56f5a9a060
Jeff Muizelaar posted some changes to the idct/reconstruction c code. This is the equivalent update for the arm assembly. This shows a good boost on v6, and a minor boost on neon. Here are some numbers for highway in qcif, 2641 frames: HEAD neon: ~161 fps new neon: ~162 fps HEAD v6: ~102 fps new v6: ~106 fps The following functions have been updated for armv6 and neon: vp8_dc_only_idct_add vp8_dequant_idct_add vp8_dequant_dc_idct_add Conflicts: vp8/decoder/arm/armv6/dequantdcidct_v6.asm vp8/decoder/arm/armv6/dequantidct_v6.asm Resolved by removing these files. When I rewrote the functions, I also moved the files to dequant_dc_idct_v6.asm/dequant_idct_v6.asm Change-Id: Ie3300df824d52474eca1a5134cf22d8b7809a5d4
72 lines
2.0 KiB
C
72 lines
2.0 KiB
C
/*
|
|
* Copyright (c) 2010 The VP8 project authors. All Rights Reserved.
|
|
*
|
|
* Use of this source code is governed by a BSD-style license
|
|
* that can be found in the LICENSE file in the root of the source
|
|
* tree. An additional intellectual property rights grant can be found
|
|
* in the file PATENTS. All contributing project authors may
|
|
* be found in the AUTHORS file in the root of the source tree.
|
|
*/
|
|
|
|
|
|
#ifndef DEQUANTIZE_H
|
|
#define DEQUANTIZE_H
|
|
#include "blockd.h"
|
|
|
|
#define prototype_dequant_block(sym) \
|
|
void sym(BLOCKD *x)
|
|
|
|
#define prototype_dequant_idct_add(sym) \
|
|
void sym(short *input, short *dq, \
|
|
unsigned char *pred, unsigned char *output, \
|
|
int pitch, int stride)
|
|
|
|
#define prototype_dequant_dc_idct_add(sym) \
|
|
void sym(short *input, short *dq, \
|
|
unsigned char *pred, unsigned char *output, \
|
|
int pitch, int stride, \
|
|
int dc)
|
|
|
|
#if ARCH_X86 || ARCH_X86_64
|
|
#include "x86/dequantize_x86.h"
|
|
#endif
|
|
|
|
#if ARCH_ARM
|
|
#include "arm/dequantize_arm.h"
|
|
#endif
|
|
|
|
#ifndef vp8_dequant_block
|
|
#define vp8_dequant_block vp8_dequantize_b_c
|
|
#endif
|
|
extern prototype_dequant_block(vp8_dequant_block);
|
|
|
|
#ifndef vp8_dequant_idct_add
|
|
#define vp8_dequant_idct_add vp8_dequant_idct_add_c
|
|
#endif
|
|
extern prototype_dequant_idct_add(vp8_dequant_idct_add);
|
|
|
|
#ifndef vp8_dequant_dc_idct_add
|
|
#define vp8_dequant_dc_idct_add vp8_dequant_dc_idct_add_c
|
|
#endif
|
|
extern prototype_dequant_dc_idct_add(vp8_dequant_dc_idct_add);
|
|
|
|
typedef prototype_dequant_block((*vp8_dequant_block_fn_t));
|
|
|
|
typedef prototype_dequant_idct_add((*vp8_dequant_idct_add_fn_t));
|
|
typedef prototype_dequant_dc_idct_add((*vp8_dequant_dc_idct_add_fn_t));
|
|
|
|
typedef struct
|
|
{
|
|
vp8_dequant_block_fn_t block;
|
|
vp8_dequant_idct_add_fn_t idct_add;
|
|
vp8_dequant_dc_idct_add_fn_t dc_idct_add;
|
|
} vp8_dequant_rtcd_vtable_t;
|
|
|
|
#if CONFIG_RUNTIME_CPU_DETECT
|
|
#define DEQUANT_INVOKE(ctx,fn) (ctx)->fn
|
|
#else
|
|
#define DEQUANT_INVOKE(ctx,fn) vp8_dequant_##fn
|
|
#endif
|
|
|
|
#endif
|