mirror of https://git.ffmpeg.org/ffmpeg.git
lavc/vp8dsp: R-V put_vp8_pixels
C908: vp8_put_pixels4_c: 78.0 vp8_put_pixels4_rvi: 33.7 vp8_put_pixels8_c: 278.0 vp8_put_pixels8_rvi: 55.0 vp8_put_pixels16_c: 999.0 vp8_put_pixels16_rvi: 86.7 Signed-off-by: Rémi Denis-Courmont <remi@remlab.net>
This commit is contained in:
parent
d7924a4f60
commit
0b8e5e5a00
|
@ -59,6 +59,7 @@ RVV-OBJS-$(CONFIG_UTVIDEO_DECODER) += riscv/utvideodsp_rvv.o
|
|||
OBJS-$(CONFIG_VC1DSP) += riscv/vc1dsp_init.o
|
||||
RVV-OBJS-$(CONFIG_VC1DSP) += riscv/vc1dsp_rvv.o
|
||||
OBJS-$(CONFIG_VP8DSP) += riscv/vp8dsp_init.o
|
||||
RV-OBJS-$(CONFIG_VP8DSP) += riscv/vp8dsp_rvi.o
|
||||
RVV-OBJS-$(CONFIG_VP8DSP) += riscv/vp8dsp_rvv.o
|
||||
OBJS-$(CONFIG_VP9_DECODER) += riscv/vp9dsp_init.o
|
||||
RVV-OBJS-$(CONFIG_VP9_DECODER) += riscv/vp9_intra_rvv.o
|
||||
|
|
|
@ -0,0 +1,75 @@
|
|||
/*
|
||||
* 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
|
||||
*/
|
||||
|
||||
#ifndef AVCODEC_RISCV_VP8DSP_H
|
||||
#define AVCODEC_RISCV_VP8DSP_H
|
||||
|
||||
#include "libavcodec/vp8dsp.h"
|
||||
|
||||
#define VP8_LF_Y(hv, inner, opt) \
|
||||
void ff_vp8_##hv##_loop_filter16##inner##_##opt(uint8_t *dst, \
|
||||
ptrdiff_t stride, \
|
||||
int flim_E, int flim_I, \
|
||||
int hev_thresh)
|
||||
|
||||
#define VP8_LF_UV(hv, inner, opt) \
|
||||
void ff_vp8_##hv##_loop_filter8uv##inner##_##opt(uint8_t *dstU, \
|
||||
uint8_t *dstV, \
|
||||
ptrdiff_t stride, \
|
||||
int flim_E, int flim_I, \
|
||||
int hev_thresh)
|
||||
|
||||
#define VP8_LF_SIMPLE(hv, opt) \
|
||||
void ff_vp8_##hv##_loop_filter16_simple_##opt(uint8_t *dst, \
|
||||
ptrdiff_t stride, \
|
||||
int flim)
|
||||
|
||||
#define VP8_LF_HV(inner, opt) \
|
||||
VP8_LF_Y(h, inner, opt); \
|
||||
VP8_LF_Y(v, inner, opt); \
|
||||
VP8_LF_UV(h, inner, opt); \
|
||||
VP8_LF_UV(v, inner, opt)
|
||||
|
||||
#define VP8_LF(opt) \
|
||||
VP8_LF_HV(, opt); \
|
||||
VP8_LF_HV(_inner, opt); \
|
||||
VP8_LF_SIMPLE(h, opt); \
|
||||
VP8_LF_SIMPLE(v, opt)
|
||||
|
||||
#define VP8_MC(n, opt) \
|
||||
void ff_put_vp8_##n##_##opt(uint8_t *dst, ptrdiff_t dststride, \
|
||||
const uint8_t *src, ptrdiff_t srcstride,\
|
||||
int h, int x, int y)
|
||||
|
||||
#define VP8_EPEL(w, opt) \
|
||||
VP8_MC(pixels ## w, opt); \
|
||||
VP8_MC(epel ## w ## _h4, opt); \
|
||||
VP8_MC(epel ## w ## _h6, opt); \
|
||||
VP8_MC(epel ## w ## _v4, opt); \
|
||||
VP8_MC(epel ## w ## _h4v4, opt); \
|
||||
VP8_MC(epel ## w ## _h6v4, opt); \
|
||||
VP8_MC(epel ## w ## _v6, opt); \
|
||||
VP8_MC(epel ## w ## _h4v6, opt); \
|
||||
VP8_MC(epel ## w ## _h6v6, opt)
|
||||
|
||||
#define VP8_BILIN(w, opt) \
|
||||
VP8_MC(bilin ## w ## _h, opt); \
|
||||
VP8_MC(bilin ## w ## _v, opt); \
|
||||
VP8_MC(bilin ## w ## _hv, opt)
|
||||
|
||||
#endif /* AVCODEC_RISCV_VP8DSP_H */
|
|
@ -24,11 +24,33 @@
|
|||
#include "libavutil/cpu.h"
|
||||
#include "libavutil/riscv/cpu.h"
|
||||
#include "libavcodec/vp8dsp.h"
|
||||
#include "vp8dsp.h"
|
||||
|
||||
void ff_vp8_idct_dc_add_rvv(uint8_t *dst, int16_t block[16], ptrdiff_t stride);
|
||||
void ff_vp8_idct_dc_add4y_rvv(uint8_t *dst, int16_t block[4][16], ptrdiff_t stride);
|
||||
void ff_vp8_idct_dc_add4uv_rvv(uint8_t *dst, int16_t block[4][16], ptrdiff_t stride);
|
||||
|
||||
VP8_EPEL(16, rvi);
|
||||
VP8_EPEL(8, rvi);
|
||||
VP8_EPEL(4, rvi);
|
||||
|
||||
av_cold void ff_vp78dsp_init_riscv(VP8DSPContext *c)
|
||||
{
|
||||
#if HAVE_RV
|
||||
int flags = av_get_cpu_flags();
|
||||
if (flags & AV_CPU_FLAG_RVI) {
|
||||
#if __riscv_xlen >= 64
|
||||
c->put_vp8_epel_pixels_tab[0][0][0] = ff_put_vp8_pixels16_rvi;
|
||||
c->put_vp8_epel_pixels_tab[1][0][0] = ff_put_vp8_pixels8_rvi;
|
||||
c->put_vp8_bilinear_pixels_tab[0][0][0] = ff_put_vp8_pixels16_rvi;
|
||||
c->put_vp8_bilinear_pixels_tab[1][0][0] = ff_put_vp8_pixels8_rvi;
|
||||
#endif
|
||||
c->put_vp8_epel_pixels_tab[2][0][0] = ff_put_vp8_pixels4_rvi;
|
||||
c->put_vp8_bilinear_pixels_tab[2][0][0] = ff_put_vp8_pixels4_rvi;
|
||||
}
|
||||
#endif
|
||||
}
|
||||
|
||||
av_cold void ff_vp8dsp_init_riscv(VP8DSPContext *c)
|
||||
{
|
||||
#if HAVE_RVV
|
||||
|
|
|
@ -0,0 +1,61 @@
|
|||
/*
|
||||
* Copyright (c) 2024 Institue of Software Chinese Academy of Sciences (ISCAS).
|
||||
*
|
||||
* 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/riscv/asm.S"
|
||||
|
||||
#if __riscv_xlen >= 64
|
||||
func ff_put_vp8_pixels16_rvi
|
||||
1:
|
||||
addi a4, a4, -1
|
||||
ld t0, (a2)
|
||||
ld t1, 8(a2)
|
||||
sd t0, (a0)
|
||||
sd t1, 8(a0)
|
||||
add a2, a2, a3
|
||||
add a0, a0, a1
|
||||
bnez a4, 1b
|
||||
|
||||
ret
|
||||
endfunc
|
||||
|
||||
func ff_put_vp8_pixels8_rvi
|
||||
1:
|
||||
addi a4, a4, -1
|
||||
ld t0, (a2)
|
||||
sd t0, (a0)
|
||||
add a2, a2, a3
|
||||
add a0, a0, a1
|
||||
bnez a4, 1b
|
||||
|
||||
ret
|
||||
endfunc
|
||||
#endif
|
||||
|
||||
func ff_put_vp8_pixels4_rvi
|
||||
1:
|
||||
addi a4, a4, -1
|
||||
lw t0, (a2)
|
||||
sw t0, (a0)
|
||||
add a2, a2, a3
|
||||
add a0, a0, a1
|
||||
bnez a4, 1b
|
||||
|
||||
ret
|
||||
endfunc
|
|
@ -681,6 +681,8 @@ av_cold void ff_vp78dsp_init(VP8DSPContext *dsp)
|
|||
ff_vp78dsp_init_arm(dsp);
|
||||
#elif ARCH_PPC
|
||||
ff_vp78dsp_init_ppc(dsp);
|
||||
#elif ARCH_RISCV
|
||||
ff_vp78dsp_init_riscv(dsp);
|
||||
#elif ARCH_X86
|
||||
ff_vp78dsp_init_x86(dsp);
|
||||
#endif
|
||||
|
|
|
@ -87,6 +87,7 @@ void ff_vp78dsp_init(VP8DSPContext *c);
|
|||
void ff_vp78dsp_init_aarch64(VP8DSPContext *c);
|
||||
void ff_vp78dsp_init_arm(VP8DSPContext *c);
|
||||
void ff_vp78dsp_init_ppc(VP8DSPContext *c);
|
||||
void ff_vp78dsp_init_riscv(VP8DSPContext *c);
|
||||
void ff_vp78dsp_init_x86(VP8DSPContext *c);
|
||||
|
||||
void ff_vp8dsp_init(VP8DSPContext *c);
|
||||
|
|
Loading…
Reference in New Issue