[FFmpeg-devel] [PATCH 4/4] avfilter/vf_v360: x86 SIMD for interpolations
James Almer
jamrial at gmail.com
Wed Sep 4 23:01:31 EEST 2019
On 9/4/2019 4:28 PM, Paul B Mahol wrote:
> diff --git a/libavfilter/x86/vf_v360.asm b/libavfilter/x86/vf_v360.asm
> new file mode 100644
> index 0000000000..46142a3bad
> --- /dev/null
> +++ b/libavfilter/x86/vf_v360.asm
> @@ -0,0 +1,104 @@
> +;*****************************************************************************
> +;* x86-optimized functions for v360 filter
> +;*
> +;* 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/x86/x86util.asm"
> +
> +SECTION_RODATA
> +
> +ALIGN 32
> +
> +pb_mask: db 0,4,8,12,5,5,5,5,5,5,5,5,5,5,5,5
> +pd_255: times 4 dd 255
> +pb_255: times 16 db 255
No need for this one. See below.
> +
> +SECTION .text
> +
> +; void ff_remap2_8bit_line_avx2(uint8_t *dst, int width, const uint8_t *src, ptrdiff_t in_linesize,
> +; const uint16_t *u, const uint16_t *v, const int16_t *ker);
> +
> +%if HAVE_AVX2_EXTERNAL
> +INIT_YMM avx2
> +cglobal remap1_8bit_line, 7, 7, 7, dst, width, src, in_linesize, u, v, x
You're loading seven regs when you only have six. x doesn't need to be
loaded.
> + movsxdifnidn widthq, widthd
> + movsxdifnidn in_linesizeq, in_linesized
Unneeded since in_linesize is ptrdiff_t.
> + xor xq, xq
> + movd xm0, in_linesized
> + VBROADCASTI128 m4, [pb_255]
pcmpeqw m4, m4
> + VBROADCASTI128 m6, [pb_mask]
> + vpbroadcastd m0, xm0
> +
> + .loop:
> + vpmovsxwd m1, [vq + xq * 2]
> + vpmovsxwd m2, [uq + xq * 2]
No need to use the v prefix in all these pre-avx instructions. INIT_YMM
takes care of it.
> +
> + vpmulld m3, m1, m0
> + vpaddd m1, m3, m2
pmulld m1, m0
paddd m1, m2
> + mova m2, m4
Pointless mova. Just use m4 in the vpgatherdd below.
> + vpgatherdd m5, [srcq + m1], m2
> + vextracti128 xm3, m5, 1
> + vpshufb m1, m5, m6
> + vpshufb m2, m3, m6
You could make these two pshufb use xmm regs, since you don't care
what's in the upper 128 bits.
> + movd [dstq+xq], xm1
> + movd [dstq+xq+4], xm2
> +
> + add xq, mmsize / 4
> + cmp xq, widthq
> + jl .loop
> + RET
> +
> +INIT_YMM avx2
> +cglobal remap2_8bit_line, 7, 9, 9, dst, width, src, in_linesize, u, v, ker, x, temp
> + movsxdifnidn widthq, widthd
> + movsxdifnidn in_linesizeq, in_linesized
Same as above.
> + xor xq, xq
> + movd xm0, in_linesized
> + VBROADCASTI128 m7, [pb_255]
Also pcmpeqw m4, m4
> + vpbroadcastd m0, xm0
> + movd xm6, [pd_255]
> + vpbroadcastd m6, xm6
VBROADCASTI128 m6, [pd_255]
> +
> + .loop:
> + vpmovsxwd m1, [kerq + xq * 8]
> + vpmovsxwd m2, [vq + xq * 8]
> + vpmovsxwd m3, [uq + xq * 8]
> +
> + vpmulld m4, m2, m0
> + vpaddd m4, m3
> + mova m3, m7
Also pointless mova. Use m7 below.
> + vpgatherdd m5, [srcq + m4], m3
> + vpand m5, m6
> + vpmulld m5, m1
> + vphaddd m2, m5, m5
> + vphaddd m5, m2, m2
> + vpsrld m5, m5, 0xd
> + vextracti128 xm3, m5, 1
> + vpshufb m5, m5, [pb_mask]
> + vpshufb m3, m3, [pb_mask]
Why aren't you loading the mask at the beginning into some register?
> +
> + movd tempd, xm5
> + mov [dstq+xq], tempb
> + movd tempd, xm3
> + mov [dstq+xq+1], tempb
did you try pextrb [mem], xm, 0?
> +
> + add xq, mmsize / 16
> + cmp xq, widthq
> + jl .loop
> + RET
> +%endif
> diff --git a/libavfilter/x86/vf_v360_init.c b/libavfilter/x86/vf_v360_init.c
> new file mode 100644
> index 0000000000..e48ee307b3
> --- /dev/null
> +++ b/libavfilter/x86/vf_v360_init.c
> @@ -0,0 +1,43 @@
> +/*
> + * 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/x86/cpu.h"
> +#include "libavfilter/v360.h"
> +
> +void ff_remap1_8bit_line_avx2(uint8_t *dst, int width, const uint8_t *src, ptrdiff_t in_linesize,
> + const uint16_t *u, const uint16_t *v, const int16_t *ker);
> +
> +void ff_remap2_8bit_line_avx2(uint8_t *dst, int width, const uint8_t *src, ptrdiff_t in_linesize,
> + const uint16_t *u, const uint16_t *v, const int16_t *ker);
> +
> +av_cold void ff_v360_init_x86(V360Context *s, int depth)
> +{
> +#if ARCH_X86_64
> + int cpu_flags = av_get_cpu_flags();
> +
> + if (EXTERNAL_AVX2(cpu_flags) && s->interp == NEAREST && depth <= 8)
> + s->remap_line = ff_remap1_8bit_line_avx2;
> +
> + if (EXTERNAL_AVX2(cpu_flags) && s->interp == BILINEAR && depth <= 8)
> + s->remap_line = ff_remap2_8bit_line_avx2;
EXTERNAL_AVX2_FAST() for both.
> +#endif
> +}
More information about the ffmpeg-devel
mailing list