virtualenv/FFmpeg
1
0
Fork 0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-18 03:19:31 +02:00
Code Issues Releases Activity
e71df841ee
FFmpeg/libavcodec/x86/rv34dsp.asm

198 lines
5.8 KiB
NASM
Raw Normal View History

rv34: DC-only inverse transform When decoding coefficients, detect whether the block is DC-only, and take advantage of this knowledge to perform DC-only inverse transform. This is achieved by: - first, changing the 108x4 element modulo_three_table into a 108 element table (kind of base4), and accessing each value using mask and shifts. - then, checking low bits for 0 (as they represent the presence of higher frequency coefficients) Also provide x86 SIMD code for the DC-only inverse transform. Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
2012-01-01 20:33:22 +03:00
;******************************************************************************
;* MMX/SSE2-optimized functions for the RV30 and RV40 decoders
;* Copyright (C) 2012 Christophe Gisquet <christophe.gisquet@gmail.com>
;*
;* This file is part of Libav.
;*
;* Libav 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.
;*
;* Libav 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 Libav; if not, write to the Free Software
;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
;******************************************************************************
%include "x86inc.asm"
%include "x86util.asm"
rv34dsp x86: implement MMX2 inverse transform 141 cycles down to 51. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2012-03-29 23:37:41 +03:00
SECTION_RODATA
pw_row_coeffs: times 4 dw 13
times 4 dw 17
times 4 dw 7
pd_512: times 2 dd 0x200
pw_col_coeffs: dw 13, 13, 13, -13
dw 17, 7, 7, -17
dw 13, -13, 13, 13
dw -7, 17, -17, -7
rv34: DC-only inverse transform When decoding coefficients, detect whether the block is DC-only, and take advantage of this knowledge to perform DC-only inverse transform. This is achieved by: - first, changing the 108x4 element modulo_three_table into a 108 element table (kind of base4), and accessing each value using mask and shifts. - then, checking low bits for 0 (as they represent the presence of higher frequency coefficients) Also provide x86 SIMD code for the DC-only inverse transform. Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
2012-01-01 20:33:22 +03:00
SECTION .text
%macro IDCT_DC_NOROUND 1
imul %1, 13*13*3
sar %1, 11
%endmacro
%macro IDCT_DC_ROUND 1
imul %1, 13*13
add %1, 0x200
sar %1, 10
%endmacro
rv34: Intra 16x16 handling Extract processing of intra 16x16 blocks from intra macroblock processing. Also implement a function performing inverse transform and block reconstruction for DC-only blocks in 1 pass instead of 2.
2012-01-03 02:22:11 +03:00
%macro rv34_idct 1
rv34: port x86 SIMD to cpuflags.
2012-07-27 08:19:19 +03:00
cglobal rv34_idct_%1, 1, 2, 0
rv34: DC-only inverse transform When decoding coefficients, detect whether the block is DC-only, and take advantage of this knowledge to perform DC-only inverse transform. This is achieved by: - first, changing the 108x4 element modulo_three_table into a 108 element table (kind of base4), and accessing each value using mask and shifts. - then, checking low bits for 0 (as they represent the presence of higher frequency coefficients) Also provide x86 SIMD code for the DC-only inverse transform. Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
2012-01-01 20:33:22 +03:00
movsx r1, word [r0]
IDCT_DC r1
Merge remote-tracking branch 'qatar/master' * qatar/master: rv34: add NEON rv34_idct_add rv34: 1-pass inter MB reconstruction add SMJPEG muxer avformat: split out common SMJPEG code pictordec: Use bytestream2 functions avconv: use avcodec_encode_audio2() pcmenc: use AVCodec.encode2() avcodec: bump minor version and add APIChanges for the new audio encoding API avcodec: Add avcodec_encode_audio2() as replacement for avcodec_encode_audio() avcodec: add a public function, avcodec_fill_audio_frame(). rv34: Intra 16x16 handling rv34: Inter/intra MB code split Conflicts: Changelog libavcodec/avcodec.h libavcodec/pictordec.c libavcodec/utils.c libavcodec/version.h libavcodec/x86/rv34dsp.asm libavformat/version.h Merged-by: Michael Niedermayer <michaelni@gmx.at>
2012-01-17 03:40:45 +03:00
movd m0, r1d
rv34: Intra 16x16 handling Extract processing of intra 16x16 blocks from intra macroblock processing. Also implement a function performing inverse transform and block reconstruction for DC-only blocks in 1 pass instead of 2.
2012-01-03 02:22:11 +03:00
pshufw m0, m0, 0
movq [r0+ 0], m0
rv34: 1-pass inter MB reconstruction Implement 1-pass inverse transform and reconstruction for inter blocks.
2012-01-03 22:38:29 +03:00
movq [r0+ 8], m0
rv34: Intra 16x16 handling Extract processing of intra 16x16 blocks from intra macroblock processing. Also implement a function performing inverse transform and block reconstruction for DC-only blocks in 1 pass instead of 2.
2012-01-03 02:22:11 +03:00
movq [r0+16], m0
rv34: 1-pass inter MB reconstruction Implement 1-pass inverse transform and reconstruction for inter blocks.
2012-01-03 22:38:29 +03:00
movq [r0+24], m0
rv34: DC-only inverse transform When decoding coefficients, detect whether the block is DC-only, and take advantage of this knowledge to perform DC-only inverse transform. This is achieved by: - first, changing the 108x4 element modulo_three_table into a 108 element table (kind of base4), and accessing each value using mask and shifts. - then, checking low bits for 0 (as they represent the presence of higher frequency coefficients) Also provide x86 SIMD code for the DC-only inverse transform. Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
2012-01-01 20:33:22 +03:00
REP_RET
%endmacro
rv34: port x86 SIMD to cpuflags.
2012-07-27 08:19:19 +03:00
INIT_MMX mmx2
rv34: DC-only inverse transform When decoding coefficients, detect whether the block is DC-only, and take advantage of this knowledge to perform DC-only inverse transform. This is achieved by: - first, changing the 108x4 element modulo_three_table into a 108 element table (kind of base4), and accessing each value using mask and shifts. - then, checking low bits for 0 (as they represent the presence of higher frequency coefficients) Also provide x86 SIMD code for the DC-only inverse transform. Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
2012-01-01 20:33:22 +03:00
%define IDCT_DC IDCT_DC_ROUND
rv34: Intra 16x16 handling Extract processing of intra 16x16 blocks from intra macroblock processing. Also implement a function performing inverse transform and block reconstruction for DC-only blocks in 1 pass instead of 2.
2012-01-03 02:22:11 +03:00
rv34_idct dc
rv34: DC-only inverse transform When decoding coefficients, detect whether the block is DC-only, and take advantage of this knowledge to perform DC-only inverse transform. This is achieved by: - first, changing the 108x4 element modulo_three_table into a 108 element table (kind of base4), and accessing each value using mask and shifts. - then, checking low bits for 0 (as they represent the presence of higher frequency coefficients) Also provide x86 SIMD code for the DC-only inverse transform. Signed-off-by: Kostya Shishkov <kostya.shishkov@gmail.com>
2012-01-01 20:33:22 +03:00
%define IDCT_DC IDCT_DC_NOROUND
rv34: Intra 16x16 handling Extract processing of intra 16x16 blocks from intra macroblock processing. Also implement a function performing inverse transform and block reconstruction for DC-only blocks in 1 pass instead of 2.
2012-01-03 02:22:11 +03:00
rv34_idct dc_noround
; ff_rv34_idct_dc_add_mmx(uint8_t *dst, int stride, int dc);
rv34: port x86 SIMD to cpuflags.
2012-07-27 08:19:19 +03:00
INIT_MMX mmx
cglobal rv34_idct_dc_add, 3, 3
rv34: Intra 16x16 handling Extract processing of intra 16x16 blocks from intra macroblock processing. Also implement a function performing inverse transform and block reconstruction for DC-only blocks in 1 pass instead of 2.
2012-01-03 02:22:11 +03:00
; calculate DC
IDCT_DC_ROUND r2
pxor m1, m1
Use correct register size. Fixes compilation with NASM. Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
2012-01-17 10:41:04 +03:00
movd m0, r2d
rv34: Intra 16x16 handling Extract processing of intra 16x16 blocks from intra macroblock processing. Also implement a function performing inverse transform and block reconstruction for DC-only blocks in 1 pass instead of 2.
2012-01-03 02:22:11 +03:00
psubw m1, m0
packuswb m0, m0
packuswb m1, m1
punpcklbw m0, m0
punpcklbw m1, m1
punpcklwd m0, m0
punpcklwd m1, m1
; add DC
lea r2, [r0+r1*2]
movh m2, [r0]
movh m3, [r0+r1]
movh m4, [r2]
movh m5, [r2+r1]
paddusb m2, m0
paddusb m3, m0
paddusb m4, m0
paddusb m5, m0
psubusb m2, m1
psubusb m3, m1
psubusb m4, m1
psubusb m5, m1
movh [r0], m2
movh [r0+r1], m3
movh [r2], m4
movh [r2+r1], m5
RET
rv34dsp x86: implement MMX2 inverse transform 141 cycles down to 51. Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>
2012-03-29 23:37:41 +03:00
; Load coeffs and perform row transform
; Output: coeffs in mm[0467], rounder in mm5
%macro ROW_TRANSFORM 1
pxor mm7, mm7
mova mm0, [%1+ 0*8]
mova mm1, [%1+ 1*8]
mova mm2, [%1+ 2*8]
mova mm3, [%1+ 3*8]
mova [%1+ 0*8], mm7
mova [%1+ 1*8], mm7
mova [%1+ 2*8], mm7
mova [%1+ 3*8], mm7
mova mm4, mm0
mova mm6, [pw_row_coeffs+ 0]
paddsw mm0, mm2 ; b0 + b2
psubsw mm4, mm2 ; b0 - b2
pmullw mm0, mm6 ; *13 = z0
pmullw mm4, mm6 ; *13 = z1
mova mm5, mm1
pmullw mm1, [pw_row_coeffs+ 8] ; b1*17
pmullw mm5, [pw_row_coeffs+16] ; b1* 7
mova mm7, mm3
pmullw mm3, [pw_row_coeffs+ 8] ; b3*17
pmullw mm7, [pw_row_coeffs+16] ; b3* 7
paddsw mm1, mm7 ; z3 = b1*17 + b3* 7
psubsw mm5, mm3 ; z2 = b1* 7 - b3*17
mova mm7, mm0
mova mm6, mm4
paddsw mm0, mm1 ; z0 + z3
psubsw mm7, mm1 ; z0 - z3
paddsw mm4, mm5 ; z1 + z2
psubsw mm6, mm5 ; z1 - z2
mova mm5, [pd_512] ; 0x200
%endmacro
; ff_rv34_idct_add_mmx2(uint8_t *dst, ptrdiff_t stride, DCTELEM *block);
%macro COL_TRANSFORM 4
pshufw mm3, %2, 0xDD ; col. 1,3,1,3
pshufw %2, %2, 0x88 ; col. 0,2,0,2
pmaddwd %2, %3 ; 13*c0+13*c2 | 13*c0-13*c2 = z0 | z1
pmaddwd mm3, %4 ; 17*c1+ 7*c3 | 7*c1-17*c3 = z3 | z2
paddd %2, mm5
pshufw mm1, %2, 01001110b ; z1 | z0
pshufw mm2, mm3, 01001110b ; z2 | z3
paddd %2, mm3 ; z0+z3 | z1+z2
psubd mm1, mm2 ; z1-z2 | z0-z3
movd mm3, %1
psrad %2, 10
pxor mm2, mm2
psrad mm1, 10
punpcklbw mm3, mm2
packssdw %2, mm1
paddw %2, mm3
packuswb %2, %2
movd %1, %2
%endmacro
INIT_MMX mmx2
cglobal rv34_idct_add, 3,3,0, d, s, b
ROW_TRANSFORM bq
COL_TRANSFORM [dq], mm0, [pw_col_coeffs+ 0], [pw_col_coeffs+ 8]
mova mm0, [pw_col_coeffs+ 0]
COL_TRANSFORM [dq+sq], mm4, mm0, [pw_col_coeffs+ 8]
mova mm4, [pw_col_coeffs+ 8]
lea dq, [dq + 2*sq]
COL_TRANSFORM [dq], mm6, mm0, mm4
COL_TRANSFORM [dq+sq], mm7, mm0, mm4
ret
rv34: Intra 16x16 handling Extract processing of intra 16x16 blocks from intra macroblock processing. Also implement a function performing inverse transform and block reconstruction for DC-only blocks in 1 pass instead of 2.
2012-01-03 02:22:11 +03:00
; ff_rv34_idct_dc_add_sse4(uint8_t *dst, int stride, int dc);
rv34: port x86 SIMD to cpuflags.
2012-07-27 08:19:19 +03:00
INIT_XMM sse4
cglobal rv34_idct_dc_add, 3, 3, 6
rv34: Intra 16x16 handling Extract processing of intra 16x16 blocks from intra macroblock processing. Also implement a function performing inverse transform and block reconstruction for DC-only blocks in 1 pass instead of 2.
2012-01-03 02:22:11 +03:00
; load data
IDCT_DC_ROUND r2
pxor m1, m1
; calculate DC
Use correct register size. Fixes compilation with NASM. Signed-off-by: Reimar Döffinger <Reimar.Doeffinger@gmx.de>
2012-01-17 10:41:04 +03:00
movd m0, r2d
rv34: Intra 16x16 handling Extract processing of intra 16x16 blocks from intra macroblock processing. Also implement a function performing inverse transform and block reconstruction for DC-only blocks in 1 pass instead of 2.
2012-01-03 02:22:11 +03:00
lea r2, [r0+r1*2]
movd m2, [r0]
movd m3, [r0+r1]
pshuflw m0, m0, 0
movd m4, [r2]
movd m5, [r2+r1]
punpcklqdq m0, m0
punpckldq m2, m3
punpckldq m4, m5
punpcklbw m2, m1
punpcklbw m4, m1
paddw m2, m0
paddw m4, m0
packuswb m2, m4
movd [r0], m2
pextrd [r0+r1], m2, 1
pextrd [r2], m2, 2
pextrd [r2+r1], m2, 3
RET
Reference in New Issue Copy Permalink