view arm/h264pred_init_arm.c @ 12492:58a960d6e34c libavcodec

Rename h264_idct_sse2.asm to h264_idct.asm; move inline IDCT asm from h264dsp_mmx.c to h264_idct.asm (as yasm code). Because the loops are now coded in asm instead of C, this is (depending on the function) up to 50% faster for cases where gcc didn't do a great job at looping. Since h264_idct_add8() is now faster than the manual loop setup in h264.c, in-asm idct calling can now be enabled for chroma as well (see r16207). For MMX, this is 5% faster. For SSE2 (which isn't done for chroma if h264.c does the looping), this makes it up to 50% faster. Speed gain overall is ~0.5-1.0%.

/*
 * Copyright (c) 2009 Mans Rullgard <[e-mail]>
 *
 * 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 <stdint.h>

#include "libavcodec/h264pred.h"

void ff_pred16x16_vert_neon(uint8_t *src, int stride);
void ff_pred16x16_hor_neon(uint8_t *src, int stride);
void ff_pred16x16_plane_neon(uint8_t *src, int stride);
void ff_pred16x16_dc_neon(uint8_t *src, int stride);
void ff_pred16x16_128_dc_neon(uint8_t *src, int stride);
void ff_pred16x16_left_dc_neon(uint8_t *src, int stride);
void ff_pred16x16_top_dc_neon(uint8_t *src, int stride);

void ff_pred8x8_vert_neon(uint8_t *src, int stride);
void ff_pred8x8_hor_neon(uint8_t *src, int stride);
void ff_pred8x8_plane_neon(uint8_t *src, int stride);
void ff_pred8x8_dc_neon(uint8_t *src, int stride);
void ff_pred8x8_128_dc_neon(uint8_t *src, int stride);
void ff_pred8x8_left_dc_neon(uint8_t *src, int stride);
void ff_pred8x8_top_dc_neon(uint8_t *src, int stride);
void ff_pred8x8_l0t_dc_neon(uint8_t *src, int stride);
void ff_pred8x8_0lt_dc_neon(uint8_t *src, int stride);
void ff_pred8x8_l00_dc_neon(uint8_t *src, int stride);
void ff_pred8x8_0l0_dc_neon(uint8_t *src, int stride);

static void ff_h264_pred_init_neon(H264PredContext *h, int codec_id)
{
    h->pred8x8[VERT_PRED8x8     ] = ff_pred8x8_vert_neon;
    h->pred8x8[HOR_PRED8x8      ] = ff_pred8x8_hor_neon;
    if (codec_id != CODEC_ID_VP8)
        h->pred8x8[PLANE_PRED8x8] = ff_pred8x8_plane_neon;
    h->pred8x8[DC_128_PRED8x8   ] = ff_pred8x8_128_dc_neon;
    if (codec_id != CODEC_ID_RV40 && codec_id != CODEC_ID_VP8) {
        h->pred8x8[DC_PRED8x8     ] = ff_pred8x8_dc_neon;
        h->pred8x8[LEFT_DC_PRED8x8] = ff_pred8x8_left_dc_neon;
        h->pred8x8[TOP_DC_PRED8x8 ] = ff_pred8x8_top_dc_neon;
        h->pred8x8[ALZHEIMER_DC_L0T_PRED8x8] = ff_pred8x8_l0t_dc_neon;
        h->pred8x8[ALZHEIMER_DC_0LT_PRED8x8] = ff_pred8x8_0lt_dc_neon;
        h->pred8x8[ALZHEIMER_DC_L00_PRED8x8] = ff_pred8x8_l00_dc_neon;
        h->pred8x8[ALZHEIMER_DC_0L0_PRED8x8] = ff_pred8x8_0l0_dc_neon;
    }

    h->pred16x16[DC_PRED8x8     ] = ff_pred16x16_dc_neon;
    h->pred16x16[VERT_PRED8x8   ] = ff_pred16x16_vert_neon;
    h->pred16x16[HOR_PRED8x8    ] = ff_pred16x16_hor_neon;
    h->pred16x16[LEFT_DC_PRED8x8] = ff_pred16x16_left_dc_neon;
    h->pred16x16[TOP_DC_PRED8x8 ] = ff_pred16x16_top_dc_neon;
    h->pred16x16[DC_128_PRED8x8 ] = ff_pred16x16_128_dc_neon;
    if (codec_id != CODEC_ID_SVQ3 && codec_id != CODEC_ID_RV40 && codec_id != CODEC_ID_VP8)
        h->pred16x16[PLANE_PRED8x8  ] = ff_pred16x16_plane_neon;
}

void ff_h264_pred_init_arm(H264PredContext *h, int codec_id)
{
    if (HAVE_NEON)    ff_h264_pred_init_neon(h, codec_id);
}