ffmpeg/libavcodec/ppc/vp8dsp_altivec.c

364 lines
12 KiB
C

/*
* VP8 compatible video decoder
*
* Copyright (C) 2010 David Conrad
*
* 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 "config.h"
#include "libavutil/attributes.h"
#include "libavutil/cpu.h"
#include "libavutil/mem_internal.h"
#include "libavutil/ppc/cpu.h"
#include "libavutil/ppc/util_altivec.h"
#include "libavcodec/vp8dsp.h"
#include "hpeldsp_altivec.h"
#if HAVE_ALTIVEC
#define REPT4(...) { __VA_ARGS__, __VA_ARGS__, __VA_ARGS__, __VA_ARGS__ }
// h subpel filter uses msum to multiply+add 4 pixel taps at once
static const vec_s8 h_subpel_filters_inner[7] =
{
REPT4( -6, 123, 12, -1),
REPT4(-11, 108, 36, -8),
REPT4( -9, 93, 50, -6),
REPT4(-16, 77, 77, -16),
REPT4( -6, 50, 93, -9),
REPT4( -8, 36, 108, -11),
REPT4( -1, 12, 123, -6),
};
// for 6tap filters, these are the outer two taps
// The zeros mask off pixels 4-7 when filtering 0-3
// and vice-versa
static const vec_s8 h_subpel_filters_outer[4] =
{
REPT4(0, 0, 2, 1),
REPT4(0, 0, 3, 3),
REPT4(0, 0, 1, 2),
REPT4(0, 0, 0, 0),
};
#define LOAD_H_SUBPEL_FILTER(i) \
vec_s8 filter_inner = h_subpel_filters_inner[i]; \
vec_s8 filter_outerh = h_subpel_filters_outer[(i)>>1]; \
vec_s8 filter_outerl = vec_sld(filter_outerh, filter_outerh, 2)
#if HAVE_BIGENDIAN
#define GET_PIXHL(offset) \
a = vec_ld((offset)-is6tap-1, src); \
b = vec_ld((offset)-is6tap-1+15, src); \
pixh = vec_perm(a, b, permh##offset); \
pixl = vec_perm(a, b, perml##offset)
#define GET_OUTER(offset) outer = vec_perm(a, b, perm_6tap##offset)
#else
#define GET_PIXHL(offset) \
a = vec_vsx_ld((offset)-is6tap-1, src); \
pixh = vec_perm(a, a, perm_inner); \
pixl = vec_perm(a, a, vec_add(perm_inner, vec_splat_u8(4)))
#define GET_OUTER(offset) outer = vec_perm(a, a, perm_outer)
#endif
#define FILTER_H(dstv, off) \
GET_PIXHL(off); \
filth = vec_msum(filter_inner, pixh, c64); \
filtl = vec_msum(filter_inner, pixl, c64); \
\
if (is6tap) { \
GET_OUTER(off); \
filth = vec_msum(filter_outerh, outer, filth); \
filtl = vec_msum(filter_outerl, outer, filtl); \
} \
if (w == 4) \
filtl = filth; /* discard pixels 4-7 */ \
dstv = vec_packs(filth, filtl); \
dstv = vec_sra(dstv, c7)
static av_always_inline
void put_vp8_epel_h_altivec_core(uint8_t *dst, ptrdiff_t dst_stride,
const uint8_t *src, ptrdiff_t src_stride,
int h, int mx, int w, int is6tap)
{
LOAD_H_SUBPEL_FILTER(mx-1);
#if HAVE_BIGENDIAN
vec_u8 align_vec0, align_vec8, permh0, permh8;
vec_u8 perm_6tap0, perm_6tap8, perml0, perml8;
vec_u8 b;
#endif
vec_u8 filt, a, pixh, pixl, outer;
vec_s16 f16h, f16l;
vec_s32 filth, filtl;
vec_u8 perm_inner6 = { 1,2,3,4, 2,3,4,5, 3,4,5,6, 4,5,6,7 };
vec_u8 perm_inner4 = { 0,1,2,3, 1,2,3,4, 2,3,4,5, 3,4,5,6 };
vec_u8 perm_inner = is6tap ? perm_inner6 : perm_inner4;
vec_u8 perm_outer = { 4,9, 0,5, 5,10, 1,6, 6,11, 2,7, 7,12, 3,8 };
vec_s32 c64 = vec_sl(vec_splat_s32(1), vec_splat_u32(6));
vec_u16 c7 = vec_splat_u16(7);
#if HAVE_BIGENDIAN
align_vec0 = vec_lvsl( -is6tap-1, src);
align_vec8 = vec_lvsl(8-is6tap-1, src);
permh0 = vec_perm(align_vec0, align_vec0, perm_inner);
permh8 = vec_perm(align_vec8, align_vec8, perm_inner);
perm_inner = vec_add(perm_inner, vec_splat_u8(4));
perml0 = vec_perm(align_vec0, align_vec0, perm_inner);
perml8 = vec_perm(align_vec8, align_vec8, perm_inner);
perm_6tap0 = vec_perm(align_vec0, align_vec0, perm_outer);
perm_6tap8 = vec_perm(align_vec8, align_vec8, perm_outer);
#endif
while (h --> 0) {
FILTER_H(f16h, 0);
if (w == 16) {
FILTER_H(f16l, 8);
filt = vec_packsu(f16h, f16l);
vec_st(filt, 0, dst);
} else {
filt = vec_packsu(f16h, f16h);
vec_ste((vec_u32)filt, 0, (uint32_t*)dst);
if (w == 8)
vec_ste((vec_u32)filt, 4, (uint32_t*)dst);
}
src += src_stride;
dst += dst_stride;
}
}
// v subpel filter does a simple vertical multiply + add
static const vec_u8 v_subpel_filters[7] =
{
{ 0, 6, 123, 12, 1, 0 },
{ 2, 11, 108, 36, 8, 1 },
{ 0, 9, 93, 50, 6, 0 },
{ 3, 16, 77, 77, 16, 3 },
{ 0, 6, 50, 93, 9, 0 },
{ 1, 8, 36, 108, 11, 2 },
{ 0, 1, 12, 123, 6, 0 },
};
#define LOAD_V_SUBPEL_FILTER(i) \
vec_u8 subpel_filter = v_subpel_filters[i]; \
vec_u8 f0 = vec_splat(subpel_filter, 0); \
vec_u8 f1 = vec_splat(subpel_filter, 1); \
vec_u8 f2 = vec_splat(subpel_filter, 2); \
vec_u8 f3 = vec_splat(subpel_filter, 3); \
vec_u8 f4 = vec_splat(subpel_filter, 4); \
vec_u8 f5 = vec_splat(subpel_filter, 5)
#define FILTER_V(dstv, vec_mul) \
s1f = (vec_s16)vec_mul(s1, f1); \
s2f = (vec_s16)vec_mul(s2, f2); \
s3f = (vec_s16)vec_mul(s3, f3); \
s4f = (vec_s16)vec_mul(s4, f4); \
s2f = vec_subs(s2f, s1f); \
s3f = vec_subs(s3f, s4f); \
if (is6tap) { \
s0f = (vec_s16)vec_mul(s0, f0); \
s5f = (vec_s16)vec_mul(s5, f5); \
s2f = vec_adds(s2f, s0f); \
s3f = vec_adds(s3f, s5f); \
} \
dstv = vec_adds(s2f, s3f); \
dstv = vec_adds(dstv, c64); \
dstv = vec_sra(dstv, c7)
#if HAVE_BIGENDIAN
#define LOAD_HL(off, s, perm) load_with_perm_vec(off, s, perm)
#else
#define LOAD_HL(off, s, perm) vec_mergeh(vec_vsx_ld(off,s), vec_vsx_ld(off+8,s))
#endif
static av_always_inline
void put_vp8_epel_v_altivec_core(uint8_t *dst, ptrdiff_t dst_stride,
const uint8_t *src, ptrdiff_t src_stride,
int h, int my, int w, int is6tap)
{
LOAD_V_SUBPEL_FILTER(my-1);
vec_u8 s0, s1, s2, s3, s4, s5, filt, align_vech, perm_vec, align_vecl;
vec_s16 s0f, s1f, s2f, s3f, s4f, s5f, f16h, f16l;
vec_s16 c64 = vec_sl(vec_splat_s16(1), vec_splat_u16(6));
vec_u16 c7 = vec_splat_u16(7);
#if HAVE_BIGENDIAN
// we want pixels 0-7 to be in the even positions and 8-15 in the odd,
// so combine this permute with the alignment permute vector
align_vech = vec_lvsl(0, src);
align_vecl = vec_sld(align_vech, align_vech, 8);
if (w ==16)
perm_vec = vec_mergeh(align_vech, align_vecl);
else
perm_vec = vec_mergeh(align_vech, align_vech);
#endif
if (is6tap)
s0 = LOAD_HL(-2*src_stride, src, perm_vec);
s1 = LOAD_HL(-1*src_stride, src, perm_vec);
s2 = LOAD_HL( 0*src_stride, src, perm_vec);
s3 = LOAD_HL( 1*src_stride, src, perm_vec);
if (is6tap)
s4 = LOAD_HL( 2*src_stride, src, perm_vec);
src += (2+is6tap)*src_stride;
while (h --> 0) {
if (is6tap)
s5 = LOAD_HL(0, src, perm_vec);
else
s4 = LOAD_HL(0, src, perm_vec);
FILTER_V(f16h, vec_mule);
if (w == 16) {
FILTER_V(f16l, vec_mulo);
filt = vec_packsu(f16h, f16l);
vec_st(filt, 0, dst);
} else {
filt = vec_packsu(f16h, f16h);
if (w == 4)
filt = (vec_u8)vec_splat((vec_u32)filt, 0);
else
vec_ste((vec_u32)filt, 4, (uint32_t*)dst);
vec_ste((vec_u32)filt, 0, (uint32_t*)dst);
}
if (is6tap)
s0 = s1;
s1 = s2;
s2 = s3;
s3 = s4;
if (is6tap)
s4 = s5;
dst += dst_stride;
src += src_stride;
}
}
#define EPEL_FUNCS(WIDTH, TAPS) \
static av_noinline \
void put_vp8_epel ## WIDTH ## _h ## TAPS ## _altivec(uint8_t *dst, ptrdiff_t dst_stride, const uint8_t *src, ptrdiff_t src_stride, int h, int mx, int my) \
{ \
put_vp8_epel_h_altivec_core(dst, dst_stride, src, src_stride, h, mx, WIDTH, TAPS == 6); \
} \
\
static av_noinline \
void put_vp8_epel ## WIDTH ## _v ## TAPS ## _altivec(uint8_t *dst, ptrdiff_t dst_stride, const uint8_t *src, ptrdiff_t src_stride, int h, int mx, int my) \
{ \
put_vp8_epel_v_altivec_core(dst, dst_stride, src, src_stride, h, my, WIDTH, TAPS == 6); \
}
#define EPEL_HV(WIDTH, HTAPS, VTAPS) \
static void put_vp8_epel ## WIDTH ## _h ## HTAPS ## v ## VTAPS ## _altivec(uint8_t *dst, ptrdiff_t dstride, const uint8_t *src, ptrdiff_t sstride, int h, int mx, int my) \
{ \
DECLARE_ALIGNED(16, uint8_t, tmp)[(2*WIDTH+5)*16]; \
if (VTAPS == 6) { \
put_vp8_epel ## WIDTH ## _h ## HTAPS ## _altivec(tmp, 16, src-2*sstride, sstride, h+5, mx, my); \
put_vp8_epel ## WIDTH ## _v ## VTAPS ## _altivec(dst, dstride, tmp+2*16, 16, h, mx, my); \
} else { \
put_vp8_epel ## WIDTH ## _h ## HTAPS ## _altivec(tmp, 16, src-sstride, sstride, h+4, mx, my); \
put_vp8_epel ## WIDTH ## _v ## VTAPS ## _altivec(dst, dstride, tmp+16, 16, h, mx, my); \
} \
}
EPEL_FUNCS(16,6)
EPEL_FUNCS(8, 6)
EPEL_FUNCS(8, 4)
EPEL_FUNCS(4, 6)
EPEL_FUNCS(4, 4)
EPEL_HV(16, 6,6)
EPEL_HV(8, 6,6)
EPEL_HV(8, 4,6)
EPEL_HV(8, 6,4)
EPEL_HV(8, 4,4)
EPEL_HV(4, 6,6)
EPEL_HV(4, 4,6)
EPEL_HV(4, 6,4)
EPEL_HV(4, 4,4)
static void put_vp8_pixels16_altivec(uint8_t *dst, ptrdiff_t dstride, const uint8_t *src, ptrdiff_t sstride, int h, int mx, int my)
{
register vector unsigned char perm;
int i;
register ptrdiff_t dstride2 = dstride << 1, sstride2 = sstride << 1;
register ptrdiff_t dstride3 = dstride2 + dstride, sstride3 = sstride + sstride2;
register ptrdiff_t dstride4 = dstride << 2, sstride4 = sstride << 2;
#if HAVE_BIGENDIAN
perm = vec_lvsl(0, src);
#endif
// hand-unrolling the loop by 4 gains about 15%
// mininum execution time goes from 74 to 60 cycles
// it's faster than -funroll-loops, but using
// -funroll-loops w/ this is bad - 74 cycles again.
// all this is on a 7450, tuning for the 7450
for (i = 0; i < h; i += 4) {
vec_st(load_with_perm_vec(0, src, perm), 0, dst);
vec_st(load_with_perm_vec(sstride, src, perm), dstride, dst);
vec_st(load_with_perm_vec(sstride2, src, perm), dstride2, dst);
vec_st(load_with_perm_vec(sstride3, src, perm), dstride3, dst);
src += sstride4;
dst += dstride4;
}
}
#endif /* HAVE_ALTIVEC */
av_cold void ff_vp78dsp_init_ppc(VP8DSPContext *c)
{
#if HAVE_ALTIVEC
if (!PPC_ALTIVEC(av_get_cpu_flags()))
return;
c->put_vp8_epel_pixels_tab[0][0][0] = put_vp8_pixels16_altivec;
c->put_vp8_epel_pixels_tab[0][0][2] = put_vp8_epel16_h6_altivec;
c->put_vp8_epel_pixels_tab[0][2][0] = put_vp8_epel16_v6_altivec;
c->put_vp8_epel_pixels_tab[0][2][2] = put_vp8_epel16_h6v6_altivec;
c->put_vp8_epel_pixels_tab[1][0][2] = put_vp8_epel8_h6_altivec;
c->put_vp8_epel_pixels_tab[1][2][0] = put_vp8_epel8_v6_altivec;
c->put_vp8_epel_pixels_tab[1][0][1] = put_vp8_epel8_h4_altivec;
c->put_vp8_epel_pixels_tab[1][1][0] = put_vp8_epel8_v4_altivec;
c->put_vp8_epel_pixels_tab[1][2][2] = put_vp8_epel8_h6v6_altivec;
c->put_vp8_epel_pixels_tab[1][1][1] = put_vp8_epel8_h4v4_altivec;
c->put_vp8_epel_pixels_tab[1][1][2] = put_vp8_epel8_h6v4_altivec;
c->put_vp8_epel_pixels_tab[1][2][1] = put_vp8_epel8_h4v6_altivec;
c->put_vp8_epel_pixels_tab[2][0][2] = put_vp8_epel4_h6_altivec;
c->put_vp8_epel_pixels_tab[2][2][0] = put_vp8_epel4_v6_altivec;
c->put_vp8_epel_pixels_tab[2][0][1] = put_vp8_epel4_h4_altivec;
c->put_vp8_epel_pixels_tab[2][1][0] = put_vp8_epel4_v4_altivec;
c->put_vp8_epel_pixels_tab[2][2][2] = put_vp8_epel4_h6v6_altivec;
c->put_vp8_epel_pixels_tab[2][1][1] = put_vp8_epel4_h4v4_altivec;
c->put_vp8_epel_pixels_tab[2][1][2] = put_vp8_epel4_h6v4_altivec;
c->put_vp8_epel_pixels_tab[2][2][1] = put_vp8_epel4_h4v6_altivec;
#endif /* HAVE_ALTIVEC */
}