ref: caca57251642b5fc2a71dee3034cde300ff0c621
dir: /src/x86/cdef.asm/
; Copyright © 2018, VideoLAN and dav1d authors
; Copyright © 2018, Two Orioles, LLC
; All rights reserved.
;
; Redistribution and use in source and binary forms, with or without
; modification, are permitted provided that the following conditions are met:
;
; 1. Redistributions of source code must retain the above copyright notice, this
; list of conditions and the following disclaimer.
;
; 2. Redistributions in binary form must reproduce the above copyright notice,
; this list of conditions and the following disclaimer in the documentation
; and/or other materials provided with the distribution.
;
; THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
; ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
; WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
; DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
; ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
; (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
; ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
; (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
; SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
%include "config.asm"
%include "ext/x86/x86inc.asm"
%if ARCH_X86_64
SECTION_RODATA 32
pd_04512763: dd 0, 4, 5, 1, 2, 7, 6, 3
div_table: dd 840, 420, 280, 210, 168, 140, 120, 105
dd 420, 210, 140, 105
pd_04261537: dd 0, 4, 2, 6, 1, 5, 3, 7
shufw_6543210x: db 12, 13, 10, 11, 8, 9, 6, 7, 4, 5, 2, 3, 0, 1, 14, 15
shufw_210xxxxx: db 4, 5, 2, 3, 0, 1, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15
pw_128: times 2 dw 128
pw_2048: times 2 dw 2048
tap_table: dw 4, 2, 3, 3, 2, 1
db -1 * 16 + 1, -2 * 16 + 2
db 0 * 16 + 1, -1 * 16 + 2
db 0 * 16 + 1, 0 * 16 + 2
db 0 * 16 + 1, 1 * 16 + 2
db 1 * 16 + 1, 2 * 16 + 2
db 1 * 16 + 0, 2 * 16 + 1
db 1 * 16 + 0, 2 * 16 + 0
db 1 * 16 + 0, 2 * 16 - 1
; the last 6 are repeats of the first 6 so we don't need to & 7
db -1 * 16 + 1, -2 * 16 + 2
db 0 * 16 + 1, -1 * 16 + 2
db 0 * 16 + 1, 0 * 16 + 2
db 0 * 16 + 1, 1 * 16 + 2
db 1 * 16 + 1, 2 * 16 + 2
db 1 * 16 + 0, 2 * 16 + 1
SECTION .text
%macro ACCUMULATE_TAP 6 ; tap_offset, shift, strength, mul_tap, w, stride
; load p0/p1
movsx offq, byte [dirq+kq+%1] ; off1
%if %5 == 4
movq xm5, [stkq+offq*2+%6*0] ; p0
movq xm6, [stkq+offq*2+%6*2]
movhps xm5, [stkq+offq*2+%6*1]
movhps xm6, [stkq+offq*2+%6*3]
vinserti128 m5, xm6, 1
%else
movu xm5, [stkq+offq*2+%6*0] ; p0
vinserti128 m5, [stkq+offq*2+%6*1], 1
%endif
neg offq ; -off1
%if %5 == 4
movq xm6, [stkq+offq*2+%6*0] ; p1
movq xm9, [stkq+offq*2+%6*2]
movhps xm6, [stkq+offq*2+%6*1]
movhps xm9, [stkq+offq*2+%6*3]
vinserti128 m6, xm9, 1
%else
movu xm6, [stkq+offq*2+%6*0] ; p1
vinserti128 m6, [stkq+offq*2+%6*1], 1
%endif
pcmpeqw m9, m14, m5
pcmpeqw m10, m14, m6
pandn m9, m5
pandn m10, m6
pmaxsw m7, m9 ; max after p0
pminsw m8, m5 ; min after p0
pmaxsw m7, m10 ; max after p1
pminsw m8, m6 ; min after p1
; accumulate sum[m15] over p0/p1
psubw m5, m4 ; diff_p0(p0 - px)
psubw m6, m4 ; diff_p1(p1 - px)
pabsw m9, m5
pabsw m10, m6
psraw m11, m9, %2
psraw m12, m10, %2
psubw m11, %3, m11
psubw m12, %3, m12
pmaxsw m11, m13
pmaxsw m12, m13
pminsw m11, m9
pminsw m12, m10
psignw m11, m5 ; constrain(diff_p0)
psignw m12, m6 ; constrain(diff_p1)
pmullw m11, %4 ; constrain(diff_p0) * pri_taps
pmullw m12, %4 ; constrain(diff_p1) * pri_taps
paddw m15, m11
paddw m15, m12
%endmacro
%macro cdef_filter_fn 3 ; w, h, stride
INIT_YMM avx2
cglobal cdef_filter_%1x%2, 4, 9, 16, 2 * 16 + (%2+4)*%3, \
dst, stride, left, top, pri, sec, stride3, dst4, edge
%define px rsp+2*16+2*%3
pcmpeqw m14, m14
psrlw m14, 1 ; 0x7fff
mov edged, r8m
; prepare pixel buffers - body/right
%if %1 == 4
INIT_XMM avx2
%endif
%if %2 == 8
lea dst4q, [dstq+strideq*4]
%endif
lea stride3q, [strideq*3]
test edged, 2 ; have_right
jz .no_right
pmovzxbw m1, [dstq+strideq*0]
pmovzxbw m2, [dstq+strideq*1]
pmovzxbw m3, [dstq+strideq*2]
pmovzxbw m4, [dstq+stride3q]
mova [px+0*%3], m1
mova [px+1*%3], m2
mova [px+2*%3], m3
mova [px+3*%3], m4
%if %2 == 8
pmovzxbw m1, [dst4q+strideq*0]
pmovzxbw m2, [dst4q+strideq*1]
pmovzxbw m3, [dst4q+strideq*2]
pmovzxbw m4, [dst4q+stride3q]
mova [px+4*%3], m1
mova [px+5*%3], m2
mova [px+6*%3], m3
mova [px+7*%3], m4
%endif
jmp .body_done
.no_right:
%if %1 == 4
movd xm1, [dstq+strideq*0]
movd xm2, [dstq+strideq*2]
pinsrd xm1, [dstq+strideq*1], 1
pinsrd xm2, [dstq+stride3q], 1
pmovzxbw xm1, xm1
pmovzxbw xm2, xm2
movq [px+0*%3], xm1
movhps [px+1*%3], xm1
movq [px+2*%3], xm2
movhps [px+3*%3], xm2
%else
pmovzxbw xm1, [dstq+strideq*0]
pmovzxbw xm2, [dstq+strideq*1]
pmovzxbw xm3, [dstq+strideq*2]
pmovzxbw xm4, [dstq+stride3q]
mova [px+0*%3], xm1
mova [px+1*%3], xm2
mova [px+2*%3], xm3
mova [px+3*%3], xm4
%endif
movd [px+0*%3+%1*2], xm14
movd [px+1*%3+%1*2], xm14
movd [px+2*%3+%1*2], xm14
movd [px+3*%3+%1*2], xm14
%if %2 == 8
; FIXME w == 4
movd [px+0*%3+%1*2], xm14
movd [px+1*%3+%1*2], xm14
movd [px+2*%3+%1*2], xm14
movd [px+3*%3+%1*2], xm14
pmovzxbw xm1, [dst4q+strideq*0]
pmovzxbw xm2, [dst4q+strideq*1]
pmovzxbw xm3, [dst4q+strideq*2]
pmovzxbw xm4, [dst4q+stride3q]
mova [px+4*%3], xm1
mova [px+5*%3], xm2
mova [px+6*%3], xm3
mova [px+7*%3], xm4
movd [px+4*%3+%1*2], xm14
movd [px+5*%3+%1*2], xm14
movd [px+6*%3+%1*2], xm14
movd [px+7*%3+%1*2], xm14
%endif
.body_done:
; top
DEFINE_ARGS dst, stride, left, top2, pri, sec, stride3, top1, edge
test edged, 4 ; have_top
jz .no_top
mov top1q, [top2q+0*gprsize]
mov top2q, [top2q+1*gprsize]
test edged, 1 ; have_left
jz .top_no_left
test edged, 2 ; have_right
jz .top_no_right
pmovzxbw m1, [top1q-(%1/2)]
pmovzxbw m2, [top2q-(%1/2)]
movu [px-2*%3-%1], m1
movu [px-1*%3-%1], m2
jmp .top_done
.top_no_right:
pmovzxbw m1, [top1q-%1]
pmovzxbw m2, [top2q-%1]
movu [px-2*%3-%1*2], m1
movu [px-1*%3-%1*2], m2
movd [px-2*%3+%1*2], xm14
movd [px-1*%3+%1*2], xm14
jmp .top_done
.top_no_left:
test edged, 2 ; have_right
jz .top_no_left_right
pmovzxbw m1, [top1q]
pmovzxbw m2, [top2q]
mova [px-2*%3+0], m1
mova [px-1*%3+0], m2
movd [px-2*%3-4], xm14
movd [px-1*%3-4], xm14
jmp .top_done
.top_no_left_right:
%if %1 == 4
movd xm1, [top1q]
pinsrd xm1, [top2q], 1
pmovzxbw xm1, xm1
movq [px-2*%3+0], xm1
movhps [px-1*%3+0], xm1
%else
pmovzxbw xm1, [top1q]
pmovzxbw xm2, [top2q]
mova [px-2*%3+0], xm1
mova [px-1*%3+0], xm2
%endif
movd [px-2*%3-4], xm14
movd [px-1*%3-4], xm14
movd [px-2*%3+%1*2], xm14
movd [px-1*%3+%1*2], xm14
jmp .top_done
.no_top:
movu [px-2*%3-%1], m14
movu [px-1*%3-%1], m14
.top_done:
; left
test edged, 1 ; have_left
jz .no_left
pmovzxbw xm1, [leftq+ 0]
%if %2 == 8
pmovzxbw xm2, [leftq+ 8]
%endif
movd [px+0*32-4], xm1
pextrd [px+1*32-4], xm1, 1
pextrd [px+2*32-4], xm1, 2
pextrd [px+3*32-4], xm1, 3
%if %2 == 8
movd [px+4*32-4], xm2
pextrd [px+5*32-4], xm2, 1
pextrd [px+6*32-4], xm2, 2
pextrd [px+7*32-4], xm2, 3
%endif
jmp .left_done
.no_left:
movd [px+0*%3-4], xm14
movd [px+1*%3-4], xm14
movd [px+2*%3-4], xm14
movd [px+3*%3-4], xm14
%if %2 == 8
movd [px+4*%3-4], xm14
movd [px+5*%3-4], xm14
movd [px+6*%3-4], xm14
movd [px+7*%3-4], xm14
%endif
.left_done:
; bottom
DEFINE_ARGS dst, stride, dst8, dummy1, pri, sec, stride3, dummy3, edge
test edged, 8 ; have_bottom
jz .no_bottom
lea dst8q, [dstq+%2*strideq]
test edged, 1 ; have_left
jz .bottom_no_left
test edged, 2 ; have_right
jz .bottom_no_right
pmovzxbw m1, [dst8q-(%1/2)]
pmovzxbw m2, [dst8q+strideq-(%1/2)]
movu [px+(%2+0)*%3-%1], m1
movu [px+(%2+1)*%3-%1], m2
jmp .bottom_done
.bottom_no_right:
pmovzxbw m1, [dst8q-%1]
pmovzxbw m2, [dst8q+strideq-%1]
movu [px+(%2+0)*%3-%1*2], m1
movu [px+(%2+1)*%3-%1*2], m2
%if %1 == 8
movd [px+(%2-1)*%3+%1*2], xm14 ; overwritten by previous movu
%endif
movd [px+(%2+0)*%3+%1*2], xm14
movd [px+(%2+1)*%3+%1*2], xm14
jmp .bottom_done
.bottom_no_left:
test edged, 2 ; have_right
jz .bottom_no_left_right
pmovzxbw m1, [dst8q]
pmovzxbw m2, [dst8q+strideq]
mova [px+(%2+0)*%3+0], m1
mova [px+(%2+1)*%3+0], m2
movd [px+(%2+0)*%3-4], xm14
movd [px+(%2+1)*%3-4], xm14
jmp .bottom_done
.bottom_no_left_right:
%if %1 == 4
movd xm1, [dst8q]
pinsrd xm1, [dst8q+strideq], 1
pmovzxbw xm1, xm1
movq [px+(%2+0)*%3+0], xm1
movhps [px+(%2+1)*%3+0], xm1
%else
pmovzxbw xm1, [dst8q]
pmovzxbw xm2, [dst8q+strideq]
mova [px+(%2+0)*%3+0], xm1
mova [px+(%2+1)*%3+0], xm2
%endif
movd [px+(%2+0)*%3-4], xm14
movd [px+(%2+1)*%3-4], xm14
movd [px+(%2+0)*%3+%1*2], xm14
movd [px+(%2+1)*%3+%1*2], xm14
jmp .bottom_done
.no_bottom:
movu [px+(%2+0)*%3-%1], m14
movu [px+(%2+1)*%3-%1], m14
.bottom_done:
; actual filter
INIT_YMM avx2
DEFINE_ARGS dst, stride, pridmp, damping, pri, sec, stride3, secdmp
%undef edged
movifnidn prid, prim
movifnidn secd, secm
mov dampingd, r7m
mov pridmpd, prid
mov secdmpd, secd
or pridmpd, 1
or secdmpd, 1
lzcnt pridmpd, pridmpd
lzcnt secdmpd, secdmpd
lea pridmpd, [pridmpd+dampingd-31]
lea secdmpd, [secdmpd+dampingd-31]
xor dampingd, dampingd
test pridmpd, pridmpd
cmovl pridmpd, dampingd
test secdmpd, secdmpd
cmovl secdmpd, dampingd
mov [rsp+0], pridmpq ; pri_shift
mov [rsp+8], secdmpq ; sec_shift
; pri/sec_taps[k] [4 total]
DEFINE_ARGS dst, stride, tap, dummy, pri, sec, stride3
movd xm0, prid
movd xm1, secd
vpbroadcastw m0, xm0 ; pri_strength
vpbroadcastw m1, xm1 ; sec_strength
and prid, 1
lea tapq, [tap_table]
lea priq, [tapq+priq*4] ; pri_taps
lea secq, [tapq+8] ; sec_taps
; off1/2/3[k] [6 total] from [tapq+12+(dir+0/2/6)*2+k]
DEFINE_ARGS dst, stride, tap, dir, pri, sec, stride3
mov dird, r6m
lea tapq, [tapq+dirq*2+12]
%if %1*%2*2/mmsize > 1
DEFINE_ARGS dst, stride, dir, stk, pri, sec, h, off, k
mov hd, %1*%2*2/mmsize
%else
DEFINE_ARGS dst, stride, dir, stk, pri, sec, stride3, off, k
%endif
lea stkq, [px]
pxor m13, m13
%if %1*%2*2/mmsize > 1
.v_loop:
%endif
mov kd, 1
%if %1 == 4
movq xm4, [stkq+%3*0]
movhps xm4, [stkq+%3*1]
movq xm5, [stkq+%3*2]
movhps xm5, [stkq+%3*3]
vinserti128 m4, xm5, 1
%else
mova xm4, [stkq+%3*0] ; px
vinserti128 m4, [stkq+%3*1], 1
%endif
pxor m15, m15 ; sum
mova m7, m4 ; max
mova m8, m4 ; min
.k_loop:
vpbroadcastw m2, [priq+kq*2] ; pri_taps
vpbroadcastw m3, [secq+kq*2] ; sec_taps
ACCUMULATE_TAP 0*2, [rsp+0], m0, m2, %1, %3
ACCUMULATE_TAP 2*2, [rsp+8], m1, m3, %1, %3
ACCUMULATE_TAP 6*2, [rsp+8], m1, m3, %1, %3
dec kq
jge .k_loop
vpbroadcastd m12, [pw_2048]
pcmpgtw m11, m13, m15
paddw m15, m11
pmulhrsw m15, m12
paddw m4, m15
pminsw m4, m7
pmaxsw m4, m8
packuswb m4, m4
vextracti128 xm5, m4, 1
%if %1 == 4
movd [dstq+strideq*0], xm4
pextrd [dstq+strideq*1], xm4, 1
movd [dstq+strideq*2], xm5
pextrd [dstq+stride3q], xm5, 1
%else
movq [dstq+strideq*0], xm4
movq [dstq+strideq*1], xm5
%endif
%if %1*%2*2/mmsize > 1
lea dstq, [dstq+strideq*2]
add stkq, %3*2
dec hd
jg .v_loop
%endif
RET
%endmacro
cdef_filter_fn 8, 8, 32
cdef_filter_fn 4, 4, 32
INIT_YMM avx2
cglobal cdef_dir, 3, 4, 15, src, stride, var, stride3
lea stride3q, [strideq*3]
movq xm0, [srcq+strideq*0]
movq xm1, [srcq+strideq*1]
movq xm2, [srcq+strideq*2]
movq xm3, [srcq+stride3q]
lea srcq, [srcq+strideq*4]
vpbroadcastq m4, [srcq+strideq*0]
vpbroadcastq m5, [srcq+strideq*1]
vpbroadcastq m6, [srcq+strideq*2]
vpbroadcastq m7, [srcq+stride3q]
vpbroadcastd m8, [pw_128]
pxor m9, m9
vpblendd m0, m0, m7, 0xf0
vpblendd m1, m1, m6, 0xf0
vpblendd m2, m2, m5, 0xf0
vpblendd m3, m3, m4, 0xf0
punpcklbw m0, m9
punpcklbw m1, m9
punpcklbw m2, m9
punpcklbw m3, m9
psubw m0, m8
psubw m1, m8
psubw m2, m8
psubw m3, m8
; shuffle registers to generate partial_sum_diag[0-1] together
vpermq m7, m0, q1032
vpermq m6, m1, q1032
vpermq m5, m2, q1032
vpermq m4, m3, q1032
; start with partial_sum_hv[0-1]
paddw m8, m0, m1
paddw m9, m2, m3
phaddw m10, m0, m1
phaddw m11, m2, m3
paddw m8, m9
phaddw m10, m11
vextracti128 xm9, m8, 1
vextracti128 xm11, m10, 1
paddw xm8, xm9 ; partial_sum_hv[1]
phaddw xm10, xm11 ; partial_sum_hv[0]
vinserti128 m8, xm10, 1
vpbroadcastd m9, [div_table+44]
pmaddwd m8, m8
pmulld m8, m9 ; cost6[2a-d] | cost2[a-d]
; create aggregates [lower half]:
; m9 = m0:01234567+m1:x0123456+m2:xx012345+m3:xxx01234+
; m4:xxxx0123+m5:xxxxx012+m6:xxxxxx01+m7:xxxxxxx0
; m10= m1:7xxxxxxx+m2:67xxxxxx+m3:567xxxxx+
; m4:4567xxxx+m5:34567xxx+m6:234567xx+m7:1234567x
; and [upper half]:
; m9 = m0:xxxxxxx0+m1:xxxxxx01+m2:xxxxx012+m3:xxxx0123+
; m4:xxx01234+m5:xx012345+m6:x0123456+m7:01234567
; m10= m0:1234567x+m1:234567xx+m2:34567xxx+m3:4567xxxx+
; m4:567xxxxx+m5:67xxxxxx+m6:7xxxxxxx
; and then shuffle m11 [shufw_6543210x], unpcklwd, pmaddwd, pmulld, paddd
pslldq m9, m1, 2
psrldq m10, m1, 14
pslldq m11, m2, 4
psrldq m12, m2, 12
pslldq m13, m3, 6
psrldq m14, m3, 10
paddw m9, m11
paddw m10, m12
paddw m9, m13
paddw m10, m14
pslldq m11, m4, 8
psrldq m12, m4, 8
pslldq m13, m5, 10
psrldq m14, m5, 6
paddw m9, m11
paddw m10, m12
paddw m9, m13
paddw m10, m14
pslldq m11, m6, 12
psrldq m12, m6, 4
pslldq m13, m7, 14
psrldq m14, m7, 2
paddw m9, m11
paddw m10, m12
paddw m9, m13
paddw m10, m14 ; partial_sum_diag[0/1][8-14,zero]
vbroadcasti128 m14, [shufw_6543210x]
vbroadcasti128 m13, [div_table+16]
vbroadcasti128 m12, [div_table+0]
paddw m9, m0 ; partial_sum_diag[0/1][0-7]
pshufb m10, m14
punpckhwd m11, m9, m10
punpcklwd m9, m10
pmaddwd m11, m11
pmaddwd m9, m9
pmulld m11, m13
pmulld m9, m12
paddd m9, m11 ; cost0[a-d] | cost4[a-d]
; merge horizontally and vertically for partial_sum_alt[0-3]
paddw m10, m0, m1
paddw m11, m2, m3
paddw m12, m4, m5
paddw m13, m6, m7
phaddw m0, m4
phaddw m1, m5
phaddw m2, m6
phaddw m3, m7
; create aggregates [lower half]:
; m4 = m10:01234567+m11:x0123456+m12:xx012345+m13:xxx01234
; m11= m11:7xxxxxxx+m12:67xxxxxx+m13:567xxxxx
; and [upper half]:
; m4 = m10:xxx01234+m11:xx012345+m12:x0123456+m13:01234567
; m11= m10:567xxxxx+m11:67xxxxxx+m12:7xxxxxxx
; and then shuffle m11 [shufw_210xxxxx], unpcklwd, pmaddwd, pmulld, paddd
vbroadcasti128 m14, [shufw_210xxxxx]
pslldq m4, m11, 2
psrldq m11, 14
pslldq m5, m12, 4
psrldq m12, 12
pslldq m6, m13, 6
psrldq m13, 10
paddw m4, m10
paddw m11, m12
vpbroadcastd m12, [div_table+44]
paddw m5, m6
paddw m11, m13 ; partial_sum_alt[3/2] right
vbroadcasti128 m13, [div_table+32]
paddw m4, m5 ; partial_sum_alt[3/2] left
pshufb m11, m14
punpckhwd m6, m4, m11
punpcklwd m4, m11
pmaddwd m6, m6
pmaddwd m4, m4
pmulld m6, m12
pmulld m4, m13
paddd m4, m6 ; cost7[a-d] | cost5[a-d]
; create aggregates [lower half]:
; m5 = m0:01234567+m1:x0123456+m2:xx012345+m3:xxx01234
; m1 = m1:7xxxxxxx+m2:67xxxxxx+m3:567xxxxx
; and [upper half]:
; m5 = m0:xxx01234+m1:xx012345+m2:x0123456+m3:01234567
; m1 = m0:567xxxxx+m1:67xxxxxx+m2:7xxxxxxx
; and then shuffle m11 [shufw_210xxxxx], unpcklwd, pmaddwd, pmulld, paddd
pslldq m5, m1, 2
psrldq m1, 14
pslldq m6, m2, 4
psrldq m2, 12
pslldq m7, m3, 6
psrldq m3, 10
paddw m5, m0
paddw m1, m2
paddw m6, m7
paddw m1, m3 ; partial_sum_alt[0/1] right
paddw m5, m6 ; partial_sum_alt[0/1] left
pshufb m1, m14
punpckhwd m6, m5, m1
punpcklwd m5, m1
pmaddwd m6, m6
pmaddwd m5, m5
pmulld m6, m12
pmulld m5, m13
paddd m5, m6 ; cost1[a-d] | cost3[a-d]
mova xm0, [pd_04512763+ 0]
mova xm1, [pd_04512763+ 16]
phaddd m9, m8
phaddd m5, m4
phaddd m9, m5
vpermd m0, m9 ; cost[0/4/2/6]
vpermd m1, m9 ; cost[1/5/3/7]
; now find the best cost, its idx^4 complement, and its idx
pcmpgtd xm2, xm1, xm0 ; [1/5/3/7] > [0/4/2/6]
pand xm3, xm2, xm1
pandn xm4, xm2, xm0
por xm3, xm4 ; higher 4 values
pshufd xm1, xm1, q2301
pshufd xm0, xm0, q2301
pand xm1, xm2, xm1
pandn xm0, xm2, xm0
por xm0, xm1 ; complementary 4 values at idx^4 offset
pand xm13, xm2, [pd_04261537+16]
pandn xm14, xm2, [pd_04261537+ 0]
por xm14, xm13 ; indices
punpckhqdq xm4, xm3, xm0
punpcklqdq xm3, xm0
pcmpgtd xm5, xm4, xm3 ; [2or3-6or7] > [0or1/4or5]
punpcklqdq xm5, xm5
pand xm6, xm5, xm4
pandn xm7, xm5, xm3
por xm6, xm7 ; { highest 2 values, complements at idx^4 }
movhlps xm13, xm14
pand xm13, xm5, xm13
pandn xm14, xm5, xm14
por xm14, xm13
pshufd xm7, xm6, q3311
pcmpgtd xm8, xm7, xm6 ; [4or5or6or7] > [0or1or2or3]
punpcklqdq xm8, xm8
pand xm9, xm8, xm7
pandn xm10, xm8, xm6
por xm9, xm10 ; max
movhlps xm10, xm9 ; complement at idx^4
psubd xm9, xm10
psrld xm9, 10
movd [varq], xm9
pshufd xm13, xm14, q1111
pand xm13, xm8, xm13
pandn xm14, xm8, xm14
por xm14, xm13
movd eax, xm14
RET
%endif ; ARCH_X86_64