mirror of https://github.com/FFmpeg/FFmpeg.git
* qatar/master: pcm-mpeg: convert to bytestream2 API Revert "h264: clear trailing bits in partially parsed NAL units" remove iwmmxt optimizations mimic: do not continue if swap_buf_size is 0 mimic: convert to bytestream2 API frwu: use MKTAG to check marker instead of AV_RL32 txd: port to bytestream2 API c93: convert to bytestream2 API iff: make .long_name more descriptive FATE: add test for cdxl demuxer rtsp: Fix a typo Conflicts: libavcodec/arm/dsputil_iwmmxt.c libavcodec/arm/dsputil_iwmmxt_rnd_template.c libavcodec/arm/mpegvideo_iwmmxt.c libavcodec/c93.c libavcodec/txd.c libavutil/arm/cpu.c tests/fate/demux.mak Merged-by: Michael Niedermayer <michaelni@gmx.at>pull/3/merge
Michael Niedermayer
13 years ago
commit
b25a265a5c
22 changed files with 142 additions and 1623 deletions
@ -1,210 +0,0 @@ |
||||
/*
|
||||
* iWMMXt optimized DSP utils |
||||
* Copyright (c) 2004 AGAWA Koji |
||||
* |
||||
* This file is part of FFmpeg. |
||||
* |
||||
* FFmpeg 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. |
||||
* |
||||
* FFmpeg 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 FFmpeg; if not, write to the Free Software |
||||
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
||||
*/ |
||||
|
||||
#include "libavutil/cpu.h" |
||||
#include "libavcodec/dsputil.h" |
||||
|
||||
#define DEF(x, y) x ## _no_rnd_ ## y ##_iwmmxt |
||||
#define SET_RND(regd) __asm__ volatile ("mov r12, #1 \n\t tbcsth " #regd ", r12":::"r12"); |
||||
#define WAVG2B "wavg2b" |
||||
#include "dsputil_iwmmxt_rnd_template.c" |
||||
#undef DEF |
||||
#undef SET_RND |
||||
#undef WAVG2B |
||||
|
||||
#define DEF(x, y) x ## _ ## y ##_iwmmxt |
||||
#define SET_RND(regd) __asm__ volatile ("mov r12, #2 \n\t tbcsth " #regd ", r12":::"r12"); |
||||
#define WAVG2B "wavg2br" |
||||
#include "dsputil_iwmmxt_rnd_template.c" |
||||
#undef DEF |
||||
#undef SET_RND |
||||
#undef WAVG2BR |
||||
|
||||
// need scheduling
|
||||
#define OP(AVG) \ |
||||
__asm__ volatile ( \
|
||||
/* alignment */ \
|
||||
"and r12, %[pixels], #7 \n\t" \
|
||||
"bic %[pixels], %[pixels], #7 \n\t" \
|
||||
"tmcr wcgr1, r12 \n\t" \
|
||||
\
|
||||
"wldrd wr0, [%[pixels]] \n\t" \
|
||||
"wldrd wr1, [%[pixels], #8] \n\t" \
|
||||
"add %[pixels], %[pixels], %[line_size] \n\t" \
|
||||
"walignr1 wr4, wr0, wr1 \n\t" \
|
||||
\
|
||||
"1: \n\t" \
|
||||
\
|
||||
"wldrd wr2, [%[pixels]] \n\t" \
|
||||
"wldrd wr3, [%[pixels], #8] \n\t" \
|
||||
"add %[pixels], %[pixels], %[line_size] \n\t" \
|
||||
"pld [%[pixels]] \n\t" \
|
||||
"walignr1 wr5, wr2, wr3 \n\t" \
|
||||
AVG " wr6, wr4, wr5 \n\t" \
|
||||
"wstrd wr6, [%[block]] \n\t" \
|
||||
"add %[block], %[block], %[line_size] \n\t" \
|
||||
\
|
||||
"wldrd wr0, [%[pixels]] \n\t" \
|
||||
"wldrd wr1, [%[pixels], #8] \n\t" \
|
||||
"add %[pixels], %[pixels], %[line_size] \n\t" \
|
||||
"walignr1 wr4, wr0, wr1 \n\t" \
|
||||
"pld [%[pixels]] \n\t" \
|
||||
AVG " wr6, wr4, wr5 \n\t" \
|
||||
"wstrd wr6, [%[block]] \n\t" \
|
||||
"add %[block], %[block], %[line_size] \n\t" \
|
||||
\
|
||||
"subs %[h], %[h], #2 \n\t" \
|
||||
"bne 1b \n\t" \
|
||||
: [block]"+r"(block), [pixels]"+r"(pixels), [h]"+r"(h) \
|
||||
: [line_size]"r"(line_size) \
|
||||
: "memory", "r12"); |
||||
void put_pixels8_y2_iwmmxt(uint8_t *block, const uint8_t *pixels, const int line_size, int h) |
||||
{ |
||||
OP("wavg2br"); |
||||
} |
||||
void put_no_rnd_pixels8_y2_iwmmxt(uint8_t *block, const uint8_t *pixels, const int line_size, int h) |
||||
{ |
||||
OP("wavg2b"); |
||||
} |
||||
#undef OP |
||||
|
||||
void add_pixels_clamped_iwmmxt(const DCTELEM *block, uint8_t *pixels, int line_size) |
||||
{ |
||||
uint8_t *pixels2 = pixels + line_size; |
||||
|
||||
__asm__ volatile ( |
||||
"mov r12, #4 \n\t" |
||||
"1: \n\t" |
||||
"pld [%[pixels], %[line_size2]] \n\t" |
||||
"pld [%[pixels2], %[line_size2]] \n\t" |
||||
"wldrd wr4, [%[pixels]] \n\t" |
||||
"wldrd wr5, [%[pixels2]] \n\t" |
||||
"pld [%[block], #32] \n\t" |
||||
"wunpckelub wr6, wr4 \n\t" |
||||
"wldrd wr0, [%[block]] \n\t" |
||||
"wunpckehub wr7, wr4 \n\t" |
||||
"wldrd wr1, [%[block], #8] \n\t" |
||||
"wunpckelub wr8, wr5 \n\t" |
||||
"wldrd wr2, [%[block], #16] \n\t" |
||||
"wunpckehub wr9, wr5 \n\t" |
||||
"wldrd wr3, [%[block], #24] \n\t" |
||||
"add %[block], %[block], #32 \n\t" |
||||
"waddhss wr10, wr0, wr6 \n\t" |
||||
"waddhss wr11, wr1, wr7 \n\t" |
||||
"waddhss wr12, wr2, wr8 \n\t" |
||||
"waddhss wr13, wr3, wr9 \n\t" |
||||
"wpackhus wr14, wr10, wr11 \n\t" |
||||
"wpackhus wr15, wr12, wr13 \n\t" |
||||
"wstrd wr14, [%[pixels]] \n\t" |
||||
"add %[pixels], %[pixels], %[line_size2] \n\t" |
||||
"subs r12, r12, #1 \n\t" |
||||
"wstrd wr15, [%[pixels2]] \n\t" |
||||
"add %[pixels2], %[pixels2], %[line_size2] \n\t" |
||||
"bne 1b \n\t" |
||||
: [block]"+r"(block), [pixels]"+r"(pixels), [pixels2]"+r"(pixels2) |
||||
: [line_size2]"r"(line_size << 1) |
||||
: "cc", "memory", "r12"); |
||||
} |
||||
|
||||
static void clear_blocks_iwmmxt(DCTELEM *blocks) |
||||
{ |
||||
__asm__ volatile( |
||||
"wzero wr0 \n\t" |
||||
"mov r1, #(128 * 6 / 32) \n\t" |
||||
"1: \n\t" |
||||
"wstrd wr0, [%0] \n\t" |
||||
"wstrd wr0, [%0, #8] \n\t" |
||||
"wstrd wr0, [%0, #16] \n\t" |
||||
"wstrd wr0, [%0, #24] \n\t" |
||||
"subs r1, r1, #1 \n\t" |
||||
"add %0, %0, #32 \n\t" |
||||
"bne 1b \n\t" |
||||
: "+r"(blocks) |
||||
: |
||||
: "r1" |
||||
); |
||||
} |
||||
|
||||
static void nop(uint8_t *block, const uint8_t *pixels, int line_size, int h) |
||||
{ |
||||
return; |
||||
} |
||||
|
||||
/* A run time test is not simple. If this file is compiled in
|
||||
* then we should install the functions |
||||
*/ |
||||
|
||||
void ff_dsputil_init_iwmmxt(DSPContext* c, AVCodecContext *avctx) |
||||
{ |
||||
int mm_flags = AV_CPU_FLAG_IWMMXT; /* multimedia extension flags */ |
||||
const int high_bit_depth = avctx->bits_per_raw_sample > 8; |
||||
|
||||
if (avctx->dsp_mask) { |
||||
if (avctx->dsp_mask & AV_CPU_FLAG_FORCE) |
||||
mm_flags |= (avctx->dsp_mask & 0xffff); |
||||
else |
||||
mm_flags &= ~(avctx->dsp_mask & 0xffff); |
||||
} |
||||
|
||||
if (!(mm_flags & AV_CPU_FLAG_IWMMXT)) return; |
||||
|
||||
c->add_pixels_clamped = add_pixels_clamped_iwmmxt; |
||||
|
||||
if (!high_bit_depth) { |
||||
c->clear_blocks = clear_blocks_iwmmxt; |
||||
|
||||
c->put_pixels_tab[0][0] = put_pixels16_iwmmxt; |
||||
c->put_pixels_tab[0][1] = put_pixels16_x2_iwmmxt; |
||||
c->put_pixels_tab[0][2] = put_pixels16_y2_iwmmxt; |
||||
c->put_pixels_tab[0][3] = put_pixels16_xy2_iwmmxt; |
||||
c->put_no_rnd_pixels_tab[0][0] = put_pixels16_iwmmxt; |
||||
c->put_no_rnd_pixels_tab[0][1] = put_no_rnd_pixels16_x2_iwmmxt; |
||||
c->put_no_rnd_pixels_tab[0][2] = put_no_rnd_pixels16_y2_iwmmxt; |
||||
c->put_no_rnd_pixels_tab[0][3] = put_no_rnd_pixels16_xy2_iwmmxt; |
||||
|
||||
c->put_pixels_tab[1][0] = put_pixels8_iwmmxt; |
||||
c->put_pixels_tab[1][1] = put_pixels8_x2_iwmmxt; |
||||
c->put_pixels_tab[1][2] = put_pixels8_y2_iwmmxt; |
||||
c->put_pixels_tab[1][3] = put_pixels8_xy2_iwmmxt; |
||||
c->put_no_rnd_pixels_tab[1][0] = put_pixels8_iwmmxt; |
||||
c->put_no_rnd_pixels_tab[1][1] = put_no_rnd_pixels8_x2_iwmmxt; |
||||
c->put_no_rnd_pixels_tab[1][2] = put_no_rnd_pixels8_y2_iwmmxt; |
||||
c->put_no_rnd_pixels_tab[1][3] = put_no_rnd_pixels8_xy2_iwmmxt; |
||||
|
||||
c->avg_pixels_tab[0][0] = avg_pixels16_iwmmxt; |
||||
c->avg_pixels_tab[0][1] = avg_pixels16_x2_iwmmxt; |
||||
c->avg_pixels_tab[0][2] = avg_pixels16_y2_iwmmxt; |
||||
c->avg_pixels_tab[0][3] = avg_pixels16_xy2_iwmmxt; |
||||
c->avg_no_rnd_pixels_tab[0][0] = avg_pixels16_iwmmxt; |
||||
c->avg_no_rnd_pixels_tab[0][1] = avg_no_rnd_pixels16_x2_iwmmxt; |
||||
c->avg_no_rnd_pixels_tab[0][2] = avg_no_rnd_pixels16_y2_iwmmxt; |
||||
c->avg_no_rnd_pixels_tab[0][3] = avg_no_rnd_pixels16_xy2_iwmmxt; |
||||
|
||||
c->avg_pixels_tab[1][0] = avg_pixels8_iwmmxt; |
||||
c->avg_pixels_tab[1][1] = avg_pixels8_x2_iwmmxt; |
||||
c->avg_pixels_tab[1][2] = avg_pixels8_y2_iwmmxt; |
||||
c->avg_pixels_tab[1][3] = avg_pixels8_xy2_iwmmxt; |
||||
c->avg_no_rnd_pixels_tab[1][0] = avg_no_rnd_pixels8_iwmmxt; |
||||
c->avg_no_rnd_pixels_tab[1][1] = avg_no_rnd_pixels8_x2_iwmmxt; |
||||
c->avg_no_rnd_pixels_tab[1][2] = avg_no_rnd_pixels8_y2_iwmmxt; |
||||
c->avg_no_rnd_pixels_tab[1][3] = avg_no_rnd_pixels8_xy2_iwmmxt; |
||||
} |
||||
} |
||||
File diff suppressed because it is too large
Load Diff
@ -1,101 +0,0 @@ |
||||
/*
|
||||
* copyright (c) 2004 AGAWA Koji |
||||
* |
||||
* This file is part of FFmpeg. |
||||
* |
||||
* FFmpeg 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. |
||||
* |
||||
* FFmpeg 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 FFmpeg; if not, write to the Free Software |
||||
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
||||
*/ |
||||
|
||||
#include "libavutil/cpu.h" |
||||
#include "libavcodec/avcodec.h" |
||||
#include "libavcodec/dsputil.h" |
||||
#include "libavcodec/mpegvideo.h" |
||||
#include "mpegvideo_arm.h" |
||||
|
||||
static void dct_unquantize_h263_intra_iwmmxt(MpegEncContext *s, |
||||
DCTELEM *block, int n, int qscale) |
||||
{ |
||||
int level, qmul, qadd; |
||||
int nCoeffs; |
||||
DCTELEM *block_orig = block; |
||||
|
||||
assert(s->block_last_index[n]>=0); |
||||
|
||||
qmul = qscale << 1; |
||||
|
||||
if (!s->h263_aic) { |
||||
if (n < 4) |
||||
level = block[0] * s->y_dc_scale; |
||||
else |
||||
level = block[0] * s->c_dc_scale; |
||||
qadd = (qscale - 1) | 1; |
||||
}else{ |
||||
qadd = 0; |
||||
level = block[0]; |
||||
} |
||||
if(s->ac_pred) |
||||
nCoeffs=63; |
||||
else |
||||
nCoeffs= s->inter_scantable.raster_end[ s->block_last_index[n] ]; |
||||
|
||||
__asm__ volatile ( |
||||
/* "movd %1, %%mm6 \n\t" //qmul */ |
||||
/* "packssdw %%mm6, %%mm6 \n\t" */ |
||||
/* "packssdw %%mm6, %%mm6 \n\t" */ |
||||
"tbcsth wr6, %[qmul] \n\t" |
||||
/* "movd %2, %%mm5 \n\t" //qadd */ |
||||
/* "packssdw %%mm5, %%mm5 \n\t" */ |
||||
/* "packssdw %%mm5, %%mm5 \n\t" */ |
||||
"tbcsth wr5, %[qadd] \n\t" |
||||
"wzero wr7 \n\t" /* "pxor %%mm7, %%mm7 \n\t" */ |
||||
"wzero wr4 \n\t" /* "pxor %%mm4, %%mm4 \n\t" */ |
||||
"wsubh wr7, wr5, wr7 \n\t" /* "psubw %%mm5, %%mm7 \n\t" */ |
||||
"1: \n\t" |
||||
"wldrd wr2, [%[block]] \n\t" /* "movq (%0, %3), %%mm0 \n\t" */ |
||||
"wldrd wr3, [%[block], #8] \n\t" /* "movq 8(%0, %3), %%mm1 \n\t" */ |
||||
"wmulsl wr0, wr6, wr2 \n\t" /* "pmullw %%mm6, %%mm0 \n\t" */ |
||||
"wmulsl wr1, wr6, wr3 \n\t" /* "pmullw %%mm6, %%mm1 \n\t" */ |
||||
/* "movq (%0, %3), %%mm2 \n\t" */ |
||||
/* "movq 8(%0, %3), %%mm3 \n\t" */ |
||||
"wcmpgtsh wr2, wr4, wr2 \n\t" /* "pcmpgtw %%mm4, %%mm2 \n\t" // block[i] < 0 ? -1 : 0 */ |
||||
"wcmpgtsh wr3, wr4, wr2 \n\t" /* "pcmpgtw %%mm4, %%mm3 \n\t" // block[i] < 0 ? -1 : 0 */ |
||||
"wxor wr0, wr2, wr0 \n\t" /* "pxor %%mm2, %%mm0 \n\t" */ |
||||
"wxor wr1, wr3, wr1 \n\t" /* "pxor %%mm3, %%mm1 \n\t" */ |
||||
"waddh wr0, wr7, wr0 \n\t" /* "paddw %%mm7, %%mm0 \n\t" */ |
||||
"waddh wr1, wr7, wr1 \n\t" /* "paddw %%mm7, %%mm1 \n\t" */ |
||||
"wxor wr2, wr0, wr2 \n\t" /* "pxor %%mm0, %%mm2 \n\t" */ |
||||
"wxor wr3, wr1, wr3 \n\t" /* "pxor %%mm1, %%mm3 \n\t" */ |
||||
"wcmpeqh wr0, wr7, wr0 \n\t" /* "pcmpeqw %%mm7, %%mm0 \n\t" // block[i] == 0 ? -1 : 0 */ |
||||
"wcmpeqh wr1, wr7, wr1 \n\t" /* "pcmpeqw %%mm7, %%mm1 \n\t" // block[i] == 0 ? -1 : 0 */ |
||||
"wandn wr0, wr2, wr0 \n\t" /* "pandn %%mm2, %%mm0 \n\t" */ |
||||
"wandn wr1, wr3, wr1 \n\t" /* "pandn %%mm3, %%mm1 \n\t" */ |
||||
"wstrd wr0, [%[block]] \n\t" /* "movq %%mm0, (%0, %3) \n\t" */ |
||||
"wstrd wr1, [%[block], #8] \n\t" /* "movq %%mm1, 8(%0, %3) \n\t" */ |
||||
"add %[block], %[block], #16 \n\t" /* "addl $16, %3 \n\t" */ |
||||
"subs %[i], %[i], #1 \n\t" |
||||
"bne 1b \n\t" /* "jng 1b \n\t" */ |
||||
:[block]"+r"(block) |
||||
:[i]"r"((nCoeffs + 8) / 8), [qmul]"r"(qmul), [qadd]"r"(qadd) |
||||
:"memory"); |
||||
|
||||
block_orig[0] = level; |
||||
} |
||||
|
||||
void ff_MPV_common_init_iwmmxt(MpegEncContext *s) |
||||
{ |
||||
if (!(mm_flags & AV_CPU_FLAG_IWMMXT)) return; |
||||
|
||||
s->dct_unquantize_h263_intra = dct_unquantize_h263_intra_iwmmxt; |
||||
} |
||||
@ -1,25 +0,0 @@ |
||||
/*
|
||||
* This file is part of FFmpeg. |
||||
* |
||||
* FFmpeg 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. |
||||
* |
||||
* FFmpeg 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 FFmpeg; if not, write to the Free Software |
||||
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
||||
*/ |
||||
|
||||
#include "libavutil/cpu.h" |
||||
#include "config.h" |
||||
|
||||
int ff_get_cpu_flags_arm(void) |
||||
{ |
||||
return HAVE_IWMMXT * AV_CPU_FLAG_IWMMXT; |
||||
} |
||||
Loading…
Reference in new issue