You've already forked FFmpeg
mirror of
https://github.com/FFmpeg/FFmpeg.git
synced 2025-10-06 05:47:18 +02:00
35aaf697e9
Using xmm registers here is very natural, as it allows to operate on eight words at a time. It also saves 48B here and does not clobber the MMX state. Old benchmarks (only tests affected by the modified function are shown): avg_h264_qpel_8_mc11_8_c: 352.2 ( 1.00x) avg_h264_qpel_8_mc11_8_sse2: 70.4 ( 5.00x) avg_h264_qpel_8_mc11_8_ssse3: 53.9 ( 6.53x) avg_h264_qpel_8_mc13_8_c: 353.3 ( 1.00x) avg_h264_qpel_8_mc13_8_sse2: 72.8 ( 4.86x) avg_h264_qpel_8_mc13_8_ssse3: 53.8 ( 6.57x) avg_h264_qpel_8_mc21_8_c: 404.0 ( 1.00x) avg_h264_qpel_8_mc21_8_sse2: 116.1 ( 3.48x) avg_h264_qpel_8_mc21_8_ssse3: 94.3 ( 4.28x) avg_h264_qpel_8_mc23_8_c: 398.9 ( 1.00x) avg_h264_qpel_8_mc23_8_sse2: 118.6 ( 3.36x) avg_h264_qpel_8_mc23_8_ssse3: 94.8 ( 4.21x) avg_h264_qpel_8_mc31_8_c: 352.7 ( 1.00x) avg_h264_qpel_8_mc31_8_sse2: 71.4 ( 4.94x) avg_h264_qpel_8_mc31_8_ssse3: 53.8 ( 6.56x) avg_h264_qpel_8_mc33_8_c: 354.0 ( 1.00x) avg_h264_qpel_8_mc33_8_sse2: 70.6 ( 5.01x) avg_h264_qpel_8_mc33_8_ssse3: 53.7 ( 6.59x) avg_h264_qpel_16_mc11_8_c: 1417.0 ( 1.00x) avg_h264_qpel_16_mc11_8_sse2: 276.9 ( 5.12x) avg_h264_qpel_16_mc11_8_ssse3: 178.8 ( 7.92x) avg_h264_qpel_16_mc13_8_c: 1427.3 ( 1.00x) avg_h264_qpel_16_mc13_8_sse2: 277.4 ( 5.14x) avg_h264_qpel_16_mc13_8_ssse3: 179.7 ( 7.94x) avg_h264_qpel_16_mc21_8_c: 1634.1 ( 1.00x) avg_h264_qpel_16_mc21_8_sse2: 421.3 ( 3.88x) avg_h264_qpel_16_mc21_8_ssse3: 291.2 ( 5.61x) avg_h264_qpel_16_mc23_8_c: 1627.0 ( 1.00x) avg_h264_qpel_16_mc23_8_sse2: 420.8 ( 3.87x) avg_h264_qpel_16_mc23_8_ssse3: 291.0 ( 5.59x) avg_h264_qpel_16_mc31_8_c: 1418.4 ( 1.00x) avg_h264_qpel_16_mc31_8_sse2: 278.5 ( 5.09x) avg_h264_qpel_16_mc31_8_ssse3: 178.6 ( 7.94x) avg_h264_qpel_16_mc33_8_c: 1407.3 ( 1.00x) avg_h264_qpel_16_mc33_8_sse2: 277.6 ( 5.07x) avg_h264_qpel_16_mc33_8_ssse3: 179.9 ( 7.82x) put_h264_qpel_8_mc11_8_c: 348.1 ( 1.00x) put_h264_qpel_8_mc11_8_sse2: 69.1 ( 5.04x) put_h264_qpel_8_mc11_8_ssse3: 53.8 ( 6.47x) put_h264_qpel_8_mc13_8_c: 349.3 ( 1.00x) put_h264_qpel_8_mc13_8_sse2: 69.7 ( 5.01x) put_h264_qpel_8_mc13_8_ssse3: 53.7 ( 6.51x) put_h264_qpel_8_mc21_8_c: 398.5 ( 1.00x) put_h264_qpel_8_mc21_8_sse2: 115.0 ( 3.46x) put_h264_qpel_8_mc21_8_ssse3: 95.3 ( 4.18x) put_h264_qpel_8_mc23_8_c: 399.9 ( 1.00x) put_h264_qpel_8_mc23_8_sse2: 120.8 ( 3.31x) put_h264_qpel_8_mc23_8_ssse3: 95.4 ( 4.19x) put_h264_qpel_8_mc31_8_c: 350.4 ( 1.00x) put_h264_qpel_8_mc31_8_sse2: 69.6 ( 5.03x) put_h264_qpel_8_mc31_8_ssse3: 54.2 ( 6.47x) put_h264_qpel_8_mc33_8_c: 353.1 ( 1.00x) put_h264_qpel_8_mc33_8_sse2: 71.0 ( 4.97x) put_h264_qpel_8_mc33_8_ssse3: 54.2 ( 6.51x) put_h264_qpel_16_mc11_8_c: 1384.2 ( 1.00x) put_h264_qpel_16_mc11_8_sse2: 272.9 ( 5.07x) put_h264_qpel_16_mc11_8_ssse3: 178.3 ( 7.76x) put_h264_qpel_16_mc13_8_c: 1393.6 ( 1.00x) put_h264_qpel_16_mc13_8_sse2: 271.1 ( 5.14x) put_h264_qpel_16_mc13_8_ssse3: 178.3 ( 7.82x) put_h264_qpel_16_mc21_8_c: 1612.6 ( 1.00x) put_h264_qpel_16_mc21_8_sse2: 416.5 ( 3.87x) put_h264_qpel_16_mc21_8_ssse3: 289.1 ( 5.58x) put_h264_qpel_16_mc23_8_c: 1621.3 ( 1.00x) put_h264_qpel_16_mc23_8_sse2: 416.9 ( 3.89x) put_h264_qpel_16_mc23_8_ssse3: 289.4 ( 5.60x) put_h264_qpel_16_mc31_8_c: 1408.4 ( 1.00x) put_h264_qpel_16_mc31_8_sse2: 273.5 ( 5.15x) put_h264_qpel_16_mc31_8_ssse3: 176.9 ( 7.96x) put_h264_qpel_16_mc33_8_c: 1396.4 ( 1.00x) put_h264_qpel_16_mc33_8_sse2: 276.3 ( 5.05x) put_h264_qpel_16_mc33_8_ssse3: 176.4 ( 7.92x) New benchmarks: avg_h264_qpel_8_mc11_8_c: 352.1 ( 1.00x) avg_h264_qpel_8_mc11_8_sse2: 52.5 ( 6.71x) avg_h264_qpel_8_mc11_8_ssse3: 53.9 ( 6.54x) avg_h264_qpel_8_mc13_8_c: 350.8 ( 1.00x) avg_h264_qpel_8_mc13_8_sse2: 54.7 ( 6.42x) avg_h264_qpel_8_mc13_8_ssse3: 54.3 ( 6.46x) avg_h264_qpel_8_mc21_8_c: 400.1 ( 1.00x) avg_h264_qpel_8_mc21_8_sse2: 98.6 ( 4.06x) avg_h264_qpel_8_mc21_8_ssse3: 95.5 ( 4.19x) avg_h264_qpel_8_mc23_8_c: 400.4 ( 1.00x) avg_h264_qpel_8_mc23_8_sse2: 101.4 ( 3.95x) avg_h264_qpel_8_mc23_8_ssse3: 95.9 ( 4.18x) avg_h264_qpel_8_mc31_8_c: 352.4 ( 1.00x) avg_h264_qpel_8_mc31_8_sse2: 52.9 ( 6.67x) avg_h264_qpel_8_mc31_8_ssse3: 54.4 ( 6.48x) avg_h264_qpel_8_mc33_8_c: 354.5 ( 1.00x) avg_h264_qpel_8_mc33_8_sse2: 52.9 ( 6.70x) avg_h264_qpel_8_mc33_8_ssse3: 54.4 ( 6.52x) avg_h264_qpel_16_mc11_8_c: 1420.4 ( 1.00x) avg_h264_qpel_16_mc11_8_sse2: 204.8 ( 6.93x) avg_h264_qpel_16_mc11_8_ssse3: 177.9 ( 7.98x) avg_h264_qpel_16_mc13_8_c: 1409.8 ( 1.00x) avg_h264_qpel_16_mc13_8_sse2: 206.4 ( 6.83x) avg_h264_qpel_16_mc13_8_ssse3: 178.0 ( 7.92x) avg_h264_qpel_16_mc21_8_c: 1634.1 ( 1.00x) avg_h264_qpel_16_mc21_8_sse2: 349.6 ( 4.67x) avg_h264_qpel_16_mc21_8_ssse3: 290.0 ( 5.63x) avg_h264_qpel_16_mc23_8_c: 1624.1 ( 1.00x) avg_h264_qpel_16_mc23_8_sse2: 350.0 ( 4.64x) avg_h264_qpel_16_mc23_8_ssse3: 291.9 ( 5.56x) avg_h264_qpel_16_mc31_8_c: 1407.2 ( 1.00x) avg_h264_qpel_16_mc31_8_sse2: 205.8 ( 6.84x) avg_h264_qpel_16_mc31_8_ssse3: 178.2 ( 7.90x) avg_h264_qpel_16_mc33_8_c: 1400.5 ( 1.00x) avg_h264_qpel_16_mc33_8_sse2: 206.3 ( 6.79x) avg_h264_qpel_16_mc33_8_ssse3: 179.4 ( 7.81x) put_h264_qpel_8_mc11_8_c: 349.7 ( 1.00x) put_h264_qpel_8_mc11_8_sse2: 50.2 ( 6.96x) put_h264_qpel_8_mc11_8_ssse3: 51.3 ( 6.82x) put_h264_qpel_8_mc13_8_c: 349.8 ( 1.00x) put_h264_qpel_8_mc13_8_sse2: 50.7 ( 6.90x) put_h264_qpel_8_mc13_8_ssse3: 51.7 ( 6.76x) put_h264_qpel_8_mc21_8_c: 398.0 ( 1.00x) put_h264_qpel_8_mc21_8_sse2: 96.5 ( 4.13x) put_h264_qpel_8_mc21_8_ssse3: 92.3 ( 4.31x) put_h264_qpel_8_mc23_8_c: 401.4 ( 1.00x) put_h264_qpel_8_mc23_8_sse2: 102.3 ( 3.92x) put_h264_qpel_8_mc23_8_ssse3: 92.8 ( 4.32x) put_h264_qpel_8_mc31_8_c: 349.4 ( 1.00x) put_h264_qpel_8_mc31_8_sse2: 50.8 ( 6.88x) put_h264_qpel_8_mc31_8_ssse3: 51.8 ( 6.75x) put_h264_qpel_8_mc33_8_c: 351.1 ( 1.00x) put_h264_qpel_8_mc33_8_sse2: 52.2 ( 6.73x) put_h264_qpel_8_mc33_8_ssse3: 51.7 ( 6.79x) put_h264_qpel_16_mc11_8_c: 1391.1 ( 1.00x) put_h264_qpel_16_mc11_8_sse2: 196.6 ( 7.07x) put_h264_qpel_16_mc11_8_ssse3: 178.2 ( 7.81x) put_h264_qpel_16_mc13_8_c: 1385.2 ( 1.00x) put_h264_qpel_16_mc13_8_sse2: 195.6 ( 7.08x) put_h264_qpel_16_mc13_8_ssse3: 176.6 ( 7.84x) put_h264_qpel_16_mc21_8_c: 1607.5 ( 1.00x) put_h264_qpel_16_mc21_8_sse2: 341.0 ( 4.71x) put_h264_qpel_16_mc21_8_ssse3: 289.1 ( 5.56x) put_h264_qpel_16_mc23_8_c: 1616.7 ( 1.00x) put_h264_qpel_16_mc23_8_sse2: 340.8 ( 4.74x) put_h264_qpel_16_mc23_8_ssse3: 288.6 ( 5.60x) put_h264_qpel_16_mc31_8_c: 1397.6 ( 1.00x) put_h264_qpel_16_mc31_8_sse2: 197.3 ( 7.08x) put_h264_qpel_16_mc31_8_ssse3: 175.4 ( 7.97x) put_h264_qpel_16_mc33_8_c: 1394.3 ( 1.00x) put_h264_qpel_16_mc33_8_sse2: 197.7 ( 7.05x) put_h264_qpel_16_mc33_8_ssse3: 175.2 ( 7.96x) As can be seen, the SSE2 version is often neck-to-neck with the SSSE3 version (which also benefits from a better hv2_lowpass SSSE3 implementation for mc21 and mc23) for eight byte block sizes. Unsurprisingly, SSSE3 beats SSE2 for 16x16 blocks: For SSE2, these blocks are processed by calling the 8x8 function four times whereas SSSE3 has a dedicated function (on x64). This implementation should also be extendable to an AVX version for 16x16 blocks. Reviewed-by: James Almer <jamrial@gmail.com> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
829 lines
20 KiB
NASM
829 lines
20 KiB
NASM
;*****************************************************************************
|
|
;* MMX/SSE2/SSSE3-optimized H.264 QPEL code
|
|
;*****************************************************************************
|
|
;* Copyright (c) 2004-2005 Michael Niedermayer, Loren Merritt
|
|
;* Copyright (C) 2012 Daniel Kang
|
|
;*
|
|
;* Authors: Daniel Kang <daniel.d.kang@gmail.com>
|
|
;*
|
|
;* 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"
|
|
|
|
cextern pw_16
|
|
cextern pw_5
|
|
cextern pb_0
|
|
|
|
SECTION .text
|
|
|
|
; void ff_avg_pixels4_mmxext(uint8_t *block, const uint8_t *pixels,
|
|
; ptrdiff_t line_size)
|
|
INIT_MMX mmxext
|
|
cglobal avg_pixels4, 3,4
|
|
lea r3, [r2*3]
|
|
movh m0, [r1]
|
|
movh m1, [r1+r2]
|
|
movh m2, [r1+r2*2]
|
|
movh m3, [r1+r3]
|
|
pavgb m0, [r0]
|
|
pavgb m1, [r0+r2]
|
|
pavgb m2, [r0+r2*2]
|
|
pavgb m3, [r0+r3]
|
|
movh [r0], m0
|
|
movh [r0+r2], m1
|
|
movh [r0+r2*2], m2
|
|
movh [r0+r3], m3
|
|
RET
|
|
|
|
%macro op_avgh 3
|
|
movh %3, %2
|
|
pavgb %1, %3
|
|
movh %2, %1
|
|
%endmacro
|
|
|
|
%macro op_avg 2-3
|
|
pavgb %1, %2
|
|
mova %2, %1
|
|
%endmacro
|
|
|
|
%macro op_puth 2-3
|
|
movh %2, %1
|
|
%endmacro
|
|
|
|
%macro op_put 2-3
|
|
mova %2, %1
|
|
%endmacro
|
|
|
|
; void ff_put/avg_pixels4_l2_mmxext(uint8_t *dst, const uint8_t *src1, const uint8_t *src2,
|
|
; ptrdiff_t stride)
|
|
%macro PIXELS4_L2 1
|
|
%define OP op_%1h
|
|
cglobal %1_pixels4_l2, 4,4
|
|
mova m0, [r1]
|
|
mova m1, [r1+r3]
|
|
lea r1, [r1+2*r3]
|
|
pavgb m0, [r2]
|
|
pavgb m1, [r2+4]
|
|
OP m0, [r0], m3
|
|
OP m1, [r0+r3], m3
|
|
lea r0, [r0+2*r3]
|
|
mova m0, [r1]
|
|
mova m1, [r1+r3]
|
|
pavgb m0, [r2+8]
|
|
pavgb m1, [r2+12]
|
|
OP m0, [r0], m3
|
|
OP m1, [r0+r3], m3
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_MMX mmxext
|
|
PIXELS4_L2 put
|
|
PIXELS4_L2 avg
|
|
|
|
%macro QPEL4_H_LOWPASS_OP 1
|
|
cglobal %1_h264_qpel4_h_lowpass, 4,5 ; dst, src, dstStride, srcStride
|
|
movsxdifnidn r2, r2d
|
|
movsxdifnidn r3, r3d
|
|
pxor m7, m7
|
|
mova m4, [pw_5]
|
|
mova m5, [pw_16]
|
|
mov r4d, 4
|
|
.loop:
|
|
movh m1, [r1-1]
|
|
movh m2, [r1+0]
|
|
movh m3, [r1+1]
|
|
movh m0, [r1+2]
|
|
punpcklbw m1, m7
|
|
punpcklbw m2, m7
|
|
punpcklbw m3, m7
|
|
punpcklbw m0, m7
|
|
paddw m1, m0
|
|
paddw m2, m3
|
|
movh m0, [r1-2]
|
|
movh m3, [r1+3]
|
|
punpcklbw m0, m7
|
|
punpcklbw m3, m7
|
|
paddw m0, m3
|
|
psllw m2, 2
|
|
psubw m2, m1
|
|
pmullw m2, m4
|
|
paddw m0, m5
|
|
paddw m0, m2
|
|
psraw m0, 5
|
|
packuswb m0, m0
|
|
op_%1h m0, [r0], m6
|
|
add r0, r2
|
|
add r1, r3
|
|
dec r4d
|
|
jg .loop
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_MMX mmxext
|
|
QPEL4_H_LOWPASS_OP put
|
|
QPEL4_H_LOWPASS_OP avg
|
|
|
|
%macro QPEL8_H_LOWPASS_OP_XMM 1
|
|
cglobal %1_h264_qpel8_h_lowpass, 4,5,8 ; dst, src, dstStride, srcStride
|
|
movsxdifnidn r2, r2d
|
|
movsxdifnidn r3, r3d
|
|
mov r4d, 8
|
|
pxor m7, m7
|
|
mova m6, [pw_5]
|
|
.loop:
|
|
movu m1, [r1-2]
|
|
mova m0, m1
|
|
punpckhbw m1, m7
|
|
punpcklbw m0, m7
|
|
mova m2, m1
|
|
mova m3, m1
|
|
mova m4, m1
|
|
mova m5, m1
|
|
palignr m4, m0, 2
|
|
palignr m3, m0, 4
|
|
palignr m2, m0, 6
|
|
palignr m1, m0, 8
|
|
palignr m5, m0, 10
|
|
paddw m0, m5
|
|
paddw m2, m3
|
|
paddw m1, m4
|
|
psllw m2, 2
|
|
psubw m2, m1
|
|
paddw m0, [pw_16]
|
|
pmullw m2, m6
|
|
paddw m2, m0
|
|
psraw m2, 5
|
|
packuswb m2, m2
|
|
op_%1h m2, [r0], m4
|
|
add r1, r3
|
|
add r0, r2
|
|
dec r4d
|
|
jne .loop
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_XMM ssse3
|
|
QPEL8_H_LOWPASS_OP_XMM put
|
|
QPEL8_H_LOWPASS_OP_XMM avg
|
|
|
|
|
|
%macro QPEL4_H_LOWPASS_L2_OP 1
|
|
cglobal %1_h264_qpel4_h_lowpass_l2, 5,6 ; dst, src, src2, dstStride, srcStride
|
|
movsxdifnidn r3, r3d
|
|
movsxdifnidn r4, r4d
|
|
pxor m7, m7
|
|
mova m4, [pw_5]
|
|
mova m5, [pw_16]
|
|
mov r5d, 4
|
|
.loop:
|
|
movh m1, [r1-1]
|
|
movh m2, [r1+0]
|
|
movh m3, [r1+1]
|
|
movh m0, [r1+2]
|
|
punpcklbw m1, m7
|
|
punpcklbw m2, m7
|
|
punpcklbw m3, m7
|
|
punpcklbw m0, m7
|
|
paddw m1, m0
|
|
paddw m2, m3
|
|
movh m0, [r1-2]
|
|
movh m3, [r1+3]
|
|
punpcklbw m0, m7
|
|
punpcklbw m3, m7
|
|
paddw m0, m3
|
|
psllw m2, 2
|
|
psubw m2, m1
|
|
pmullw m2, m4
|
|
paddw m0, m5
|
|
paddw m0, m2
|
|
movh m3, [r2]
|
|
psraw m0, 5
|
|
packuswb m0, m0
|
|
pavgb m0, m3
|
|
op_%1h m0, [r0], m6
|
|
add r0, r3
|
|
add r1, r3
|
|
add r2, r4
|
|
dec r5d
|
|
jg .loop
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_MMX mmxext
|
|
QPEL4_H_LOWPASS_L2_OP put
|
|
QPEL4_H_LOWPASS_L2_OP avg
|
|
|
|
|
|
%macro QPEL8_H_LOWPASS_L2_OP 1
|
|
cglobal %1_h264_qpel8_h_lowpass_l2, 5,6,6 ; dst, src, src2, dstStride, srcStride
|
|
movsxdifnidn r3, r3d
|
|
movsxdifnidn r4, r4d
|
|
mova m3, [pw_16]
|
|
mov r5d, 8
|
|
pxor m5, m5
|
|
mova m4, [pw_5]
|
|
.loop:
|
|
movh m0, [r1]
|
|
movh m1, [r1+1]
|
|
punpcklbw m0, m5
|
|
punpcklbw m1, m5
|
|
paddw m0, m1
|
|
psllw m0, 2
|
|
movh m1, [r1-1]
|
|
movh m2, [r1+2]
|
|
punpcklbw m1, m5
|
|
punpcklbw m2, m5
|
|
paddw m1, m2
|
|
psubw m0, m1
|
|
pmullw m0, m4
|
|
movh m1, [r1-2]
|
|
movh m2, [r1+3]
|
|
punpcklbw m1, m5
|
|
punpcklbw m2, m5
|
|
paddw m0, m1
|
|
paddw m0, m2
|
|
paddw m0, m3
|
|
psraw m0, 5
|
|
packuswb m0, m5
|
|
movh m2, [r2]
|
|
pavgb m0, m2
|
|
op_%1h m0, [r0], m2
|
|
add r0, r3
|
|
add r1, r3
|
|
add r2, r4
|
|
dec r5d
|
|
jg .loop
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_XMM sse2
|
|
QPEL8_H_LOWPASS_L2_OP put
|
|
QPEL8_H_LOWPASS_L2_OP avg
|
|
|
|
|
|
%macro QPEL8_H_LOWPASS_L2_OP_XMM 1
|
|
cglobal %1_h264_qpel8_h_lowpass_l2, 5,6,8 ; dst, src, src2, dstStride, src2Stride
|
|
movsxdifnidn r3, r3d
|
|
movsxdifnidn r4, r4d
|
|
mov r5d, 8
|
|
pxor m7, m7
|
|
mova m6, [pw_5]
|
|
.loop:
|
|
lddqu m1, [r1-2]
|
|
mova m0, m1
|
|
punpckhbw m1, m7
|
|
punpcklbw m0, m7
|
|
mova m2, m1
|
|
mova m3, m1
|
|
mova m4, m1
|
|
mova m5, m1
|
|
palignr m4, m0, 2
|
|
palignr m3, m0, 4
|
|
palignr m2, m0, 6
|
|
palignr m1, m0, 8
|
|
palignr m5, m0, 10
|
|
paddw m0, m5
|
|
paddw m2, m3
|
|
paddw m1, m4
|
|
psllw m2, 2
|
|
movh m3, [r2]
|
|
psubw m2, m1
|
|
paddw m0, [pw_16]
|
|
pmullw m2, m6
|
|
paddw m2, m0
|
|
psraw m2, 5
|
|
packuswb m2, m2
|
|
pavgb m2, m3
|
|
op_%1h m2, [r0], m4
|
|
add r1, r3
|
|
add r0, r3
|
|
add r2, r4
|
|
dec r5d
|
|
jg .loop
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_XMM ssse3
|
|
QPEL8_H_LOWPASS_L2_OP_XMM put
|
|
QPEL8_H_LOWPASS_L2_OP_XMM avg
|
|
|
|
|
|
; All functions that call this are required to have function arguments of
|
|
; dst, src, dstStride, srcStride
|
|
%macro FILT_V 1
|
|
mova m6, m2
|
|
movh m5, [r1]
|
|
paddw m6, m3
|
|
psllw m6, 2
|
|
psubw m6, m1
|
|
psubw m6, m4
|
|
punpcklbw m5, m7
|
|
pmullw m6, [pw_5]
|
|
paddw m0, [pw_16]
|
|
add r1, r3
|
|
paddw m0, m5
|
|
paddw m6, m0
|
|
psraw m6, 5
|
|
packuswb m6, m6
|
|
op_%1h m6, [r0], m0 ; 1
|
|
add r0, r2
|
|
SWAP 0, 1, 2, 3, 4, 5
|
|
%endmacro
|
|
|
|
%macro QPEL4_V_LOWPASS_OP 1
|
|
cglobal %1_h264_qpel4_v_lowpass, 4,4 ; dst, src, dstStride, srcStride
|
|
movsxdifnidn r2, r2d
|
|
movsxdifnidn r3, r3d
|
|
sub r1, r3
|
|
sub r1, r3
|
|
pxor m7, m7
|
|
movh m0, [r1]
|
|
movh m1, [r1+r3]
|
|
lea r1, [r1+2*r3]
|
|
movh m2, [r1]
|
|
movh m3, [r1+r3]
|
|
lea r1, [r1+2*r3]
|
|
movh m4, [r1]
|
|
add r1, r3
|
|
punpcklbw m0, m7
|
|
punpcklbw m1, m7
|
|
punpcklbw m2, m7
|
|
punpcklbw m3, m7
|
|
punpcklbw m4, m7
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_MMX mmxext
|
|
QPEL4_V_LOWPASS_OP put
|
|
QPEL4_V_LOWPASS_OP avg
|
|
|
|
|
|
|
|
%macro QPEL8OR16_V_LOWPASS_OP 1
|
|
%if cpuflag(sse2)
|
|
cglobal %1_h264_qpel8or16_v_lowpass, 5,5,8 ; dst, src, dstStride, srcStride, h
|
|
movsxdifnidn r2, r2d
|
|
movsxdifnidn r3, r3d
|
|
sub r1, r3
|
|
sub r1, r3
|
|
%else
|
|
cglobal %1_h264_qpel8or16_v_lowpass_op, 5,5,8 ; dst, src, dstStride, srcStride, h
|
|
movsxdifnidn r2, r2d
|
|
movsxdifnidn r3, r3d
|
|
%endif
|
|
pxor m7, m7
|
|
movh m0, [r1]
|
|
movh m1, [r1+r3]
|
|
lea r1, [r1+2*r3]
|
|
movh m2, [r1]
|
|
movh m3, [r1+r3]
|
|
lea r1, [r1+2*r3]
|
|
movh m4, [r1]
|
|
add r1, r3
|
|
punpcklbw m0, m7
|
|
punpcklbw m1, m7
|
|
punpcklbw m2, m7
|
|
punpcklbw m3, m7
|
|
punpcklbw m4, m7
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
cmp r4d, 16
|
|
jne .end
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
FILT_V %1
|
|
.end:
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_XMM sse2
|
|
QPEL8OR16_V_LOWPASS_OP put
|
|
QPEL8OR16_V_LOWPASS_OP avg
|
|
|
|
|
|
; All functions that use this are required to have args:
|
|
; src, tmp, srcSize
|
|
%macro FILT_HV 1 ; offset
|
|
mova m6, m2
|
|
movh m5, [r0]
|
|
paddw m6, m3
|
|
psllw m6, 2
|
|
paddw m0, [pw_16]
|
|
psubw m6, m1
|
|
psubw m6, m4
|
|
punpcklbw m5, m7
|
|
pmullw m6, [pw_5]
|
|
paddw m0, m5
|
|
add r0, r2
|
|
paddw m6, m0
|
|
mova [r1+%1], m6
|
|
SWAP 0, 1, 2, 3, 4, 5
|
|
%endmacro
|
|
|
|
%macro QPEL4_HV1_LOWPASS_OP 1
|
|
cglobal %1_h264_qpel4_hv_lowpass_v, 3,3 ; src, tmp, srcStride
|
|
movsxdifnidn r2, r2d
|
|
pxor m7, m7
|
|
movh m0, [r0]
|
|
movh m1, [r0+r2]
|
|
lea r0, [r0+2*r2]
|
|
movh m2, [r0]
|
|
movh m3, [r0+r2]
|
|
lea r0, [r0+2*r2]
|
|
movh m4, [r0]
|
|
add r0, r2
|
|
punpcklbw m0, m7
|
|
punpcklbw m1, m7
|
|
punpcklbw m2, m7
|
|
punpcklbw m3, m7
|
|
punpcklbw m4, m7
|
|
FILT_HV 0*24
|
|
FILT_HV 1*24
|
|
FILT_HV 2*24
|
|
FILT_HV 3*24
|
|
RET
|
|
|
|
cglobal %1_h264_qpel4_hv_lowpass_h, 3,4 ; tmp, dst, dstStride
|
|
movsxdifnidn r2, r2d
|
|
mov r3d, 4
|
|
.loop:
|
|
mova m0, [r0]
|
|
paddw m0, [r0+10]
|
|
mova m1, [r0+2]
|
|
paddw m1, [r0+8]
|
|
mova m2, [r0+4]
|
|
paddw m2, [r0+6]
|
|
psubw m0, m1
|
|
psraw m0, 2
|
|
psubw m0, m1
|
|
paddsw m0, m2
|
|
psraw m0, 2
|
|
paddw m0, m2
|
|
psraw m0, 6
|
|
packuswb m0, m0
|
|
op_%1h m0, [r1], m7
|
|
add r0, 24
|
|
add r1, r2
|
|
dec r3d
|
|
jnz .loop
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_MMX mmxext
|
|
QPEL4_HV1_LOWPASS_OP put
|
|
QPEL4_HV1_LOWPASS_OP avg
|
|
|
|
%macro QPEL8OR16_HV1_LOWPASS_OP 1
|
|
cglobal %1_h264_qpel8or16_hv1_lowpass_op, 4,4,8 ; src, tmp, srcStride, size
|
|
movsxdifnidn r2, r2d
|
|
pxor m7, m7
|
|
movh m0, [r0]
|
|
movh m1, [r0+r2]
|
|
lea r0, [r0+2*r2]
|
|
movh m2, [r0]
|
|
movh m3, [r0+r2]
|
|
lea r0, [r0+2*r2]
|
|
movh m4, [r0]
|
|
add r0, r2
|
|
punpcklbw m0, m7
|
|
punpcklbw m1, m7
|
|
punpcklbw m2, m7
|
|
punpcklbw m3, m7
|
|
punpcklbw m4, m7
|
|
FILT_HV 0*48
|
|
FILT_HV 1*48
|
|
FILT_HV 2*48
|
|
FILT_HV 3*48
|
|
FILT_HV 4*48
|
|
FILT_HV 5*48
|
|
FILT_HV 6*48
|
|
FILT_HV 7*48
|
|
cmp r3d, 16
|
|
jne .end
|
|
FILT_HV 8*48
|
|
FILT_HV 9*48
|
|
FILT_HV 10*48
|
|
FILT_HV 11*48
|
|
FILT_HV 12*48
|
|
FILT_HV 13*48
|
|
FILT_HV 14*48
|
|
FILT_HV 15*48
|
|
.end:
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_XMM sse2
|
|
QPEL8OR16_HV1_LOWPASS_OP put
|
|
|
|
|
|
|
|
%macro QPEL8OR16_HV2_LOWPASS_OP 1
|
|
cglobal %1_h264_qpel8or16_hv2_lowpass_op, 4,4 ; dst, tmp, dstStride, h
|
|
movsxdifnidn r2, r2d
|
|
.loop:
|
|
mova m0, [r1]
|
|
mova m3, [r1+8]
|
|
mova m1, [r1+2]
|
|
mova m4, [r1+10]
|
|
paddw m0, m4
|
|
paddw m1, m3
|
|
paddw m3, [r1+18]
|
|
paddw m4, [r1+16]
|
|
mova m2, [r1+4]
|
|
mova m5, [r1+12]
|
|
paddw m2, [r1+6]
|
|
paddw m5, [r1+14]
|
|
psubw m0, m1
|
|
psubw m3, m4
|
|
psraw m0, 2
|
|
psraw m3, 2
|
|
psubw m0, m1
|
|
psubw m3, m4
|
|
paddsw m0, m2
|
|
paddsw m3, m5
|
|
psraw m0, 2
|
|
psraw m3, 2
|
|
paddw m0, m2
|
|
paddw m3, m5
|
|
psraw m0, 6
|
|
psraw m3, 6
|
|
packuswb m0, m3
|
|
op_%1 m0, [r0], m7
|
|
add r1, 48
|
|
add r0, r2
|
|
dec r3d
|
|
jne .loop
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_MMX mmxext
|
|
QPEL8OR16_HV2_LOWPASS_OP put
|
|
QPEL8OR16_HV2_LOWPASS_OP avg
|
|
|
|
%macro QPEL8OR16_HV2_LOWPASS_OP_XMM 1
|
|
%ifidn %1, avg
|
|
cglobal %1_h264_qpel8_hv2_lowpass, 3,4,7 ; dst, tmp, dstStride
|
|
%else
|
|
cglobal %1_h264_qpel8_hv2_lowpass, 3,4,6 ; dst, tmp, dstStride
|
|
%endif
|
|
movsxdifnidn r2, r2d
|
|
mov r3d, 8
|
|
.loop:
|
|
mova m1, [r1+16]
|
|
mova m0, [r1]
|
|
mova m2, m1
|
|
mova m3, m1
|
|
mova m4, m1
|
|
mova m5, m1
|
|
palignr m5, m0, 10
|
|
palignr m4, m0, 8
|
|
palignr m3, m0, 6
|
|
palignr m2, m0, 4
|
|
palignr m1, m0, 2
|
|
paddw m0, m5
|
|
paddw m1, m4
|
|
paddw m2, m3
|
|
psubw m0, m1
|
|
psraw m0, 2
|
|
psubw m0, m1
|
|
paddw m0, m2
|
|
psraw m0, 2
|
|
paddw m0, m2
|
|
psraw m0, 6
|
|
packuswb m0, m0
|
|
op_%1h m0, [r0], m6
|
|
add r1, 48
|
|
add r0, r2
|
|
dec r3d
|
|
jne .loop
|
|
RET
|
|
|
|
cglobal %1_h264_qpel16_hv2_lowpass, 3,4,8 ; dst, tmp, dstStride
|
|
movsxdifnidn r2, r2d
|
|
mov r3d, 16
|
|
.loop:
|
|
mova m4, [r1+32]
|
|
mova m5, [r1+16]
|
|
mova m7, [r1]
|
|
mova m3, m4
|
|
mova m2, m4
|
|
mova m1, m4
|
|
mova m0, m4
|
|
palignr m0, m5, 10
|
|
palignr m1, m5, 8
|
|
palignr m2, m5, 6
|
|
palignr m3, m5, 4
|
|
palignr m4, m5, 2
|
|
paddw m0, m5
|
|
paddw m1, m4
|
|
paddw m2, m3
|
|
mova m6, m5
|
|
mova m4, m5
|
|
mova m3, m5
|
|
palignr m4, m7, 8
|
|
palignr m6, m7, 2
|
|
palignr m3, m7, 10
|
|
paddw m4, m6
|
|
mova m6, m5
|
|
palignr m5, m7, 6
|
|
palignr m6, m7, 4
|
|
paddw m3, m7
|
|
paddw m5, m6
|
|
psubw m0, m1
|
|
psubw m3, m4
|
|
psraw m0, 2
|
|
psraw m3, 2
|
|
psubw m0, m1
|
|
psubw m3, m4
|
|
paddw m0, m2
|
|
paddw m3, m5
|
|
psraw m0, 2
|
|
psraw m3, 2
|
|
paddw m0, m2
|
|
paddw m3, m5
|
|
psraw m0, 6
|
|
psraw m3, 6
|
|
packuswb m3, m0
|
|
op_%1 m3, [r0], m7
|
|
add r1, 48
|
|
add r0, r2
|
|
dec r3d
|
|
jne .loop
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_XMM ssse3
|
|
QPEL8OR16_HV2_LOWPASS_OP_XMM put
|
|
QPEL8OR16_HV2_LOWPASS_OP_XMM avg
|
|
|
|
|
|
%macro PIXELS4_L2_SHIFT5 1
|
|
cglobal %1_pixels4_l2_shift5,4,4 ; dst, src16, src8, dstStride
|
|
movsxdifnidn r3, r3d
|
|
mova m0, [r1]
|
|
mova m1, [r1+24]
|
|
psraw m0, 5
|
|
psraw m1, 5
|
|
packuswb m0, m0
|
|
packuswb m1, m1
|
|
pavgb m0, [r2]
|
|
pavgb m1, [r2+4]
|
|
op_%1h m0, [r0], m4
|
|
op_%1h m1, [r0+r3], m5
|
|
lea r0, [r0+r3*2]
|
|
mova m0, [r1+48]
|
|
mova m1, [r1+72]
|
|
psraw m0, 5
|
|
psraw m1, 5
|
|
packuswb m0, m0
|
|
packuswb m1, m1
|
|
pavgb m0, [r2+2*4]
|
|
pavgb m1, [r2+3*4]
|
|
op_%1h m0, [r0], m4
|
|
op_%1h m1, [r0+r3], m5
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_MMX mmxext
|
|
PIXELS4_L2_SHIFT5 put
|
|
PIXELS4_L2_SHIFT5 avg
|
|
|
|
|
|
%macro PIXELS_L2_SHIFT5 2
|
|
%if cpuflag(sse2)
|
|
cglobal %1_pixels%2_l2_shift5, 5, 5, 4 ; dst, src16, src8, dstStride
|
|
%else
|
|
cglobal %1_pixels%2_l2_shift5, 5, 5 ; dst, src16, src8, dstStride
|
|
%endif
|
|
movsxdifnidn r3, r3d
|
|
mov r4d, %2
|
|
.loop:
|
|
movu m0, [r1]
|
|
movu m1, [r1+%2]
|
|
movu m2, [r1+48]
|
|
movu m3, [r1+48+%2]
|
|
psraw m0, 5
|
|
psraw m1, 5
|
|
psraw m2, 5
|
|
psraw m3, 5
|
|
packuswb m0, m1
|
|
packuswb m2, m3
|
|
pavgb m0, [r2]
|
|
pavgb m2, [r2+%2]
|
|
op_%1 m0, [r0], m1
|
|
op_%1 m2, [r0+r3], m1
|
|
lea r2, [r2+2*%2]
|
|
add r1, 48*2
|
|
lea r0, [r0+2*r3]
|
|
sub r4d, 2
|
|
jne .loop
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_MMX mmxext
|
|
PIXELS_L2_SHIFT5 put, 8
|
|
PIXELS_L2_SHIFT5 avg, 8
|
|
|
|
INIT_XMM sse2
|
|
PIXELS_L2_SHIFT5 put, 16
|
|
PIXELS_L2_SHIFT5 avg, 16
|
|
|
|
%if ARCH_X86_64
|
|
%macro QPEL16_H_LOWPASS_L2_OP 1
|
|
cglobal %1_h264_qpel16_h_lowpass_l2, 5, 6, 16 ; dst, src, src2, dstStride, src2Stride
|
|
movsxdifnidn r3, r3d
|
|
movsxdifnidn r4, r4d
|
|
mov r5d, 16
|
|
pxor m15, m15
|
|
mova m14, [pw_5]
|
|
mova m13, [pw_16]
|
|
.loop:
|
|
lddqu m1, [r1+6]
|
|
lddqu m7, [r1-2]
|
|
mova m0, m1
|
|
punpckhbw m1, m15
|
|
punpcklbw m0, m15
|
|
punpcklbw m7, m15
|
|
mova m2, m1
|
|
mova m6, m0
|
|
mova m3, m1
|
|
mova m8, m0
|
|
mova m4, m1
|
|
mova m9, m0
|
|
mova m12, m0
|
|
mova m11, m1
|
|
palignr m11, m0, 10
|
|
palignr m12, m7, 10
|
|
palignr m4, m0, 2
|
|
palignr m9, m7, 2
|
|
palignr m3, m0, 4
|
|
palignr m8, m7, 4
|
|
palignr m2, m0, 6
|
|
palignr m6, m7, 6
|
|
paddw m11, m0
|
|
palignr m1, m0, 8
|
|
palignr m0, m7, 8
|
|
paddw m7, m12
|
|
paddw m2, m3
|
|
paddw m6, m8
|
|
paddw m1, m4
|
|
paddw m0, m9
|
|
psllw m2, 2
|
|
psllw m6, 2
|
|
psubw m2, m1
|
|
psubw m6, m0
|
|
paddw m11, m13
|
|
paddw m7, m13
|
|
pmullw m2, m14
|
|
pmullw m6, m14
|
|
lddqu m3, [r2]
|
|
paddw m2, m11
|
|
paddw m6, m7
|
|
psraw m2, 5
|
|
psraw m6, 5
|
|
packuswb m6, m2
|
|
pavgb m6, m3
|
|
op_%1 m6, [r0], m11
|
|
add r1, r3
|
|
add r0, r3
|
|
add r2, r4
|
|
dec r5d
|
|
jg .loop
|
|
RET
|
|
%endmacro
|
|
|
|
INIT_XMM ssse3
|
|
QPEL16_H_LOWPASS_L2_OP put
|
|
QPEL16_H_LOWPASS_L2_OP avg
|
|
%endif
|