RepoMirrors/ffmpeg
1
0
Fork 0
mirror of https://git.ffmpeg.org/ffmpeg.git synced 2024-12-18 13:35:13 +00:00
Code Issues Projects Releases Wiki Activity

lavc/aarch64: Move non-neon vp9 copy functions out of neon source file.

Browse Source 
Fixes part of ticket #8565.
This commit is contained in:
Carl Eugen Hoyos 2020-03-11 13:01:02 +01:00
parent 613de37a6c
commit 9a21754904
4 changed files with 82 additions and 55 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
libavcodec/aarch64/Makefile
View File

@ -21,6 +21,7 @@ OBJS-$(CONFIG_VC1DSP) += aarch64/vc1dsp_init_aarch64.o
OBJS-$(CONFIG_VORBIS_DECODER) += aarch64/vorbisdsp_init.o
OBJS-$(CONFIG_VP9_DECODER) += aarch64/vp9dsp_init_10bpp_aarch64.o \
aarch64/vp9dsp_init_12bpp_aarch64.o \
aarch64/vp9mc_aarch64.o \
aarch64/vp9dsp_init_aarch64.o
# ARMv8 optimizations

25
libavcodec/aarch64/vp9mc_16bpp_neon.S
View File

@ -25,31 +25,6 @@
// const uint8_t *ref, ptrdiff_t ref_stride,
// int h, int mx, int my);
function ff_vp9_copy128_aarch64, export=1
1:
ldp x5, x6, [x2]
ldp x7, x8, [x2, #16]
stp x5, x6, [x0]
ldp x9, x10, [x2, #32]
stp x7, x8, [x0, #16]
subs w4, w4, #1
ldp x11, x12, [x2, #48]
stp x9, x10, [x0, #32]
stp x11, x12, [x0, #48]
ldp x5, x6, [x2, #64]
ldp x7, x8, [x2, #80]
stp x5, x6, [x0, #64]
ldp x9, x10, [x2, #96]
stp x7, x8, [x0, #80]
ldp x11, x12, [x2, #112]
stp x9, x10, [x0, #96]
stp x11, x12, [x0, #112]
add x2, x2, x3
add x0, x0, x1
b.ne 1b
ret
endfunc
function ff_vp9_avg64_16_neon, export=1
mov x5, x0
sub x1, x1, #64

81
libavcodec/aarch64/vp9mc_aarch64.c Normal file
View File

@ -0,0 +1,81 @@
/*
* Copyright (c) 2016 Google Inc.
*
* 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/aarch64/asm.S"
// All public functions in this file have the following signature:
// typedef void (*vp9_mc_func)(uint8_t *dst, ptrdiff_t dst_stride,
// const uint8_t *ref, ptrdiff_t ref_stride,
// int h, int mx, int my);
function ff_vp9_copy128_aarch64, export=1
1:
ldp x5, x6, [x2]
ldp x7, x8, [x2, #16]
stp x5, x6, [x0]
ldp x9, x10, [x2, #32]
stp x7, x8, [x0, #16]
subs w4, w4, #1
ldp x11, x12, [x2, #48]
stp x9, x10, [x0, #32]
stp x11, x12, [x0, #48]
ldp x5, x6, [x2, #64]
ldp x7, x8, [x2, #80]
stp x5, x6, [x0, #64]
ldp x9, x10, [x2, #96]
stp x7, x8, [x0, #80]
ldp x11, x12, [x2, #112]
stp x9, x10, [x0, #96]
stp x11, x12, [x0, #112]
add x2, x2, x3
add x0, x0, x1
b.ne 1b
ret
endfunc
function ff_vp9_copy64_aarch64, export=1
1:
ldp x5, x6, [x2]
ldp x7, x8, [x2, #16]
stp x5, x6, [x0]
ldp x9, x10, [x2, #32]
stp x7, x8, [x0, #16]
subs w4, w4, #1
ldp x11, x12, [x2, #48]
stp x9, x10, [x0, #32]
stp x11, x12, [x0, #48]
add x2, x2, x3
add x0, x0, x1
b.ne 1b
ret
endfunc
function ff_vp9_copy32_aarch64, export=1
1:
ldp x5, x6, [x2]
ldp x7, x8, [x2, #16]
stp x5, x6, [x0]
subs w4, w4, #1
stp x7, x8, [x0, #16]
add x2, x2, x3
add x0, x0, x1
b.ne 1b
ret
endfunc

30
libavcodec/aarch64/vp9mc_neon.S
View File

@ -25,23 +25,6 @@
// const uint8_t *ref, ptrdiff_t ref_stride,
// int h, int mx, int my);
function ff_vp9_copy64_aarch64, export=1
1:
ldp x5, x6, [x2]
ldp x7, x8, [x2, #16]
stp x5, x6, [x0]
ldp x9, x10, [x2, #32]
stp x7, x8, [x0, #16]
subs w4, w4, #1
ldp x11, x12, [x2, #48]
stp x9, x10, [x0, #32]
stp x11, x12, [x0, #48]
add x2, x2, x3
add x0, x0, x1
b.ne 1b
ret
endfunc
function ff_vp9_avg64_neon, export=1
mov x5, x0
1:
@ -64,19 +47,6 @@ function ff_vp9_avg64_neon, export=1
ret
endfunc
function ff_vp9_copy32_aarch64, export=1
1:
ldp x5, x6, [x2]
ldp x7, x8, [x2, #16]
stp x5, x6, [x0]
subs w4, w4, #1
stp x7, x8, [x0, #16]
add x2, x2, x3
add x0, x0, x1
b.ne 1b
ret
endfunc
function ff_vp9_avg32_neon, export=1
1:
ld1 {v2.16b, v3.16b}, [x2], x3