generic-library/ffmpeg
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
Files
dd2d26dc82b8f0fcbcad3104cbee03e7ec80b263
ffmpeg/libswresample/x86/resample.asm
James Almer 5750d6c5e9 x86: move XOP emulation code back to x86inc 
Only two functions that use xop multiply-accumulate instructions where the
first operand is the same as the fourth actually took advantage of the macros.

This further reduces differences with x264's x86inc.

Reviewed-by: Ronald S. Bultje <rsbultje@gmail.com>
Signed-off-by: James Almer <jamrial@gmail.com>
2015-08-03 17:11:13 -03:00

606 lines
23 KiB
NASM
Raw Blame History

;******************************************************************************
;* Copyright (c) 2012 Michael Niedermayer
;* Copyright (c) 2014 James Almer <jamrial <at> gmail.com>
;* Copyright (c) 2014 Ronald S. Bultje <rsbultje@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"
%if ARCH_X86_64
%define pointer resq
%else
%define pointer resd
%endif
struc ResampleContext
.av_class: pointer 1
.filter_bank: pointer 1
.filter_length: resd 1
.filter_alloc: resd 1
.ideal_dst_incr: resd 1
.dst_incr: resd 1
.dst_incr_div: resd 1
.dst_incr_mod: resd 1
.index: resd 1
.frac: resd 1
.src_incr: resd 1
.compensation_distance: resd 1
.phase_shift: resd 1
.phase_mask: resd 1
; there's a few more here but we only care about the first few
endstruc
SECTION_RODATA
pf_1: dd 1.0
pdbl_1: dq 1.0
pd_0x4000: dd 0x4000
SECTION .text
%macro RESAMPLE_FNS 3-5 ; format [float or int16], bps, log2_bps, float op suffix [s or d], 1.0 constant
; int resample_common_$format(ResampleContext *ctx, $format *dst,
; const $format *src, int size, int update_ctx)
%if ARCH_X86_64 ; unix64 and win64
cglobal resample_common_%1, 0, 15, 2, ctx, dst, src, phase_shift, index, frac, \
dst_incr_mod, size, min_filter_count_x4, \
min_filter_len_x4, dst_incr_div, src_incr, \
phase_mask, dst_end, filter_bank
; use red-zone for variable storage
%define ctx_stackq [rsp-0x8]
%define src_stackq [rsp-0x10]
%if WIN64
%define update_context_stackd r4m
%else ; unix64
%define update_context_stackd [rsp-0x14]
%endif
; load as many variables in registers as possible; for the rest, store
; on stack so that we have 'ctx' available as one extra register
mov sized, r3d
mov phase_maskd, [ctxq+ResampleContext.phase_mask]
%if UNIX64
mov update_context_stackd, r4d
%endif
mov indexd, [ctxq+ResampleContext.index]
mov fracd, [ctxq+ResampleContext.frac]
mov dst_incr_modd, [ctxq+ResampleContext.dst_incr_mod]
mov filter_bankq, [ctxq+ResampleContext.filter_bank]
mov src_incrd, [ctxq+ResampleContext.src_incr]
mov ctx_stackq, ctxq
mov min_filter_len_x4d, [ctxq+ResampleContext.filter_length]
mov dst_incr_divd, [ctxq+ResampleContext.dst_incr_div]
shl min_filter_len_x4d, %3
lea dst_endq, [dstq+sizeq*%2]
%if UNIX64
mov ecx, [ctxq+ResampleContext.phase_shift]
mov edi, [ctxq+ResampleContext.filter_alloc]
DEFINE_ARGS filter_alloc, dst, src, phase_shift, index, frac, dst_incr_mod, \
filter, min_filter_count_x4, min_filter_len_x4, dst_incr_div, \
src_incr, phase_mask, dst_end, filter_bank
%elif WIN64
mov R9d, [ctxq+ResampleContext.filter_alloc]
mov ecx, [ctxq+ResampleContext.phase_shift]
DEFINE_ARGS phase_shift, dst, src, filter_alloc, index, frac, dst_incr_mod, \
filter, min_filter_count_x4, min_filter_len_x4, dst_incr_div, \
src_incr, phase_mask, dst_end, filter_bank
%endif
neg min_filter_len_x4q
sub filter_bankq, min_filter_len_x4q
sub srcq, min_filter_len_x4q
mov src_stackq, srcq
%else ; x86-32
cglobal resample_common_%1, 1, 7, 2, ctx, phase_shift, dst, frac, \
index, min_filter_length_x4, filter_bank
; push temp variables to stack
%define ctx_stackq r0mp
%define src_stackq r2mp
%define update_context_stackd r4m
mov dstq, r1mp
mov r3, r3mp
lea r3, [dstq+r3*%2]
PUSH dword [ctxq+ResampleContext.dst_incr_div]
PUSH dword [ctxq+ResampleContext.dst_incr_mod]
PUSH dword [ctxq+ResampleContext.filter_alloc]
PUSH r3
PUSH dword [ctxq+ResampleContext.phase_mask]
PUSH dword [ctxq+ResampleContext.src_incr]
mov min_filter_length_x4d, [ctxq+ResampleContext.filter_length]
mov indexd, [ctxq+ResampleContext.index]
shl min_filter_length_x4d, %3
mov fracd, [ctxq+ResampleContext.frac]
neg min_filter_length_x4q
mov filter_bankq, [ctxq+ResampleContext.filter_bank]
sub r2mp, min_filter_length_x4q
sub filter_bankq, min_filter_length_x4q
PUSH min_filter_length_x4q
PUSH filter_bankq
mov phase_shiftd, [ctxq+ResampleContext.phase_shift]
DEFINE_ARGS src, phase_shift, dst, frac, index, min_filter_count_x4, filter
%define filter_bankq dword [rsp+0x0]
%define min_filter_length_x4q dword [rsp+0x4]
%define src_incrd dword [rsp+0x8]
%define phase_maskd dword [rsp+0xc]
%define dst_endq dword [rsp+0x10]
%define filter_allocd dword [rsp+0x14]
%define dst_incr_modd dword [rsp+0x18]
%define dst_incr_divd dword [rsp+0x1c]
mov srcq, r2mp
%endif
.loop:
mov filterd, filter_allocd
imul filterd, indexd
%if ARCH_X86_64
mov min_filter_count_x4q, min_filter_len_x4q
lea filterq, [filter_bankq+filterq*%2]
%else ; x86-32
mov min_filter_count_x4q, filter_bankq
lea filterq, [min_filter_count_x4q+filterq*%2]
mov min_filter_count_x4q, min_filter_length_x4q
%endif
%ifidn %1, int16
movd m0, [pd_0x4000]
%else ; float/double
xorps m0, m0, m0
%endif
align 16
.inner_loop:
movu m1, [srcq+min_filter_count_x4q*1]
%ifidn %1, int16
%if cpuflag(xop)
vpmadcswd m0, m1, [filterq+min_filter_count_x4q*1], m0
%else
pmaddwd m1, [filterq+min_filter_count_x4q*1]
paddd m0, m1
%endif
%else ; float/double
%if cpuflag(fma4) || cpuflag(fma3)
fmaddp%4 m0, m1, [filterq+min_filter_count_x4q*1], m0
%else
mulp%4 m1, m1, [filterq+min_filter_count_x4q*1]
addp%4 m0, m0, m1
%endif ; cpuflag
%endif
add min_filter_count_x4q, mmsize
js .inner_loop
%ifidn %1, int16
HADDD m0, m1
psrad m0, 15
add fracd, dst_incr_modd
packssdw m0, m0
add indexd, dst_incr_divd
movd [dstq], m0
%else ; float/double
; horizontal sum & store
%if mmsize == 32
vextractf128 xm1, m0, 0x1
addps xm0, xm1
%endif
movhlps xm1, xm0
%ifidn %1, float
addps xm0, xm1
shufps xm1, xm0, xm0, q0001
%endif
add fracd, dst_incr_modd
addp%4 xm0, xm1
add indexd, dst_incr_divd
movs%4 [dstq], xm0
%endif
cmp fracd, src_incrd
jl .skip
sub fracd, src_incrd
inc indexd
%if UNIX64
DEFINE_ARGS filter_alloc, dst, src, phase_shift, index, frac, dst_incr_mod, \
index_incr, min_filter_count_x4, min_filter_len_x4, dst_incr_div, \
src_incr, phase_mask, dst_end, filter_bank
%elif WIN64
DEFINE_ARGS phase_shift, dst, src, filter_alloc, index, frac, dst_incr_mod, \
index_incr, min_filter_count_x4, min_filter_len_x4, dst_incr_div, \
src_incr, phase_mask, dst_end, filter_bank
%else ; x86-32
DEFINE_ARGS src, phase_shift, dst, frac, index, index_incr
%endif
.skip:
mov index_incrd, indexd
add dstq, %2
and indexd, phase_maskd
sar index_incrd, phase_shiftb
lea srcq, [srcq+index_incrq*%2]
cmp dstq, dst_endq
jne .loop
%if ARCH_X86_64
DEFINE_ARGS ctx, dst, src, phase_shift, index, frac
%else ; x86-32
DEFINE_ARGS src, ctx, update_context, frac, index
%endif
cmp dword update_context_stackd, 0
jz .skip_store
; strictly speaking, the function should always return the consumed
; number of bytes; however, we only use the value if update_context
; is true, so let's just leave it uninitialized otherwise
mov ctxq, ctx_stackq
movifnidn rax, srcq
mov [ctxq+ResampleContext.frac ], fracd
sub rax, src_stackq
mov [ctxq+ResampleContext.index], indexd
shr rax, %3
.skip_store:
%if ARCH_X86_32
ADD rsp, 0x20
%endif
RET
; int resample_linear_$format(ResampleContext *ctx, float *dst,
; const float *src, int size, int update_ctx)
%if ARCH_X86_64 ; unix64 and win64
%if UNIX64
cglobal resample_linear_%1, 0, 15, 5, ctx, dst, phase_mask, phase_shift, index, frac, \
size, dst_incr_mod, min_filter_count_x4, \
min_filter_len_x4, dst_incr_div, src_incr, \
src, dst_end, filter_bank
mov srcq, r2mp
%else ; win64
cglobal resample_linear_%1, 0, 15, 5, ctx, phase_mask, src, phase_shift, index, frac, \
size, dst_incr_mod, min_filter_count_x4, \
min_filter_len_x4, dst_incr_div, src_incr, \
dst, dst_end, filter_bank
mov dstq, r1mp
%endif
; use red-zone for variable storage
%define ctx_stackq [rsp-0x8]
%define src_stackq [rsp-0x10]
%define phase_mask_stackd [rsp-0x14]
%if WIN64
%define update_context_stackd r4m
%else ; unix64
%define update_context_stackd [rsp-0x18]
%endif
; load as many variables in registers as possible; for the rest, store
; on stack so that we have 'ctx' available as one extra register
mov sized, r3d
mov phase_maskd, [ctxq+ResampleContext.phase_mask]
%if UNIX64
mov update_context_stackd, r4d
%endif
mov indexd, [ctxq+ResampleContext.index]
mov fracd, [ctxq+ResampleContext.frac]
mov dst_incr_modd, [ctxq+ResampleContext.dst_incr_mod]
mov filter_bankq, [ctxq+ResampleContext.filter_bank]
mov src_incrd, [ctxq+ResampleContext.src_incr]
mov ctx_stackq, ctxq
mov phase_mask_stackd, phase_maskd
mov min_filter_len_x4d, [ctxq+ResampleContext.filter_length]
%ifidn %1, int16
movd m4, [pd_0x4000]
%else ; float/double
cvtsi2s%4 xm0, src_incrd
movs%4 xm4, [%5]
divs%4 xm4, xm0
%endif
mov dst_incr_divd, [ctxq+ResampleContext.dst_incr_div]
shl min_filter_len_x4d, %3
lea dst_endq, [dstq+sizeq*%2]
%if UNIX64
mov ecx, [ctxq+ResampleContext.phase_shift]
mov edi, [ctxq+ResampleContext.filter_alloc]
DEFINE_ARGS filter_alloc, dst, filter2, phase_shift, index, frac, filter1, \
dst_incr_mod, min_filter_count_x4, min_filter_len_x4, \
dst_incr_div, src_incr, src, dst_end, filter_bank
%elif WIN64
mov R9d, [ctxq+ResampleContext.filter_alloc]
mov ecx, [ctxq+ResampleContext.phase_shift]
DEFINE_ARGS phase_shift, filter2, src, filter_alloc, index, frac, filter1, \
dst_incr_mod, min_filter_count_x4, min_filter_len_x4, \
dst_incr_div, src_incr, dst, dst_end, filter_bank
%endif
neg min_filter_len_x4q
sub filter_bankq, min_filter_len_x4q
sub srcq, min_filter_len_x4q
mov src_stackq, srcq
%else ; x86-32
cglobal resample_linear_%1, 1, 7, 5, ctx, min_filter_length_x4, filter2, \
frac, index, dst, filter_bank
; push temp variables to stack
%define ctx_stackq r0mp
%define src_stackq r2mp
%define update_context_stackd r4m
mov dstq, r1mp
mov r3, r3mp
lea r3, [dstq+r3*%2]
PUSH dword [ctxq+ResampleContext.dst_incr_div]
PUSH r3
mov r3, dword [ctxq+ResampleContext.filter_alloc]
PUSH dword [ctxq+ResampleContext.dst_incr_mod]
PUSH r3
shl r3, %3
PUSH r3
mov r3, dword [ctxq+ResampleContext.src_incr]
PUSH dword [ctxq+ResampleContext.phase_mask]
PUSH r3d
%ifidn %1, int16
movd m4, [pd_0x4000]
%else ; float/double
cvtsi2s%4 xm0, r3d
movs%4 xm4, [%5]
divs%4 xm4, xm0
%endif
mov min_filter_length_x4d, [ctxq+ResampleContext.filter_length]
mov indexd, [ctxq+ResampleContext.index]
shl min_filter_length_x4d, %3
mov fracd, [ctxq+ResampleContext.frac]
neg min_filter_length_x4q
mov filter_bankq, [ctxq+ResampleContext.filter_bank]
sub r2mp, min_filter_length_x4q
sub filter_bankq, min_filter_length_x4q
PUSH min_filter_length_x4q
PUSH filter_bankq
PUSH dword [ctxq+ResampleContext.phase_shift]
DEFINE_ARGS filter1, min_filter_count_x4, filter2, frac, index, dst, src
%define phase_shift_stackd dword [rsp+0x0]
%define filter_bankq dword [rsp+0x4]
%define min_filter_length_x4q dword [rsp+0x8]
%define src_incrd dword [rsp+0xc]
%define phase_mask_stackd dword [rsp+0x10]
%define filter_alloc_x4q dword [rsp+0x14]
%define filter_allocd dword [rsp+0x18]
%define dst_incr_modd dword [rsp+0x1c]
%define dst_endq dword [rsp+0x20]
%define dst_incr_divd dword [rsp+0x24]
mov srcq, r2mp
%endif
.loop:
mov filter1d, filter_allocd
imul filter1d, indexd
%if ARCH_X86_64
mov min_filter_count_x4q, min_filter_len_x4q
lea filter1q, [filter_bankq+filter1q*%2]
lea filter2q, [filter1q+filter_allocq*%2]
%else ; x86-32
mov min_filter_count_x4q, filter_bankq
lea filter1q, [min_filter_count_x4q+filter1q*%2]
mov min_filter_count_x4q, min_filter_length_x4q
mov filter2q, filter1q
add filter2q, filter_alloc_x4q
%endif
%ifidn %1, int16
mova m0, m4
mova m2, m4
%else ; float/double
xorps m0, m0, m0
xorps m2, m2, m2
%endif
align 16
.inner_loop:
movu m1, [srcq+min_filter_count_x4q*1]
%ifidn %1, int16
%if cpuflag(xop)
vpmadcswd m2, m1, [filter2q+min_filter_count_x4q*1], m2
vpmadcswd m0, m1, [filter1q+min_filter_count_x4q*1], m0
%else
pmaddwd m3, m1, [filter2q+min_filter_count_x4q*1]
pmaddwd m1, [filter1q+min_filter_count_x4q*1]
paddd m2, m3
paddd m0, m1
%endif ; cpuflag
%else ; float/double
%if cpuflag(fma4) || cpuflag(fma3)
fmaddp%4 m2, m1, [filter2q+min_filter_count_x4q*1], m2
fmaddp%4 m0, m1, [filter1q+min_filter_count_x4q*1], m0
%else
mulp%4 m3, m1, [filter2q+min_filter_count_x4q*1]
mulp%4 m1, m1, [filter1q+min_filter_count_x4q*1]
addp%4 m2, m2, m3
addp%4 m0, m0, m1
%endif ; cpuflag
%endif
add min_filter_count_x4q, mmsize
js .inner_loop
%ifidn %1, int16
%if mmsize == 16
%if cpuflag(xop)
vphadddq m2, m2
vphadddq m0, m0
%endif
pshufd m3, m2, q0032
pshufd m1, m0, q0032
paddd m2, m3
paddd m0, m1
%endif
%if notcpuflag(xop)
PSHUFLW m3, m2, q0032
PSHUFLW m1, m0, q0032
paddd m2, m3
paddd m0, m1
%endif
psubd m2, m0
; This is probably a really bad idea on atom and other machines with a
; long transfer latency between GPRs and XMMs (atom). However, it does
; make the clip a lot simpler...
movd eax, m2
add indexd, dst_incr_divd
imul fracd
idiv src_incrd
movd m1, eax
add fracd, dst_incr_modd
paddd m0, m1
psrad m0, 15
packssdw m0, m0
movd [dstq], m0
; note that for imul/idiv, I need to move filter to edx/eax for each:
; - 32bit: eax=r0[filter1], edx=r2[filter2]
; - win64: eax=r6[filter1], edx=r1[todo]
; - unix64: eax=r6[filter1], edx=r2[todo]
%else ; float/double
; val += (v2 - val) * (FELEML) frac / c->src_incr;
%if mmsize == 32
vextractf128 xm1, m0, 0x1
vextractf128 xm3, m2, 0x1
addps xm0, xm1
addps xm2, xm3
%endif
cvtsi2s%4 xm1, fracd
subp%4 xm2, xm0
mulp%4 xm1, xm4
shufp%4 xm1, xm1, q0000
%if cpuflag(fma4) || cpuflag(fma3)
fmaddp%4 xm0, xm2, xm1, xm0
%else
mulp%4 xm2, xm1
addp%4 xm0, xm2
%endif ; cpuflag
; horizontal sum & store
movhlps xm1, xm0
%ifidn %1, float
addps xm0, xm1
shufps xm1, xm0, xm0, q0001
%endif
add fracd, dst_incr_modd
addp%4 xm0, xm1
add indexd, dst_incr_divd
movs%4 [dstq], xm0
%endif
cmp fracd, src_incrd
jl .skip
sub fracd, src_incrd
inc indexd
%if UNIX64
DEFINE_ARGS filter_alloc, dst, filter2, phase_shift, index, frac, index_incr, \
dst_incr_mod, min_filter_count_x4, min_filter_len_x4, \
dst_incr_div, src_incr, src, dst_end, filter_bank
%elif WIN64
DEFINE_ARGS phase_shift, filter2, src, filter_alloc, index, frac, index_incr, \
dst_incr_mod, min_filter_count_x4, min_filter_len_x4, \
dst_incr_div, src_incr, dst, dst_end, filter_bank
%else ; x86-32
DEFINE_ARGS filter1, phase_shift, index_incr, frac, index, dst, src
%endif
.skip:
%if ARCH_X86_32
mov phase_shiftd, phase_shift_stackd
%endif
mov index_incrd, indexd
add dstq, %2
and indexd, phase_mask_stackd
sar index_incrd, phase_shiftb
lea srcq, [srcq+index_incrq*%2]
cmp dstq, dst_endq
jne .loop
%if UNIX64
DEFINE_ARGS ctx, dst, filter2, phase_shift, index, frac, index_incr, \
dst_incr_mod, min_filter_count_x4, min_filter_len_x4, \
dst_incr_div, src_incr, src, dst_end, filter_bank
%elif WIN64
DEFINE_ARGS ctx, filter2, src, phase_shift, index, frac, index_incr, \
dst_incr_mod, min_filter_count_x4, min_filter_len_x4, \
dst_incr_div, src_incr, dst, dst_end, filter_bank
%else ; x86-32
DEFINE_ARGS filter1, ctx, update_context, frac, index, dst, src
%endif
cmp dword update_context_stackd, 0
jz .skip_store
; strictly speaking, the function should always return the consumed
; number of bytes; however, we only use the value if update_context
; is true, so let's just leave it uninitialized otherwise
mov ctxq, ctx_stackq
movifnidn rax, srcq
mov [ctxq+ResampleContext.frac ], fracd
sub rax, src_stackq
mov [ctxq+ResampleContext.index], indexd
shr rax, %3
.skip_store:
%if ARCH_X86_32
ADD rsp, 0x28
%endif
RET
%endmacro
INIT_XMM sse
RESAMPLE_FNS float, 4, 2, s, pf_1
%if HAVE_AVX_EXTERNAL
INIT_YMM avx
RESAMPLE_FNS float, 4, 2, s, pf_1
%endif
%if HAVE_FMA3_EXTERNAL
INIT_YMM fma3
RESAMPLE_FNS float, 4, 2, s, pf_1
%endif
%if HAVE_FMA4_EXTERNAL
INIT_XMM fma4
RESAMPLE_FNS float, 4, 2, s, pf_1
%endif
%if ARCH_X86_32
INIT_MMX mmxext
RESAMPLE_FNS int16, 2, 1
%endif
INIT_XMM sse2
RESAMPLE_FNS int16, 2, 1
%if HAVE_XOP_EXTERNAL
INIT_XMM xop
RESAMPLE_FNS int16, 2, 1
%endif
INIT_XMM sse2
RESAMPLE_FNS double, 8, 3, d, pdbl_1
Reference in New Issue View Git Blame Copy Permalink