#include "../dsputil.h"
#include <mlib_types.h>
#include <mlib_status.h>
#include <mlib_sys.h>
#include <mlib_video.h>
static void put_pixels_mlib (uint8_t * dest, const uint8_t * ref,
int stride, int height)
{
assert(height == 16 || height == 8);
if (height == 16)
mlib_VideoCopyRef_U8_U8_8x16(dest, (uint8_t *)ref, stride);
else
mlib_VideoCopyRef_U8_U8_8x8 (dest, (uint8_t *)ref, stride);
}
static void put_pixels_x2_mlib (uint8_t * dest, const uint8_t * ref,
int stride, int height)
{
assert(height == 16 || height == 8);
if (height == 16)
mlib_VideoInterpX_U8_U8_8x16(dest, (uint8_t *)ref, stride, stride);
else
mlib_VideoInterpX_U8_U8_8x8 (dest, (uint8_t *)ref, stride, stride);
}
static void put_pixels_y2_mlib (uint8_t * dest, const uint8_t * ref,
int stride, int height)
{
assert(height == 16 || height == 8);
if (height == 16)
mlib_VideoInterpY_U8_U8_8x16(dest, (uint8_t *)ref, stride, stride);
else
mlib_VideoInterpY_U8_U8_8x8 (dest, (uint8_t *)ref, stride, stride);
}
static void put_pixels_xy2_mlib(uint8_t * dest, const uint8_t * ref,
int stride, int height)
{
assert(height == 16 || height == 8);
if (height == 16)
mlib_VideoInterpXY_U8_U8_8x16(dest, (uint8_t *)ref, stride, stride);
else
mlib_VideoInterpXY_U8_U8_8x8 (dest, (uint8_t *)ref, stride, stride);
}
static void avg_pixels_mlib (uint8_t * dest, const uint8_t * ref,
int stride, int height)
{
assert(height == 16 || height == 8);
if (height == 16)
mlib_VideoCopyRefAve_U8_U8_8x16(dest, (uint8_t *)ref, stride);
else
mlib_VideoCopyRefAve_U8_U8_8x8 (dest, (uint8_t *)ref, stride);
}
static void avg_pixels_x2_mlib (uint8_t * dest, const uint8_t * ref,
int stride, int height)
{
assert(height == 16 || height == 8);
if (height == 16)
mlib_VideoInterpAveX_U8_U8_8x16(dest, (uint8_t *)ref, stride, stride);
else
mlib_VideoInterpAveX_U8_U8_8x8 (dest, (uint8_t *)ref, stride, stride);
}
static void avg_pixels_y2_mlib (uint8_t * dest, const uint8_t * ref,
int stride, int height)
{
assert(height == 16 || height == 8);
if (height == 16)
mlib_VideoInterpAveY_U8_U8_8x16(dest, (uint8_t *)ref, stride, stride);
else
mlib_VideoInterpAveY_U8_U8_8x8 (dest, (uint8_t *)ref, stride, stride);
}
static void avg_pixels_xy2_mlib (uint8_t * dest, const uint8_t * ref,
int stride, int height)
{
assert(height == 16 || height == 8);
if (height == 16)
mlib_VideoInterpAveXY_U8_U8_8x16(dest, (uint8_t *)ref, stride, stride);
else
mlib_VideoInterpAveXY_U8_U8_8x8 (dest, (uint8_t *)ref, stride, stride);
}
static void add_pixels_clamped_mlib(const DCTELEM *block, UINT8 *pixels, int line_size)
{
mlib_VideoAddBlock_U8_S16(pixels, (mlib_s16 *)block, line_size);
}
void ff_idct_mlib(DCTELEM *data)
{
mlib_VideoIDCT8x8_S16_S16 (data, data);
}
void ff_fdct_mlib(DCTELEM *data)
{
mlib_VideoDCT8x8_S16_S16 (data, data);
}
void dsputil_init_mlib(void)
{
av_fdct = ff_fdct_mlib;
ff_idct = ff_idct_mlib;
put_pixels_tab[0] = put_pixels_mlib;
put_pixels_tab[1] = put_pixels_x2_mlib;
put_pixels_tab[2] = put_pixels_y2_mlib;
put_pixels_tab[3] = put_pixels_xy2_mlib;
avg_pixels_tab[0] = avg_pixels_mlib;
avg_pixels_tab[1] = avg_pixels_x2_mlib;
avg_pixels_tab[2] = avg_pixels_y2_mlib;
avg_pixels_tab[3] = avg_pixels_xy2_mlib;
put_no_rnd_pixels_tab[0] = put_pixels_mlib;
add_pixels_clamped = add_pixels_clamped_mlib;
}