tzhash/tz/avx2_amd64.s
Evgenii Stratonikov c8a32b25ec Optimize AVX2 implementation
We use 6 instructions only to calculate mask based on single bit value.
Use only 3 now and calculate multiple masks in parallel.

Also `VPSUB*` is faster than VPBROADCAST*,
see https://www.intel.com/content/www/us/en/docs/intrinsics-guide/index.html .

```
name                     old time/op    new time/op    delta
Sum/AVX2Inline_digest-8    1.83ms ± 0%    1.62ms ± 1%  -11.23%  (p=0.000 n=46+42)

name                     old speed      new speed      delta
Sum/AVX2Inline_digest-8  54.7MB/s ± 0%  61.6MB/s ± 1%  +12.65%  (p=0.000 n=46+42)
```

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2021-12-29 13:23:05 +03:00

75 lines
1.7 KiB
ArmAsm

#include "textflag.h"
#define mask(bit, tmp, to) \
VPSRLW bit, Y10, tmp \
VPAND Y12, tmp, to \ // to2 = 0x000<bit>000<bit>...
VPSUBW to, Y13, to // to2 = 0xFFFF.. or 0x0000 depending on bit
#define mulBit(bit) \
VPSLLQ $1, Y0, Y1 \
VPALIGNR $8, Y1, Y0, Y2 \
VPSRLQ $63, Y2, Y2 \
VPXOR Y1, Y2, Y2 \
VPSRLQ $63, Y1, Y3 \
VPSLLQ $63, Y3, Y3 \
VPUNPCKHQDQ Y3, Y3, Y3 \
VPXOR Y2, Y3, Y3 \
mask(bit, Y11, Y2) \
VPXOR Y3, Y8, Y3 \
VPAND Y3, Y2, Y4 \
VPXOR Y4, Y0, Y8 \
VMOVDQA Y3, Y0
// func mulByteRightx2(c00c10, c01c11 *[4]uint64, b byte)
TEXT ·mulByteRightx2(SB),NOSPLIT,$0
MOVQ c00c10+0(FP), AX
VMOVDQA (AX), Y0
MOVQ c01c11+8(FP), BX
VMOVDQA (BX), Y8
VPXOR Y13, Y13, Y13 // Y13 = 0x0000...
VPCMPEQB Y12, Y12, Y12 // Y12 = 0xFFFF...
VPSUBW Y12, Y13, Y12 // Y12 = 0x00010001... (packed words of 1)
VPBROADCASTB b+16(FP), X10 // X10 = packed bytes of b.
VPMOVZXBW X10, Y10 // Extend with zeroes to packed words.
mulBit($7)
mulBit($6)
mulBit($5)
mulBit($4)
mulBit($3)
mulBit($2)
mulBit($1)
mulBit($0)
VMOVDQA Y8, (BX)
VMOVDQA Y0, (AX)
RET
// func mulBitRightx2(c00c10, c01c11 *[4]uint64, e *[2]uint64)
TEXT ·mulBitRightx2(SB),NOSPLIT,$0
MOVQ c00c10+0(FP), AX
VMOVDQA (AX), Y0
MOVQ c01c11+8(FP), BX
VMOVDQA (BX), Y8
VPSLLQ $1, Y0, Y1
VPALIGNR $8, Y1, Y0, Y2
VPSRLQ $63, Y2, Y2
VPXOR Y1, Y2, Y2
VPSRLQ $63, Y1, Y3
VPSLLQ $63, Y3, Y3
VPUNPCKHQDQ Y3, Y3, Y3
VPXOR Y2, Y3, Y3
MOVQ e+16(FP), CX
VBROADCASTI128 (CX), Y2
VPXOR Y3, Y8, Y3
VPAND Y3, Y2, Y4
VPXOR Y4, Y0, Y8
VMOVDQA Y8, (BX)
VMOVDQA Y3, (AX)
RET