README
¶
vectorcmp
This library provides functions to do comparions on many rows using AVX2's VPCMP instruction. Fallback code is transparently used if AVX2 is not available.
VectorEquals
The easiest way to explain is to just show the fallback code.
func goVectorEquals[T uint8 | uint16 | uint32 | uint64](dst []byte, search T, rows []T) {
clear(dst)
for i, v := range rows {
if search == v {
dst[i/8] |= 1 << (i % 8)
}
}
}
Similar to VectorEquals we also implement VectorGreaterThan, VectorLessThan, VectorGreaterEquals and VectorLesserEquals.
Benchmark
goos: linux
goarch: amd64
pkg: github.com/Jille/vectorcmp
cpu: 13th Gen Intel(R) Core(TM) i9-13900
│ purego │ asm │
│ sec/op │ sec/op vs base │
VectorEquals8-32 164.600µ ± 2% 1.208µ ± 16% -99.27% (p=0.000 n=10)
VectorEquals16-32 125.375µ ± 2% 2.429µ ± 16% -98.06% (p=0.000 n=10)
VectorEquals32-32 163.694µ ± 3% 4.645µ ± 12% -97.16% (p=0.000 n=10)
VectorEquals64-32 127.200µ ± 2% 8.495µ ± 15% -93.32% (p=0.000 n=10)
geomean 144.0µ 3.280µ -97.72%
Documentation
¶
Index ¶
- func VectorEquals[T uint8 | uint16 | uint32 | uint64](dst []byte, search T, rows []T)
- func VectorEquals16(dstMask []byte, b uint16, rows []uint16)
- func VectorEquals32(dstMask []byte, b uint32, rows []uint32)
- func VectorEquals64(dstMask []byte, b uint64, rows []uint64)
- func VectorEquals8(dstMask []byte, b uint8, rows []uint8)
- func VectorGreaterEquals[T uint8 | uint16 | uint32 | uint64](dst []byte, search T, rows []T)
- func VectorGreaterEquals16(dstMask []byte, b uint16, rows []uint16)
- func VectorGreaterEquals32(dstMask []byte, b uint32, rows []uint32)
- func VectorGreaterEquals64(dstMask []byte, b uint64, rows []uint64)
- func VectorGreaterEquals8(dstMask []byte, b uint8, rows []uint8)
- func VectorGreaterThan[T uint8 | uint16 | uint32 | uint64](dst []byte, search T, rows []T)
- func VectorGreaterThan16(dstMask []byte, b uint16, rows []uint16)
- func VectorGreaterThan32(dstMask []byte, b uint32, rows []uint32)
- func VectorGreaterThan64(dstMask []byte, b uint64, rows []uint64)
- func VectorGreaterThan8(dstMask []byte, b uint8, rows []uint8)
- func VectorLessThan[T uint8 | uint16 | uint32 | uint64](dst []byte, search T, rows []T)
- func VectorLessThan16(dstMask []byte, b uint16, rows []uint16)
- func VectorLessThan32(dstMask []byte, b uint32, rows []uint32)
- func VectorLessThan64(dstMask []byte, b uint64, rows []uint64)
- func VectorLessThan8(dstMask []byte, b uint8, rows []uint8)
- func VectorLesserEquals[T uint8 | uint16 | uint32 | uint64](dst []byte, search T, rows []T)
- func VectorLesserEquals16(dstMask []byte, b uint16, rows []uint16)
- func VectorLesserEquals32(dstMask []byte, b uint32, rows []uint32)
- func VectorLesserEquals64(dstMask []byte, b uint64, rows []uint64)
- func VectorLesserEquals8(dstMask []byte, b uint8, rows []uint8)
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func VectorEquals ¶
VectorEquals compares each entry in rows to search and enables the corresponding bit in dst if equal. Prefer calling VectorEquals8/16/32/64 directly.
clear(dst)
for i, v := range rows {
if search == v {
dst[i/8] |= 1 << (i % 8)
}
}
func VectorEquals16 ¶
func VectorEquals32 ¶
func VectorEquals64 ¶
func VectorEquals8 ¶
func VectorGreaterEquals ¶
VectorGreaterEquals compares each entry in rows to search and enables the corresponding bit in dst if search >= rows[i]. Prefer calling VectorGreaterEquals8/16/32/64 directly.
clear(dst)
for i, v := range rows {
if search >= v {
dst[i/8] |= 1 << (i % 8)
}
}
func VectorGreaterEquals16 ¶
func VectorGreaterEquals32 ¶
func VectorGreaterEquals64 ¶
func VectorGreaterEquals8 ¶
func VectorGreaterThan ¶
VectorGreaterThan compares each entry in rows to search and enables the corresponding bit in dst if search > rows[i]. Prefer calling VectorGreaterThan8/16/32/64 directly.
clear(dst)
for i, v := range rows {
if search > v {
dst[i/8] |= 1 << (i % 8)
}
}
func VectorGreaterThan16 ¶
func VectorGreaterThan32 ¶
func VectorGreaterThan64 ¶
func VectorGreaterThan8 ¶
func VectorLessThan ¶
VectorLessThan compares each entry in rows to search and enables the corresponding bit in dst if search < rows[i]. Prefer calling VectorLessThan8/16/32/64 directly.
clear(dst)
for i, v := range rows {
if search < v {
dst[i/8] |= 1 << (i % 8)
}
}
func VectorLessThan16 ¶
func VectorLessThan32 ¶
func VectorLessThan64 ¶
func VectorLessThan8 ¶
func VectorLesserEquals ¶
VectorLesserEquals compares each entry in rows to search and enables the corresponding bit in dst if search <= rows[i]. Prefer calling VectorLesserEquals8/16/32/64 directly.
clear(dst)
for i, v := range rows {
if search <= v {
dst[i/8] |= 1 << (i % 8)
}
}
func VectorLesserEquals16 ¶
func VectorLesserEquals32 ¶
func VectorLesserEquals64 ¶
func VectorLesserEquals8 ¶
Types ¶
This section is empty.
Source Files